A model for combinatorial regulation of the petunia R2R3-MYB transcription factor ODORANT1

Abstract

The emission of floral volatiles requires coordinated expression of biosynthetic genes. In the regulatory network of the volatile benzenoid/phenylpropanoid pathway in Petunia hybrida two master regulators of the pathway have been identified. The R2R3-MYB transcription factor EMISSION OF BENZENOIDS II (EOBII) utilizes a specific MYB binding site to activate the expression of the R2R3-MYB ODORANT1 (ODO1). However, because EOBII is expressed early in flower development, when ODO1 is not, there must be other factors that play a role in regulating expression of ODO1. Through functional analyses of ODO1 promoter fragments from fragrant and non-fragrant flowers, we provide evidence for additional players and present a model for combinatorial regulation of ODO1 expression in Petunia.

Floral volatile production and emission in plants is a dynamic process, depending on various factors such as the developmental program and time of day.¹ Consequently, transcript accumulation of floral fragrance genes increases during petal development, peaking at anthesis,^2-7 and generally follows a diurnal/nocturnal rhythm corresponding to the activity of its pollinators.⁸ Where and when genes are expressed largely depends on transcription factors (TFs) that bind to their promoters. The hierarchical organization of TFs is a common feature in transcriptional regulatory networks and the identification of master regulators and their target genes is thus crucial for understanding and ultimately manipulating complex biological processes such as plant metabolism.⁹ Although genes can be activated by a single TF, often multiple regulators are involved, enabling modulation of the appropriate level and time of expression. In addition, combinatorial regulation of genes entails that a given TF can regulate distinct genes involved in the same, but also in different processes.¹⁰

Recently, TFs regulating floral fragrance biosynthesis and some of their target genes have been identified in Petunia hybrida.^11-13 Silencing of the Petunia R2R3-MYB TF ODO1 resulted in a severe reduction in production and emission of most volatile benzenoids and phenylpropanoids.¹³ ODO1 can trans-activate the 5-enolpyruvylshikimate-3-phosphate synthase (EPSPS) promoter, confirming its importance for substrate supply via the shikimate pathway.¹³ Silencing of the R2R3-MYB TF EMISSION OF BENZENOIDS II (EOBII) in Petunia downregulates genes of the shikimate and phenylpropanoid pathways, and importantly, ODO1.¹² However, transient overexpression of EBOII in petunia petals did not result in higher ODO1 transcript levels at the peak of ODO1 expression.¹² Nevertheless EOBII was identified as a direct activator of ODO1 and was shown to bind and activate the ODO1 promoter at a MYB binding site located in an enhancer region between approximately 1.2 and 1kbp upstream of the start codon.¹⁴ In addition, EOBII has been shown to be involved in petal opening.¹⁵ Because EOBII is expressed early in flower development,¹² the expression pattern of EOBII does not fully overlap with that of ODO1 and another target gene, isoeugenol synthase (IGS). This suggests that additional factors are needed to activate the ODO1 promoter. Alternatively, ODO1 expression could be repressed during early stages of flower development or the ODO1 promoter could not be accessible for activators due to the chromatin conformation.

In order to test activation at the ODO1 promoter, reporter constructs were made with the ODO1 promoter (M10:GUS, M12:GUS) of an ODO1-expressing Petunia line (P. hybrida cv Mitchell), an ODO1 non-expressing Petunia line (P. hybrida cv R27, R12:GUS), two chimeric constructs (R2M10:GUS and M2R10:GUS) and a promoter construct with a mutation in the EOBII-binding site (mutM12:GUS) (Fig. 1). These were on the one hand agroinfiltrated in Petunia x hybrida cv Mitchell petal limbs and on the other hand in leaves of Nicotiana benthamiana, together with EOBII driven by the 35S promoter (Fig. 1).¹⁴ The M12:GUS construct is highly active in Petunia and is strongly activated by EOBII in N. benthamiana. In contrast, the R12:GUS construct, which does not contain the EOBII-binding site, is only weakly activated (Fig. 1).¹⁴ In our proposed model, the unknown factor Y co-activates the M12, but not the R12 promoter, outside the previously identified enhancer region. This factor is necessary but not sufficient for high activity and can be either the Petunia EOBII or a homologous N. benthamiana factor (Fig. 1A). However, there is no obvious EOBII binding-site in this part of the Mitchell promoter.¹⁴ The chimeric M2R10:GUS construct is activated in petunia petals but not in N. benthamiana leaves by EOBII (Fig. 1A and B). This suggests that in addition to EOBII and Y, an additional petunia factor X is needed for high promoter activity, suggesting combinatorial regulation at the ODO1 promoter. From this result, we conclude (1) that factor X recognizes an element within the enhancer region and (2) that it is dependent on the binding of EOBII (Fig. 1B). In a yeast-one-hybrid experiment using the 240 bp enhancer region (located between app. 1.2 and 1 kbp upstream of the start codon) as a bait, we were unable to fish out EOBII or any other factor from a fragrant petunia petal cDNA library (Van Moerkercke, unpublished results). This suggests that EOBII binding could require a full-length promoter, i.e., the 1.2 kb upstream region, for binding. In addition, the proposed factor X may only be able to bind the promoter fragment in the presence of EOBII.¹⁶ These new findings may be of help to develop new strategies aiming to identify new regulators of ODO1 and the floral scent pathway in Petunia.

Figure 1. Model for combinatorial regulation at the ODO1 promoter. Trans-activation of selected ODO1 promoter:GUS constructs by ectopically expressed EOBII in (A) N. benthamiana leaves and (B) petunia petal limbs. Red and gray bars correspond to R27 and Mitchell promoter fragments, respectively. The bar on top shows the length of the promoter fragments upstream of the ATG start codon. The strength of the GUS activities of the reporter constructs is proportional with the number of plus signs (+), but should not be compared between experiments depicted in (A and B). The blue lines in M12:GUS and red lines in mutM12:GUS represent the wild type and mutated EOBII activation site, respectively. The enhancer region in the Mitchell promoter is indicated and located between 1.2 and 1 kbp upstream of the start codon. E, EOBII; Y, unidentified factor in both N. benthamiana leaves and Petunia petals and X, unknown petunia-specific factor. M, promoter fragment corresponding to the Mitchell promoter; R, promoter fragment corresponding to the R27 promoter.

Conclusion

Given the strict regulation of volatile emission from petunia flowers, tight control of the expression of the relevant genes, especially of those that are master regulators of the pathway is crucial. Here, we propose that in addition to EOBII an unknown factor X is needed for high ODO1 expression. This could explain why ODO1 is not expressed early in flower development when EOBII is expressed. In addition, if the unknown factor is limiting for ODO1 activation in petunia petals, it could partly explain why overexpression of EOBII did not result in significant higher ODO1 levels, as was reported by Spitzer-Rimon et al. (2010).¹² As an alternative to, but not excluding this hypothesis, a repressor that inhibits ODO1 expression and putatively interferes with factor X might be present during early stages of flower development.

Glossary

Abbreviations:

ODO1

ODORANT1

EOBII

EMISSION OF BENZENOIDS II

TF

transcription factor

GUS

β-glucuronidase

Van Moerkercke A, Haring MA, Schuurink RC. The transcription factor EMISSION OF BENZENOIDS II activates the MYB ODORANT1 promoter at a MYB binding site specific for fragrant petunias. Plant J. 2011;67:917–28. doi: 10.1111/j.1365-313X.2011.04644.x.