Does zaxinone counteract strigolactones in shaping rice architecture?

ABSTRACT

The cleavage of plant carotenoids leads to apocarotenoids, a group of metabolites including precursors of the hormones strigolactones (SLs) and abscisic acid, regulatory and signaling molecules. Zaxinone is a recently discovered apocarotenoid growth regulator that improves growth and suppress SL biosynthesis in rice (Oryza sativa). To test if zaxinone also counteracts the growth regulatory effects of SLs in rice, we co-supplied zaxinone and the synthetic SL analog rac-GR24 to the rice SL-deficient DWARF17 (d17) mutant. Results showed that co-application of GR24 and zaxinone still rescued d17 phenotype, indicating that zaxinone and GR24 act independently in regulating root and shoot growth and development in rice.

KEYWORDS:

• Strigolactones
• Zaxinone
• Apocarotenoids
• Rice (Oryza sativa)

Strigolactones (SLs) are a novel plant hormone that determines plant architecture and mediates communications in the rhizosphere.^1,2 Structurally, SLs are carotenoid derivatives recognized by a lactone ring (D-ring, Figure1a) that is linked by an enol-ether bridge in R-configuration to a second moiety.¹ While the D-ring and the enol-ether bridge are strictly conserved in all natural SLs and essential for SL activity,³ there is large variations in the structure of the second moiety, which is the basis for distinguishing the canonical SLs that contain a tricyclic lactone (ABC-ring) from the non-canonical ones that contain various structures as a second moiety (Figure 1a). The biosynthesis of SLs begins with the reversible isomerization of all-trans- into 9-cis-β-carotene by the isomerase DWARF27.^4,5 Consecutive cleavage and rearrangement reactions accomplished by the Carotenoid Cleavage Dioxygenase 7 (CCD7/D17) and CCD8 (D10) lead to carlactone (CL), the central intermediate in SL biosynthesis.^2,4,6–8 CL is the substrate of cytochrome P450 monooxygenases (CYP), including MORE AXILLARY GROWTH1 (MAX1) enzymes that belong to the CYP711A clade, which are involved in the canonical and non-canonical SL formation⁹ (Figure 1a).

Figure 1. Characterization of compound effects on the rice d17 mutant at seedling stages. (a) Proposed biosynthesis pathway of canonical SLs in rice. Abbreviations: D, Dwarf; CCD, Carotenoid Cleavage Dioxygenase; MAX, More Axillary Growth; Os900, OsMAX1-900; Os1400, OsMAX1-1400. (b-c) Phenotypic characterization of d17 mutant fed with zaxinone (Zax) and rac-GR24. Roots of hydroponically grown d17 mutant seedlings in the absence (Mock), presence of zaxinone (2.5 µM), rac-GR24 (1 µM) and of zaxinone (2.5 µM) combined with rac-GR24 (1 µM). Data represent the mean ± SD for 6 biological replicates. Statistical analysis was performed using one-way analysis of variance (ANOVA) and Tukey’s post hoc test. Different letters denote significant differences (P < .05). Scale bars at (b): up 10 cm and down 1 cm.

All types of SLs were generally considered as shoot branching inhibitors, as SL-deficient mutants are characterized by a high-branching and dwarf phenotype.^1,10,11 However, we recently reported that canonical SLs, i.e. 4-deoxyorobanchol (4DO) and orobanchol, are important rhizospheric signals but not the predominant tillering regulators in rice (Oryza sativa).¹² Consistently, the orobanchol-deficient cyp722C tomato (Solanum lycopersicum) mutant does not display high branching and dwarf phenotype.¹³ In fact, canonical SLs were originally identified as rhizospheric signals inducing seed germination of root parasitic plants and released into soil, particularly upon phosphate deficiency, to successfully recruit arbuscular mycorrhizal (AM) fungi for establishing AM symbiosis.^14–16

The apocarotenoid growth-regulator zaxinone, produced by Zaxinone Synthase (ZAS) in rice, was identified as a negative and positive regulator of SL biosynthesis in rice and Arabidopsis, respectively.^17,18 Interestingly, its activity in improving rice performance requires functional SL biosynthesis and signaling machinery.¹⁷ In addition, rice zas mutants showed lower tillering phenotypes accompanied by increased 4DO exudation, which could be restored by exogenous zaxinone application.^17,19 Rice contains five MAX1 enzymes, among which Os900 mediates the formation of the rice canonical SLs, 4DO and orobanchol. So far, max1-900 (Os900) mutant lines are the first rice SL-biosynthetic mutants that do not exhibit a pronounced SL-deficiency-related architectural phenotype.¹² This observation implies that 4-DO and orobanchol are not the main regulator of rice shoot architecture and that this trait is rather determined by non-canonical SLs.^12,20 The functionality of the SL signaling pathway in the Os900 mutants was confirmed by zaxinone application that led to improved root and shoot growth¹² and reduced the transcript level of SL biosynthetic genes.^17,21 However, the question how zaxinone causes this effect on SL biosynthesis remained elusive. In this communication, we asked whether zaxinone generally acts as antagonist of SL in regulating growth and development of rice at early stages, particularly tillering.

To answer this question, we treated seedlings of the SL-deficient d17 (ccd7) mutant with 2.5 μM zaxinone, 1 μM rac-GR24, a synthetic SL analog¹ or with a combination of both (Figure 1b). As expected, application of GR24 alone fully rescued d17 phenotype, increasing shoot and root length while decreasing the number of tillers. Co-application of GR24 and zaxinone led to the same effect, while application of zaxinone alone failed to rescue the d17 phenotype (Figure 1c). The latter result confirmed that the growth-promoting activity of zaxinone depends on the presence of intact SL biosynthesis, but it also showed that zaxinone likely does not counteract the effect of SLs, as it did not affect the rescuing effect of GR24 in the d17 mutant. This observation suggested that zaxinone might not directly interfere with SL signaling response or that the effect of SLs is much more dominant than that of zaxinone. Indeed, rac-GR24 also function on karrikin pathway²² and, possibly, the observed effects might be the results of karrikin signaling.

Notably, SL biosynthesis is barely expressed under normal growth conditions.¹⁷ To determine the impact of zaxinone on the expression of SL biosynthetic genes under normal conditions, we re-analyzed our related RNAseq dataset.²³ Interestingly, none of SL biosynthesis and signaling transcripts were significantly affected by zaxinone (Figuire 2). Although we cannot rule out the possibility that zaxinone indirectly affects SL pathways via sugar or cytokinin metabolism under normal conditions,²³ our data demonstrate that zaxinone does not antagonize the shoot architecture determining activity of SLs in rice, at least at the concentrations applied. Future investigations are required to shed light on the interaction between SLs and zaxinone at molecular level.

Figure 2. Transcriptome analysis of wild-type rice root tissues in response to zaxinone treatment at different time points. Differentially expressed genes (DEGs), following Deseq2 analysis, revealed the SL biosynthesis and signaling gene expression pattern with log₂FoldChange (Log₂FC). Data were extracted from Wang et al. ²³ Numbers in the color box were Log₂FC upon zaxinone treatment. Abbreviations: D, Dwarf; CCD, Carotenoid Cleavage Dioxygenase; MAX, More Axillary Growth.

Material and methods

Plant material

d17²⁴ and its corresponding WT Nipponbare rice plants were surface-sterilized, cultivated, and grown under controlled conditions (a 12 h photoperiod, 200-µmol photons m⁻² s⁻¹ and day/night temperature of 27/25°C) according to the published protocol.²⁵ 7-day-old seedlings were transferred into black falcon tubes filled half-strength modified Hoagland nutrient solution (nutrient compositions listed in²⁵) with adjusted pH to 5.8 with KOH.

Exogenous applications of zaxinone and GR24

For investigating the effect of zaxinone and GR24 on d17 seedlings, 7-day-old rice seedlings were grown hydroponically in 1/2 Hoagland nutrient solution containing 2.5 µM zaxinone (obtained from Buchem B.V.; Apeldoorn, The Netherlands), 1 µM rac-GR24 (purchased from StrigoLab; Turin, Italy), or the corresponding volume of the solvent (mock; acetone) for 2 weeks. The solution was changed twice per week, adding the chemical at each renewal.

RNA-seq data analysis

RNAseq dataset was extracted from Wang et al. ²³ and the reads were aligned to the O. sativa genome v7.0 (http://phytozome.jgi.doe.gov/; Phytozome v12.1). Data processing and analysis were performed using the LSTrAP workflow,²⁶ and differential gene expression (DGE), read counts from HTSeq, were analyzed by DESeq2.²⁷

Author Contributions

S.A.-B. and J.Y.W. proposed the concept and designed the experiments. J.Y.W. and J.B. conducted experiments. J.Y.W., J.B., and S. A.-B. analyzed and discussed the data. All authors read, edited, and approved the manuscript.

Acknowledgments

We are grateful to the members of the Bioactives lab in KAUST for their helpful discussions and assistance.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

All data generated or analyzed during this study are included in this published article.

Additional information

Funding

This work was supported by baseline funding from King Abdullah University of Science and Technology and by the Bill & Melinda Gates Foundation (grant number OPP1136424) given to S. A.-B.