A DOF transcriptional repressor-gibberellin feedback loop plays a crucial role in modulating light-independent seed germination

Abstract

Plants have evolved several strategies to cope with the ever-changing environment. One example of this is given by seed germination, which must occur when environmental conditions are suitable for plant life. In the model system Arabidopsis thaliana seed germination is induced by light; however, in nature, seeds of several plant species can germinate regardless of this stimulus. While the molecular mechanisms underlying light-induced seed germination are well understood, those governing germination in the dark are still vague, mostly due to the lack of suitable model systems. Here, we employ Cardamine hirsuta, a close relative of Arabidopsis, as a powerful model system to uncover the molecular mechanisms underlying light-independent germination. By comparing Cardamine and Arabidopsis, we show that maintenance of the pro-germination hormone gibberellin (GA) levels prompt Cardamine seeds to germinate under both dark and light conditions. Using genetic and molecular biology experiments, we show that the Cardamine DOF transcriptional repressor DOF AFFECTING GERMINATION 1 (ChDAG1), homologous to the Arabidopsis transcription factor DAG1, is involved in this process functioning to mitigate GA levels by negatively regulating GA biosynthetic genes ChGA3OX1 and ChGA3OX2, independently of light conditions. We also demonstrate that this mechanism is likely conserved in other Brassicaceae species capable of germinating in dark conditions, such as Lepidium sativum and Camelina sativa. Our data support Cardamine as a new model system suitable for studying light-independent germination studies. Exploiting this system, we have also resolved a long-standing question about the mechanisms controlling light-independent germination in plants, opening new frontiers for future research.

Keywords: Cardamine hirsuta; DOF AFFECTING GERMINATION1; gibberellins; light; seed germination.

Figure 1

Light and hormonal requirements for the germination of Cardamine, Capsella, Lepidium and Camelina seeds. (A) Seed germination of Cardamine hirsuta, Capsella rubella, Lepidium sativum and Camelina sativa seeds, and Arabidopsis thaliana as a control from 0 to 24 HAI. (B–E) Germination rates: in white light (B), total darkness (C), with increasing ABA concentrations (D), or PAC concentrations (E). Germination rate was measured at different HAI (6, 12, 24, 36, 48, 60, 72 and 120) in (B and C), and at 120 HAI in (D and E). The values are means of three biological replicates, with SD values. Significant differences were analyzed by t-test (∗∗∗p ≤ 0.001, ∗∗p ≤ 0.005, ∗p ≤ 0.05). PAC, paclobutrazol; HAI, hours after imbibition. Control is referred to “mock treatment control” with ethanol. The diagram on top depicts the light treatment scheme; STRAT, stratification (2 days at 4°C, dark), WL, white light, D, dark.

Figure 2

A high GA/ABA ratio enables the germination of Cardamine seeds in darkness. (A and B) Relative expression level of: ChGA20OX3, ChGA3OX1, ChGA3OX2, ChGA2OX2, ChGA2OX3(A), ChABA1, ChNCED6, ChNCED9, ChCYP707A2(B) in Cardamine wild-type (Ox) seeds at 12 and 24 HAI (Hours After Imbibition), under light and dark conditions. Expression levels as log10 respect to the dry condition, set to 0 (X axis). The values are means of three biological replicates, with SD values. Significant differences were analyzed by t-test (∗∗∗p ≤ 0.001, ∗∗p ≤ 0.005, ∗p ≤ 0.05). (C) Germination of seeds ChUBQ10>>GA2OX2, issued from the cross UBQ10::GAL4 × UAS::GA2OX2, under dark conditions, compared with the wild-type (Ox) and the UBQ10-GAL4 line. Germination rates were measured at 120 HAI (Hours After Imbibition). The values are means of three biological replicates, with SD values. Significant differences were analyzed by one-way ANOVA with post hoc Tukey multiple comparison test (∗∗p ≤ 0.005). (D) Ratio of bioactive GAs/ABA in wild-type Ox seeds; the analyses were performed on dry and 24-h imbibed seeds in light and dark conditions. The values are the mean of three biological replicates, with SD values.

Figure 3

ChDAG1 is involved in light-independent seed germination. (A and B) Relative expression level of: ChPIF1, ChGAI, and ChRGA(A), ChDAG1and AtDAG1(B) in Cardamine and Arabidopsis wild-type (Ox, Ws, respectively) seeds at 12 and 24 HAI (Hours After Imbibition), under light and dark conditions. Expression levels as log10 respect to the dry condition, set to 0 (X axis). The values are means of three biological replicates, with SD values. Significant differences were analyzed by t-test (∗∗∗p ≤ 0.001, ∗∗p ≤ 0.005, ∗p ≤ 0.05). (C) Histochemical staining of pChDAG1::GUS and pAtDAG1::GUS seeds dry (DRY) or imbibed 24 h, under white light (WL) or in dark (D). Scale bar, 1 mm. (D) Sequence of the Chdag1-1 and Chdag1-2 mutant alleles. (E–G) Germination rates of wild-type (Ox) and both Chdag1-1 and Chdag1-2 mutant seeds: in white light (E), in total darkness (F), and in the presence of PAC 100 μM + increasing concentrations of GAs (G). Germination rates were measured at different HAI (12, 24, 36, 48, 60, 72, and 120) in (E and F), and at 120 HAI in (G). The values are means of three biological replicates, with SD values. Significant differences were analyzed by t-test (∗∗∗p ≤ 0.001, ∗∗p ≤ 0.005, ∗p ≤ 0.05). PAC, paclobutrazol; HAI, hours after imbibition. Control is referred to “mock treatment control” with ethanol. The diagram on top depicts the light treatment scheme; STRAT, stratification (2 days at 4°C, dark), WL, white light; D, dark.

Figure 4

Expression profiles of GA and ABA genes in Chdag1-1 and Chdag1-2 mutant seeds. (A–C) Relative expression level of: ChGA20OX3, ChGA3OX1, ChGA3OX2, ChGA2OX2, ChGA2OX3(A), ChABA1, ChNCED6, ChNCED9, ChCYP707A2(B), ChPIF1, ChGAI, ChRGA(C), in Chdag1-1 and Chdag1-2 mutant seeds compared to the wild-type, at 12 and 24 HAI (Hours After Imbibition), under light and dark conditions. Expression levels as log10 respect to the dry condition, set to 0 (X axis). The values of relative expression levels are means of three biological replicates, with SD values. Significant differences were analyzed by t-test (∗∗p ≤ 0.001, ∗∗p ≤ 0.005, ∗p ≤ 0.05).

Figure 5

GAs promote ChDAG1 expression in dark-imbibed seeds. (A and B) Relative expression level of ChDAG1(A) and AtDAG1(B) in 24-h imbibed wild-type seeds (Ox and Ws, respectively), in the presence of water (control) or GA₄₊₇ (100 μM), in white light or in darkness. The values of relative expression levels are the mean of three biological replicates, with SD values. Expression levels were normalized with that of the ChUBQ10 and AtUBQ10 genes for Cardamine and Arabidopsis samples, respectively. The values are the mean of three biological replicates, with SD values. Significant differences were analyzed by t-test (∗∗p ≤ 0.005, ∗p ≤ 0.05). (C) Histochemical staining of pChDAG1::GUS and of pAtDAG1::GUS seeds imbibed 24 h, with/without addition of GAs, under white light (WL) or in dark (D). Scale bar, 1 mm. (D) Ratio of bioactive GAs/ABA in Chdag1-1 mutant seeds compared to Ox seeds (Figure 2D). The analyses were performed on dry and 24-h imbibed seeds in light and dark conditions. The values are the mean of three biological replicates, with SD values.

Figure 6

Activity of ChDAG1 in the Atdag1 background. (A–C) Germination rates of Atdag1,pChDAG1::ChDAG1, Atdag1, and wild-type (Ws) seeds in white light (A), in total darkness (B), and in the presence of PAC 100 μM + increasing concentrations of GA₄₊₇(C). Germination rates were measured at different HAI (12, 24, 36, 48, 60, 72, and 120) in (A and B), and at 120 HAI in (C). The values are the mean of three biological replicates, with SD values. Significant differences were analyzed by t-test (∗∗p ≤ 0.005, ∗p ≤ 0.05). PAC, paclobutrazol; HAI, hours after imbibition. Control is referred to “mock treatment control” with ethanol. The diagram on top depicts the light treatment scheme; STRAT, stratification (2 days at 4°C, dark), WL, white light; D, dark. (D and E) Relative expression level of ChDAG1 and AtDAG1(D and E) and of AtGA3OX1 and AtCYP707A2 (from top to bottom). Seeds of Atdag1,pChDAG1::ChDAG1-a(D) and Atdag1,pAtDAG1::AtDAG1-a(E) at 12 and 24 HAI, under light and dark conditions. Expression levels as log10 respect to the dry condition, set to 0 (X axis). The values of relative expression levels are means of three biological replicates, with SD values. Significant differences were analyzed by t-test (∗∗∗p ≤ 0.001, ∗∗p ≤ 0.005, ∗p ≤ 0.05).

Figure 7

Scheme of the molecular mechanism underlying seed germination in Cardamine. (A and B) Scheme of the main elements involved in seed germination in Cardamine(A) and Arabidopsis(B). Red and black arrows are referred to light and dark conditions, respectively. The arrows' thickness is referred to the expression level of the corresponding genes. (A) In seeds of Cardamine, the transcript levels of ChGA3OX1 and ChGA3OX2 are higher in the dark than in the light, thus increasing GA levels. ChDAG1 represses these two GAs’ biosynthetic genes which, in the absence of ChDAG1, are upregulated, particularly in light-imbibed seeds. The transcript level of ChCYP707A2 is not altered by the inactivation of ChDAG1, suggesting that ChDAG1 is not involved in its regulation. GAs increase ChDAG1 transcript level in the dark, while decreasing it in the light. (B) In seeds of Arabidopsis, AtDAG1 represses both AtGA3OX1 and AtCYP707A2, mainly in the dark (Gabriele et al., 2010; Boccaccini et al., 2016). GAs increase AtDAG1 transcript level in the light, while decreasing it in the dark.