SUPPRESSOR OF GAMMA RESPONSE1 Links DNA Damage Response to Organ Regeneration

Abstract

In Arabidopsis, DNA damage-induced programmed cell death is limited to the meristematic stem cell niche and its early descendants. The significance of this cell-type-specific programmed cell death is unclear. Here, we demonstrate in roots that it is the programmed destruction of the mitotically compromised stem cell niche that triggers its regeneration, enabling growth recovery. In contrast to wild-type plants, sog1 plants, which are defective in damage-induced programmed cell death, maintain the cell identities and stereotypical structure of the stem cell niche after irradiation, but these cells fail to undergo cell division, terminating root growth. We propose DNA damage-induced programmed cell death is employed by plants as a developmental response, contrasting with its role as an anticarcinogenic response in animals. This role in plants may have evolved to restore the growth of embryos after the accumulation of DNA damage in seeds.

Figure 1.

Growth recovery after IR. A, Root growth rate per day (growth since previous day) after 150 Gy IR, as a fraction of root growth rate for the mock-irradiated control, in wild type (WT), sog1-1, and sog1-1 + tgSOG1. ***sog1-1 samples are significantly different from wild-type (P < 0.001); *wild-type samples at 7 d after IR are significantly different from those 6 d after IR (P < 0.05). B, Wild-type (left) and sog1-1 (right) roots at 8 d after IR. Arrowhead indicates root tip position at time of IR. C, Five-day-old wild-type, sog1-1, or sog1-1 + SOG1 root tips were stained with propidium iodide and imaged up to 7 d after 150 Gy IR.

Figure 2.

Effects of IR on QC marker expression. A, Five-day-old seedlings carrying pWOX5:_ERGFP in wild-type (WT), sog1-1, or sog1-9 backgrounds were irradiated and imaged up to 8 d after 150 Gy IR. False color: black, propidium iodide; purple, GFP. B, Five-day-old seedlings carrying QC25 (top) or QC46 (bottom) GUS enhancer trap lines were stained and imaged up to 8 d after 150 Gy IR.

Figure 3.

Effects of SOG1 on cell cycle arrest in the RAM after irradiation. A, DNA replication in the mitotic zone measured as EdU-labeled nuclei. Five-day-old seedlings were irradiated to 150 Gy and labeled with EdU for 4 h beginning at 6 h after the completion of IR. Error bars are se (wild-type [WT], n = 9; sog1-1, n = 6; sog1-1 + tgSOG1, n = 6; wild type + IR, n = 11; sog1-1 + IR, n = 10; sog1-1 + tgSOG1 + IR, n = 10). *+ IR samples are significantly different from mock-IR controls (P < 0.05). B, The number of metaphase cells per root tip observed in anticlinal versus periclinal orientations, up to 7 d after 5-d-old wild-type or sog1-9 seedlings, both carrying pRPS5a::H2B:eCFP, were irradiated with 150 Gy IR. Error bars are se (wild type, n = 8; sog1-9, n = 8). * and **sog1-9 samples are significantly different from wild type (P values are < 0.05 and < 0.01, respectively).

Figure 4.

Effects of IR on ERF115 expression and PCD. Five-day-old seedlings carrying pERF115:NLS-GUS/GFP in wild-type backgrounds were irradiated and imaged up to 7 h after 150 Gy IR. False color: black, propidium iodide; purple, GFP.

Figure 5.

pERF115:GUS expression after cell death. A, Five-day-old seedlings carrying pERF115:NLS-GUS/GFP were irradiated and imaged 8 and 32 h after 150 Gy IR (or mock irradiation). B, Five-day-old seedlings were irradiated with 150 Gy. Seedlings were stained with propidium iodide to visualize PCD up to 32 h after IR. C, Five-day-old wild-type (WT) or sog1-1 seedlings carrying pERF115:NLS-GUS/GFP were either cut in the meristematic zone or left intact, then stained with propidium iodide (top) up to 8 h after cutting, and immediately following visualization; the same seedlings were stained for GUS activity (bottom).

Figure 6.

WOX5, ERF115 induction during the onset of PCD. WOX5, ERF115, ERF114, PAT1, ACT1 (ACTIN1) fold change in GFP+ (stele-specific) or whole-root-tip protoplasts during the onset of PCD after 100 Gy IR. Error bars are se. *Irradiated samples are significantly differentially expressed from mock-irradiated controls (P < 0.05).