ShHTL7 requires functional brassinosteroid signaling to initiate GA-independent germination

ABSTRACT

In the model plant Arabidopsis thaliana, two mutually antagonistic hormones regulate germination: abscisic acid (ABA) which promotes dormancy and gibberellins (GA) which breaks dormancy. Mutants auxotrophic for or insensitive to GA do not germinate. However, changes in the signaling flux through other hormone pathways will permit GA-independent germination. These changes include increased brassinosteroid (BR) signaling and decreased ABA signaling. Recently, strigolactone (SL) was also shown to enable GA-independent germination, provided the seeds express the SL receptor ShHTL7 from the parasitic plant Striga hermonthica. Here we show that a mutation which reduces sensitivity to BR (bri1-6) prevents ShHTL7 from promoting GA-independent germination. Further, we show that neither ShHTL7 nor the constitutive karrikin signaling mutant smax1-2 confer insensitivity to ABA. These results suggest ShHTL7 requires functional BR perception to bypass the GA requirement for germination.

Introduction

Plants tie their seed dormancy to external cues such as light, moisture, and temperature to ensure germination only occurs in favorable environments.^1,2 We still do not fully understand how seeds transduce these environmental signals and integrate them into a single decision. However, we do know that the environment alters the balance of two, mutually antagonistic hormones: the dormancy promoting abscisic acid (ABA) and dormancy breaking gibberellins (GA).^1,2 Physiological analysis has shown that stresses such as high temperature cause a significant increase in the ABA/GA ratio, which in turn prevents germination.³ Mutant analysis also lends support to the view that the ABA/GA ratio controls germination. Seeds auxotrophic for or insensitive to gibberellins will not germinate; but, a corresponding decrease in ABA synthesis or perception will enable these GA auxotrophs to germinate.^4-6 Thus, the ratio and not the absolute concentrations of these two hormones determines whether a seed will germinate.

Despite the importance of the ABA/GA ratio, other hormones also play a role in Arabidopsis germination. Among these is brassinosteroid (BR), a steroid hormone with diverse roles throughout plant development, including seedling morphogenesis, root development, and pollen maturation.⁷ Prior research has shown that that seeds auxotrophic for or insensitive to BR have enhanced dormancy and sensitivity to ABA. Interestingly, exogenous application of BR can rescue the germination defect of seeds auxotrophic for GA.⁸ Recent work has begun to show how BR promotes germination. Zhao et al. have found that the positive regulator of BR signaling BRI1 EMS SUPPRESSOR 1 (BES1) can directly bind to and inhibit the dormancy promoting transcription factor ABA INSENSITIVE 5 (ABI5).⁹ Conversely, the negative regulator of BR signaling BRASSINOSTEROID-INSENSITIVE 2 (BIN2) can phosphorylate and thereby stabilize ABI5 during seed germination.¹⁰ In addition, BES1 also represses ABA INSENSITIVE 3 (ABI3) expression, which in turn regulates ABI5 expression in the seed.¹¹ Together, these lines of evidence suggest that BR promotes germination by antagonizing ABA signaling.

Unlike Arabidopsis, the parasitic plant Striga hermonthica uses the hormone strigolactone (SL) to break dormancy.¹² However, SL can also trigger germination in Arabidopsis, provided the seeds express the SL receptor HYPOSENSITIVE TO LIGHT 7 (ShHTL7) from Striga.^13,14 We know that ShHTL7 functions in Arabidopsis by hijacking the karrikin signaling pathway.¹⁵ Upon binding its ligand, ShHTL7 and the F-box, MORE AXILLARY BRANCHES 2 (MAX2) inhibit the activity of the downstream repressor SUPPRESSOR OF MAX2 1 (SMAX1).¹⁵ Importantly, application of SL to ShHTL7 expressing seeds enables them to germinate without GA.¹⁵ Because changes in BR or ABA signaling are sufficient to germinate GA auxotrophs, we asked whether SL modulates the activities of these hormone pathways during germination and seedling development. Here we show that ShHTL7 requires functional BR signaling to initiate germination and that it does not confer insensitivity to ABA.

Results and discussion

To determine whether SL dependent germination requires BR perception, we crossed a ShHTL7 overexpressing line to the BR insensitive mutant brassinosteroid insensitive 1–6 (bri1-6). From this cross we generated the doubly homozygous bri1-6; ShHTL7 mutant. We then tested the germination of this mutant in the presence of the GA biosynthesis inhibitor paclobutrazol (PAC) and the synthetic strigolactone GR24. As shown in Figure 1a, we found that the bri1-6; ShHTL7 double mutant had significantly reduced germination compared to the parental ShHTL7 line. Despite this result, however, both the bri1-6 and bri1-6; ShHTL7 lines could germinate on minimal media, which indicates that BRI1 is not absolutely required for germination.

Figure 1.

ShHTL7 interactions with BR and ABA. (a) Germination of the bri1-6 mutant overexpressing ShHTL7 on 20 µM PAC and 1 µM GR24. Note, all ShHTL7 overexpressing lines arise from the same transgenic event. Asterisks indicate a significant difference from ShHTL7 (P < .01, analysis of variance (ANOVA) with post-hoc Tukey’s honest significant difference (HSD) test). (b) Cotyledon greening of smax1-2 and ShHTL7 in the presence of 2 µM ABA. Asterisks indicate a significant difference from Col-0 (P < .01, analysis of variance (ANOVA) with post-hoc Tukey’s honest significant difference (HSD) test). (c) Photographs of seedling phenotypes from (b). Bar represents 1 mm

Next, we asked whether SL signaling conferred resistance to ABA. We tested the effect of 2 µM ABA on the constitutive karrikin signaling mutant smax1-2 and the ShHTL7 mutant sown in the presence of GR24. As shown in Figure 1b and c, both the smax1-2 and ShHTL7 mutants retained wild type sensitivity to ABA. By comparison, both the ABA insensitive mutants abi1-1 and abi4-5 were fully resistant to the effects of 2 µM ABA.

Here we have shown that ShHTL7 requires BRI1 to initiate GA-independent germination. Three points about this result require emphasis. First, even though seeds do not require BRI1 to germinate under standard conditions, its function becomes indispensable under GA limiting conditions. This suggests that in favorable environments seeds may have multiple, redundant routes to initiate germination; but if one pathway becomes blocked, the functionality of the alternative pathways become essential. This phenomenon is in many ways analogous to synthetic lethality observed in yeast, where two non-lethal gene deletions become lethal when combined in a double mutant.¹⁶ In this case, seeds may dispense with either the GA pathway or the BR pathway and still germinate; but the combined absence of both pathways renders germination impossible.

Second, the positive interaction between ShHTL7 and BR signaling is likely limited to the seed. Here we have shown that both ShHTL7 and BRI1 work in the same direction: promoting germination. However, these two pathways have opposing effects during other developmental stages. For example, both ShHTL7 and its Arabidopsis homolog HYPOSENSITIVE TO LIGHT/KARRIKIN INSESNSITIVE 2 (HTL/KAI2) enhance photomorphogenesis by inhibiting hypocotyl elongation.¹⁷ By contrast, BRs promote skotomorphogenesis and hypocotyl elongation.⁷ As a result, bri1 loss-of-function mutants have short hypocotyls, whereas htl/kai2 loss-of-function mutants have long hypocotyls.^17,18 Additionally, knockdown of the positive BR response regulator BES1 suppresses the long hypocotyl phenotype of max2-1 loss-of-function mutants.¹⁹ Taken together, these findings suggest that the relationship between karrikin and BRs changes during development. In some stages both pathways work together, but in other stages they oppose each other.

Third, although bri1-6 suppresses the ShHTL7 overexpression phenotype, this does not necessarily imply that BRI1 functions downstream of ShHTL7. Prior work has shown that BRs inhibit the expression of ABI3, a gene which helps embryos establish dormancy.¹¹ Consequently, bri1-6 seeds may have elevated ABI3 expression and therefore a deeper dormancy which ShHTL7 cannot overcome. This possibility, along with the opposing effects of BRs and karrikins in other developmental stages suggest bri1-6 suppresses ShHTL7 but is not necessarily epistatic to ShHTL7.

In addition to establishing a link between BR and SL during germination, we have also shown that ShHTL7 does not confer ABA insensitivity to Arabidopsis. Previous studies have shown that mutants insensitive to ABA can germinate in the absence of GA.⁶ Our results suggest that ShHTL7 initiates GA-independent germination through a different mechanism than that employed by ABA insensitive mutants.

Materials and methods

Plant materials and growth conditions

All mutants, except abi1-1 and bri1-6, are in the Col-0 background. abi1-1 and bri1-6 are in the Ler-0 and En-2 ecotypes respectively. All plants were grown under a 24 h. light regime. Harvested seeds were after ripened in darkness at 25°C for at least one month.

Strain construction

The generation of the ShHTL7 mutant is previously described.¹³ All crosses were set up using the standard procedure.²⁰ F₁ seeds were collected, re-planted, and allowed to self-fertilize. The bri1-6 x ShHTL7 F₂ seed population was sown on ½ Murashige and Skoog (MS) minimal agar medium and stratified for 4 days at 4°C. The seeds were then exposed to 6 hours of white light then placed in a dark cabinet for 5 days. Seedlings with the characteristic bri1 short hypocotyl were transferred to plates with 20 µg/mL glufosinate ammonium and grown under the light for 5 days. Resistant seedlings were propagated; all adult plants were checked for the bri1-6 dwarf phenotype. The bri1-6; ShHTL7 homozygote was identified by progeny testing F₃ populations for segregation of the BASTA resistance marker.

Germination assays

Germination assays were carried out on ½ MS minimal agar media. Stocks of GR24 were prepared in DMSO, PAC in ethanol, and ABA in methanol. For each experiment, a working stock 1,000 times the experimental concentration was prepared. Each stock was then diluted 1:1,000 in ½ MS to a percentage of 0.1 (v/v). Seeds surface sterilized with 70% ethanol were then added to the plates. The seeds were stratified for 4 days at 4°C then placed under continuous white light at 25–26°C for 7 days. The plates were then scored for either germination or resistance to ABA.

Funding Statement

This work was supported by the Natural Sciences and Engineering Research Council of Canada [507992, 06752] andthe New Frontiers in Research Fund Exploration Award [2018-00356].

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.