Data from: Stress induced gene expression drives transient DNA methylation changes at adjacent repetitive elements

Figure 1-source data 1.xlsx (16.93 Mb):
Figure 1-source data 1: FPKM and fold change of all genes in rice roots and shoots upon Pi treatments.

Figure 2-source data 1.xlsx (131.2 Kb):
Figure 2-source data 1: List of 175 PSI DMRs identified in the root, with DNA methylation levels and transcription levels of DMR-associated genes.

Figure 2-source data 2.xlsx (28.38 Kb):
Figure 2-source data 2: Transcription levels of TEs overlapping PSI DMRS after 7 d and 21 d of Pi deprivation.

Figure 2-source data 3.xlsx (55.15 Kb):
Figure 2-source data 3: List of 341 PSI DMRs identified in the shoots, with DNA methylation levels and transcription levels of DMR-associated genes.

Figure 3-source data 1.xlsx (75.92 Kb):
Figure 3-source data 1: Levels of DNA methylation and FDR values for the 100 PSI DMRs associated with changes in proximal gene transcription.

Figure 4-source data 1.xlsx (14.4 Kb):
Figure 4-source data 1: List of 14 root persisting PSI DMRs, despite 31 d of Pi resupply.

Figure 4-source data 2.xlsx (22.46 Kb):
Figure 4-source data 2: List of 36 panicle PSI DMRs.

Figure 4-source data 3.xlsx (61.38 Kb):
Figure 4-source data 3: Average methylation levels observed in the roots of the progeny of stressed- and non-stressed plants in all contexts and methylation levels for each replicate in CNN in the 175 root PSI DMRs.

Figure 5-source data 1.xlsx (6.470 Mb):
Figure 5-source data 1: FPKM and fold change of all genes between WT +Pi and DCL3a RNAi line +Pi as well as WT -Pi and DCL3a RNAi line -Pi (FDR < 0.05). List of DMRs identified between WT +Pi and DCL3a RNAi line +Pi as well as WT -Pi and DCL3a RNAi line -Pi is shown as well as the average methylation levels observed in all samples (FDR < 0.05).

Figure 5-source data 2.xlsx (58.18 Kb):
Figure 5-source data 2: 11 DCL3a-dependant PSI DMRs, with methylation levels and associated gene transcription.

Figure 6-source data 1.xlsx (6.269 Mb):
Figure 6-source data 1: FPKM and fold change of all genes in Arabidopsis roots upon Pi starvation.

Figure 5-source data 3.xlsx (39.49 Kb):
409 PSI DMRs in the DCL3a RNAi line, and 88 PSI DMRs in the DCL3a RNAi line, associated with significant changes in nearby gene expression.

Cytosine DNA methylation (mC) is a genome modification that can regulate the expression of coding and non-coding genetic elements. However, little is known about the involvement of mC in response to environmental cues. Using whole genome bisulfite sequencing to assess the spatio-temporal dynamics of mC in rice grown under phosphate starvation and recovery conditions, we identified widespread phosphate starvation-induced changes in mC, preferentially localized in transposable elements (TEs) close to highly induced genes. These changes in mC occurred after changes in nearby gene transcription, were mostly DCL3a-independent, could partially be propagated through mitosis, however no evidence of meiotic transmission was observed. Similar analyses performed in Arabidopsis revealed a very limited effect of phosphate starvation on mC, suggesting a species-specific mechanism. Overall, this suggests that TEs in proximity to environmentally induced genes are silenced via hypermethylation, and establishes the temporal hierarchy of transcriptional and epigenomic changes in response to stress.

图1-源数据集1.xlsx（16.93 Mb）：
图1-源数据集1：水稻根与地上部经磷（Pi）处理后所有基因的FPKM值与倍数变化（fold change）。
图2-源数据集1.xlsx（131.2 Kb）：
图2-源数据集1：在根中鉴定到的175个磷饥饿诱导差异甲基化区域（PSI DMRs）列表，包含DMR关联基因的DNA甲基化水平与转录水平。
图2-源数据集2.xlsx（28.38 Kb）：
图2-源数据集2：磷饥饿处理7天与21天后，与PSI DMRs重叠的转座因子（TEs，Transposable Elements）的转录水平。
图2-源数据集3.xlsx（55.15 Kb）：
图2-源数据集3：在地上部鉴定到的341个PSI DMRs列表，包含DMR关联基因的DNA甲基化水平与转录水平。
图3-源数据集1.xlsx（75.92 Kb）：
图3-源数据集1：与近端基因转录变化相关的100个PSI DMRs的DNA甲基化水平与错误发现率（FDR，False Discovery Rate）值。
图4-源数据集1.xlsx（14.4 Kb）：
图4-源数据集1：经31天磷恢复处理后仍持续存在的14个根特异性PSI DMRs列表。
图4-源数据集2.xlsx（22.46 Kb）：
图4-源数据集2：36个穗部PSI DMRs列表。
图4-源数据集3.xlsx（61.38 Kb）：
图4-源数据集3：在所有甲基化语境下，胁迫与非胁迫植株子代根中的平均甲基化水平，以及175个根PSI DMRs中卷积神经网络（CNN，Convolutional Neural Network）重复样本的甲基化水平。
图5-源数据集1.xlsx（6.470 Mb）：
图5-源数据集1：野生型（WT，Wild Type）+Pi组与DCL3a RNA干扰（RNAi，RNA interference）株系+Pi组，以及野生型（WT）-Pi组与DCL3a RNAi株系-Pi组间所有基因的FPKM值与倍数变化（fold change）；同时列出了上述两组间鉴定到的DMRs，以及所有样本中的平均甲基化水平（FDR < 0.05）。
图5-源数据集2.xlsx（58.18 Kb）：
图5-源数据集2：11个依赖于DCL3a的PSI DMRs，包含其甲基化水平与关联基因的转录水平。
图6-源数据集1.xlsx（6.269 Mb）：
图6-源数据集1：拟南芥根经磷饥饿处理后所有基因的FPKM值与倍数变化（fold change）。
图5-源数据集3.xlsx（39.49 Kb）：
图5-源数据集3：DCL3a RNAi株系中的409个PSI DMRs，以及该株系中与邻近基因表达显著变化相关的88个PSI DMRs。
胞嘧啶DNA甲基化（mC，Cytosine DNA methylation）是一种可调控编码与非编码遗传元件表达的基因组修饰。然而，目前关于mC参与环境应答的相关机制仍知之甚少。本研究通过全基因组亚硫酸氢盐测序，评估了磷饥饿与恢复条件下生长的水稻中mC的时空动态特征，鉴定到大量磷饥饿诱导的mC变化，这些变化优先定位于高度诱导基因附近的转座因子（TEs）中。上述mC变化发生于邻近基因转录改变之后，且大多不依赖于DCL3a，可通过有丝分裂部分传递，但未观察到减数分裂传递的证据。在拟南芥中开展的类似分析显示，磷饥饿对mC的影响极为有限，这表明该过程存在物种特异性机制。总体而言，本研究结果表明，环境诱导基因邻近的转座因子可通过超甲基化被沉默，并阐明了响应胁迫的转录与表观基因组变化的时间层级关系。