Arabidopsis GLR3.5-modulated seed germination involves GA and ROS signaling

ABSTRACT

Seed germination, a crucial developmental step, is regulated by multiple plant endogenous signals, among which phytohormones absisic acid (ABA) and gibberellin (GA) act antagonistically. Reactive oxygen species (ROS) interact with the two hormones to coordinate germination. We have previously reported that Arabidopsis glutamate receptor homolog3.5 (AtGLR3.5) modulates calcium signal to attenuate the repression effect of ABA on seed germination and that amino acid L-methionine functions upstream of AtGLR3.5, resulting in calcium influx. Here, we show that AtGLR3.5 modulates GA and ROS signaling during seed germination. Our findings provide a more complete picture as to the molecular mechanisms of AtGLR3.5 in seed germination control.

Seed germination is a critical developmental step regulated by multiple plant endogenously produced signals, such as phytohormones, reactive oxygen species (ROS), and calcium (Ca²⁺).^1-5 Absisic acid (ABA) and gibberellin (GA) are two prominent hormones acting antagonistically in seed germination control, with the former inhibiting while the latter promoting germination.^1,2 With the onset of germination process, the ABA level in seeds decreases and bioactive GA content increases, and the expression of many genes in metabolic and signaling pathways of the two hormones alter accordingly.^2,6 Emerging lines of evidence show that ROS, which favor germination initiation, accumulate after seed imbibition,^3,7 and that the function of ROS in germination regulation relies largely on their interactions with ABA and GA.⁸ We recently reported that Ca²⁺ signal stimulates seed germination, and that Ca²⁺ signal impedes the ABA repression on germination.⁴ We showed that a Ca²⁺ permeable channel named Arabidopsis glutamate receptor homolog3.5 (AtGLR3.5)^4,9 mediates the cytosolic Ca²⁺ influx that acts against a core regulator in seed ABA response, ABSCISIC ACID INSENSITIVE4 (ABI4),¹⁰ to foster germination.^4,5

In this study, we first investigated whether AtGLR3.5-mediated Ca²⁺ is related to GA signaling in germination control. We tested the expression of crucial GA regulators¹¹ in the AtGLR3.5 RNA interference (RNAi) lines, which have a defect in free cytosolic Ca²⁺ level fluctuation and germinate more slowly than the wild type.⁴ The data showed that the expression of GA biosynthesis genes GA REQUIRING 3 (GA3) and GA 20-oxidase2 (GA20ox2) was down-regulated in germinating AtGLR3.5 RNAi seeds compared with the wild type, whereas the expression of RESTORATION OF GROWTH ON AMMONIA 2 (RGA2) and RGA-LIKE 2 (RGL2), two negative regulators in GA signaling, was up-regulated (Figure 1), suggesting impaired GA metabolism and GA sensitivity in germinating AtGLR3.5 RNAi seeds. These results indicate that AtGLR3.5 positively regulates GA biosynthesis and GA signaling during seed germination.

Figure 1.

Expression of GA biosynthesis and signaling genes in wild-type (WT) and AtGLR3.5 RNAi seeds at 0 and 24 h after incubation under germination conditions. Seeds were incubated at 4°C for 3 d and transferred to a growth chamber for indicated time periods prior to the analysis. Relative gene expression levels were calculated using ACTIN2 as a normalized reference gene. Data are shown as means ± SE, n = 3. Asterisks indicate significant differences determined by Student’s t-test (*P < .05; **P < .01).

In additon to being a key regulator in ABA biogenesis and ABA signaling, transcription factor ABI4 attenuates GA biosynthesis during germination.¹¹ We then tested the expression of above GA regulators in abi4-1 mutant, where ABI4 transcript amount was largely reduced,¹¹ under our experimental conditions. Consistent with the reported notion, we found that the transcripts of GA3 and GA20ox2 were enhanced in germinating abi4-1 mutants compared with the wild type, while the transcripts of RGA2 and RGL2 were reduced in the mutants (Figure 2). Because germinating AtGLR3.5 RNAi seeds harbored elevated ABI4 transcripts,⁴ the reduced expression of GA3 and GA20ox2 and enhanced expression of RGA2 and RGL2 we observed in these seeds (Figure 1) may be attributable to the high amounts of ABI4. Taken together, these results indicate that the regulation of AtGLR3.5 to the expression of GA biosynthesis and signaling genes during germination may be largely achieved through ABI4.

Figure 2.

Expression of genes involved in GA biosynthesis and signaling pathways in wild-type (WT) and abi4-1 seeds. After incubation at 4°C for 3 d, the seeds were transferred to a growth chamber for 24 h prior to the analysis. Relative gene expression levels were calculated using ACTIN2 as a normalized reference gene. Data are shown as means ± SE, n = 3. Asterisks indicate significant differences determined by student’s t-test (**P < .01).

We next examined the modulation of ROS generation by AtGLR3.5 in germinating seeds. Interestingly, ROS production in the embryonic axis, a position where ROS preferentially accumulate in germinating embryos,¹² was significantly lower in the AtGLR3.5 RNAi lines than the wild type (Figure 3a-c). ROS can be generated from different sources in hydrated seeds, such as by lipid and purine catabolism, respiratory action, or through the activity of various enzymes including NADPH oxidases.³ Consistent with ROS staining results (Figure 3a-c), the expression level of the NADPH oxidase rbohD, known to be highly transcribed in the endosperm during germination,¹³ was lower in germinating AtGLR3.5 RNAi seeds than the wild type (Figure 3d). A study has shown that ABI4 represses lipid breakdown occurred in the embryo during germination,¹³ so the high amounts of ABI4 may also contribute to the impaired ROS production we observed in germinating AtGLR3.5 RNAi seeds. These data suggest that AtGLR3.5 may affect different sources of ROS production during germination.

Figure 3.

ROS production in germinating wild-type (WT) and AtGLR3.5 RNAi seeds. (a and b) ROS production in the embryonic axes of the samples. Seeds were stratified and transferred to a growth chamber for 24 h, and ROS staining was performed using fluorescent dye DCFH-DA after the removal of each each coat. (c) Quantification of ROS production in samples in (a) and (b). Data are shown as means ± SE, n = 19 for wild-type seeds and n = 16 for AtGLR3.5 RNAi seeds. (d) qRT-PCR analyses of rbohD expression in wild-type (WT) and AtGLR3.5 RNAi seeds after incubation under germination conditions for 24 h. Relative gene expression levels were calculated using ACTIN2 as a normalized reference gene. Data are shown as means ± SE, n = 3. Asterisks indicate significant differences determined by Student’s t-test (**P < .01).

Using pharomological, molecular, and biochemical approaches, we previously demonstrated the negative modulation of AtGLR3.5 and it-mediated Ca²⁺ signal to ABA effect on germination.⁴ Our work here suggests that the AtGLR3.5 regulation to germination involves GA and ROS signaling, opposite to its action mode on ABA. Combining these information, we conclude that AtGLR3.5 and it-mediated Ca²⁺ signal work upstream of ABA, GA, and ROS, with ABI4 being an important integration node among these signals. A recent report shows that amino acid methionine, biosynthesized via methionine synthase 1, modulates the AtGLR3.5-mediated Ca²⁺ signal and ABI4 function in germination.⁵ With the addition of the these new findings, the picture as to the molecular mechanisms of AtGLR3.5 and it-mediated Ca²⁺ signal in seed germination control is more clear.

Funding Statement

This work was supported by the Beijing Municipal Education Commission [19530050120];National Natural Science Foundation of China [31770296];Beijing Natural Science Foundation [6172002];

Acknowledgments

We would like to thank Dr. Ruth Finkelstein (University of California, Santa Barbara) for providing seeds of the abi4-1 mutant.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.