Modulation of Phytohormone Signaling: A Primary Function of Flavonoids in Plant–Environment Interactions

1Flavonoids, a vast class of phenylpropanoids comprising more than 8,000 structures with a wide range of decorations, have the ability to absorb the most energetic solar wavelengths reaching the leaf surface. Flavonoids have long been considered as primarily synthesized to constitute an effective shield against the penetration of UV-B radiation to sensitive leaf tissues, and greatly involved in protecting plants challenged by the depletion of stratospheric ozone layer (for a review, see ). Though this is obvious, as the UV-B (and UV-A) screening properties of flavonoids are long known, recent evidence suggests flavonoids may serve other ‘eco-physiological’ functions in early as well as in modern terrestrial plants.

2First, it is noted that all phenylpropanoids, not only flavonoids, have the ability to absorb wavelengths over the UV-B region of the solar spectrum. For instance, hydroxycinnamic acid derivatives have greater molar extinction coefficients (𝜀) than flavonoids over the UV-B waveband. Nonetheless, the biosynthesis of flavonoids occur at the expense of hydroxycinnamate biosynthesis in UV-B-treated leaves.

3Second, flavonoids have maximum 𝜀s over the 330–355 nm, UV-A region of the solar spectrum, so that the induction spectrum most effective for their biosynthesis (UV-B) does not overlap with their absorption spectrum maxima, thereby suggesting that flavonoids are not synthesized to primarily serve UV-B screening functions in UV-B-treated tissues. This conforms to the observation that the biosynthesis of flavonoids is activated to a similar degree by high solar irradiance, in the presence or in the absence of UV-radiation. Moreover, blue light has been recently reported as being more effective than UV-B light in promoting the biosynthesis of flavonoids. Finally, flavonoids accumulate to a very similar extent in plants exposed to excess both NaCl and Cu2+ or upon exposure to high UV-B irradiance. This is in line with the notion that changes in the redox potential of the cell activate flavonoid (particularly flavonol) biosynthesis, consistent with relatively old findings that R2R3MYB transcription factors that regulate the biosynthesis of flavonoids, are themselves redox controlled (reviewed in ).

4Third, the biosynthesis of flavonoids with ortho-dihydroxy B-ring substitution is strongly favored as compared to the biosynthesis of monohydroxy B-ring-substituted flavonoids in response to excess light, irrespective of the relative proportions of solar wavelengths reaching the leaf surface (for a review, see ). Monohydroxy and dihydroxy flavonoids do not display different UV-screening abilities, but largely differ in their ability to scavenge reactive oxygen species (ROS), i.e., to behave as antioxidants. This is in line with old and recent views that flavonoids may play major functions as antioxidants in plants exposed to excess light and, hence, to photooxidative stress of largely different origin. It is a part of the folklore in plant photobiology and plant ecology that UV-B-treated leaves display a steeply higher quercetin to kaempferol (or apigenin) ratio than untreated leaves, and this is for equipping leaves with a more effective antioxidant, not UV-screening potential.

5The antioxidant properties of flavonoids have long been invoked to explain their health promoting effects in animals (;, for critical reviews), though flavonoid aglycones may also behave as pro-oxidants in high concentrations. Since flavonoids are usually glycosylated in plant cells, thereby increasing their solubility and facilitating their transport from the endoplasmic reticulum (ER, the site of their biosynthesis) to different cellular organelles, their pro-oxidant actions are likely of minor significance in plants. There is an increasing body of evidence indicating that flavonoids may exert complex functions in both animal and plant cell metabolism, going well-beyond the mere chemical quenching of ROS. Certain flavonoids, especially but not limited to quercetin derivatives, have the ability to modulate signaling cascades that regulate cell growth and differentiation. For instance, flavonoids may strongly affect the protein kinase [e.g., mitogen-activated protein kinases (MAPKs)] signaling cascades, which are at the base of many disorders, including carcinogenesis, in humans. In other words, flavonoids may behave as signaling molecules, and in a concentration range much lower than that required to both scavenge ROS and effectively absorb UV-radiation. Though the ability of flavonoids to regulate cell growth and differentiation in plants has been known for decades, the functional significance in the adaptive mechanisms of plants to unfavorable habitats has been explored in less detail.

6Here, we focus our discussion on old and recent evidence of the strong relationship between flavonols (the ancient class of flavonoids already present in mosses, ), auxin (IAA,; ) and abscisic acid (ABA;, ). We hypothesize that the regulation of phytohormone signaling by flavonoids might have represented a function of primary significance for the conquest of land by plants and it is still of major value for the successful acclimation of modern terrestrial plants to a severe excess of radiant energy.

7One of the molecular innovations that accompanied the water-to-land transition of plants was the replacement of mycosporine like aminoacid (MAA) with the flavonoid, particularly flavonol metabolism. The functional reason(s) for such drastic metabolic modification was unlikely used for equipping early land plants with a more effective UV-B shield as compared to algae: MAA are at least as efficient as flavonoids to absorb UV-B radiation. The replacement of ‘nitrogen-rich’ with carbon-based UV-screening pigments might have been functional when early land plants moved to nutrient poor environments. There is consensus that flavonols, particularly derivatives of quercetin, may have enhanced the ability of early land plants to take-up water and nutrients, acting on soil chemistry and, particularly by promoting the symbiosis with nitrogen-fixing bacteria and the plant–mycorrhiza association. The plant–mycorrhiza association represented a central event at the origin of land flora. Notably, flavonols play the peculiar role of inhibiting auxin transport during nodulation, thus determining increases in local auxin levels that, in turn, promote nodule organogenesis. This adds further support to the old hypothesis that the origin of land plants resulted from “a union of alga and fungus advanced by flavonoids”.

8Flavonols are well-suited to modulate auxin transport and signaling, given their ability to affect the activities of a wide range of proteins, as well as to scavenge ROS. Ortho-dihydroxy B-ring-substituted flavonoids, such as quercetin, are particularly effective in affecting the activity serine–threonine PINOID (PID) proteins that control the localization of PINFORMED (PIN) auxin efflux facilitator proteins (PIN) proteins (Figure 1;,;; ). Nonetheless, the significance of flavonoids as direct or indirect inhibitors of auxin transporters in the modulation of auxin gradients at cellular and inter-cellular levels and, hence, of auxin signaling has been questioned in some instances (e.g., ). Instead, flavonoids might determine gradients in auxin concentration at both cellular and tissue level by primarily modulating auxin catabolism, i.e., acting as direct and/or indirect scavengers of ROS. However, how the ROS-scavenging capacities of flavonols may have an impact on auxin signaling is a matter that deserves further investigation. Antioxidant flavonoids may limit IAA-oxidation, by inhibiting the activity of DIOXYGENASE for AUXIN OXIDATION1 protein (DAO1, ), a member of the 2-oxoglutarate and Fe(II)-dependent [2OG Fe(II)] oxygenase superfamily (Figure 1). Furthermore, flavonols may both reduce IAA radicals generated during IAA-oxidation and chelate Mn(II), which is a cofactor of IAA-oxidase. As noted above, this may be responsible for determining auxin gradients and the consequential developmental processes. There is evidence, however, that increased IAA levels may promote ROS formation and then stimulate IAA oxidation, thereby transiently repressing auxin signaling. Flavonoids might act as general buffers of cellular ROS levels (though how and how much flavonoids may regulate the cytoplasmic ROS levels involved in IAA-oxidation and IAA-signaling is far from being fully elucidated (Figure 1; ), thus rendering plants in a “receptive state” capable to respond promptly to changes in environmental conditions. Similarly, stress-induced increase in ROS, e.g., H2O2, levels may activate specific MAPK, such as ANP1 kinase in Arabidopsis and NPK1 in tobacco which, in turn, repress auxin signaling while transducing oxidative stress signaling. In other words, the massive generation of H2O2 triggers ANP1-mediated MAPK cascade, and may help stressed plants to divert energy from auxin-related activities to stress protection.

9We speculate that drastic changes in cellular redox homeostasis activate flavonoid (particularly flavonol) biosynthesis and, in turn, antioxidant flavonoids might act as components of a regulatory circuit of auxin signaling pathway. It is worth noting that flavonoids accumulate greatly indeed in regions of high auxin concentration. Further support to our hypothesis is the observation that antioxidant flavonoids have long been reported as having a nuclear location (and perhaps synthesized in the nucleus, for a review see; ) and, hence, optimally suited to strongly affect MAPK activities. This may be of great significance under stressful conditions, when MAPKs re-distribute from the cytoplasm to the nucleus, thus assisting cell re-programming (; Figure 1).

10There is relatively recent evidence that a subclade of PIN proteins, characterized by a shorter hydrophilic domain, such as PIN5, PIN6 and PIN8, as compared to ‘long’ plasma membrane (PM) PINs, are located at the ER. Notably, the auxin metabolic pathway is also compartmentalized to the ER, as evidenced by the presence of several auxin metabolic enzymes and regulatory proteins in the ER. These ER-located PIN proteins were already present in early land plants, such as Physcomitrella patens and Selaginella moellendorffii, thereby suggesting that mediating auxin homeostasis at the ER is the ancestral function of PIN proteins. In particular, PIN5 escorts auxin from the cytoplasmic face of the ER (the site of auxin biosynthesis) to the ER lumen (Figure 1), therefore both increasing auxin compartmentation and establishing auxin gradients at cellular level. Cell-to-cell auxin transport at the PM likely occurred at a later stage during land plant evolution and involved long PINs (e.g., PIN1 and PIN2;;; Figure 1). Notably, the cytoplasmic face of the ER is also the site of flavonoid biosynthesis, and both ATP binding cassette (ABC)-type and multidrug resistance and toxic ion extrusion (MATE) proteins assist the ‘transport’ of flavonoids to the ER lumen (reviewed in; Figure 1). It is conceivable that the strong relationship between flavonoids (particularly flavonols) and auxin may constitute an ancestral feature of land plants, possibly contributing to enhance their ability to adapt, i.e., evolvability sensu stricto, to an ever-changing environment.

11We hypothesize that the regulation of auxin both movement and signaling, through the regulation of ROS levels and protein activities (Figure 1), i.e., signaling functions sensu lato was a function of primary significance served by flavonoids during the colonization of land by plants. These signaling functions are of key significance in the ecology of modern terrestrial plants as well, as capable of greatly affecting the organ as well as the whole-plant morphology. For instance, Arabidopsis transparent testa (tt) mutants, deficient in flavonoid biosynthesis, have elevated auxin transport and display phenotypes with largely impaired apical dominance. It has been speculated that UVR8-induced preferential activation of quercetin biosynthesis may well-contribute to the regulation of auxin signaling and the compact architecture (bushy phenotypes), of plants long-exposed to UV-B or full solar radiation. Antioxidant flavonoids have the potential to strongly control the plant architecture, and likely play an important role in the stress-induced (particularly high light stress) morphogenic responses (SIMR), the flight strategy of sessile organisms. It may be not a mere coincidence that environmental stimuli that result into a severe excess of radiant energy, and induce large alterations in the shape of individual organs and the whole-plant, almost exclusively promotes the biosynthesis of quercetin derivatives in all order of taxa, in the presence or in the absence of UV-radiation. This is also consistent with the observation that both the moss P. patens and the angiosperm A. thaliana respond very similarly to elevated doses of UV-B radiation, mostly enhancing the biosynthesis of quercetin derivatives. There is very recent evidence indeed that (1) both the moss P. patens and the liverwort M. polymorpha have functional UVR8 proteins that are regulated similarly to the Arabidopsis UVR8, and (2) in both bryophytes, UVR8 proteins mediate HY5 (ELONGATED HYPOCOTYL5) transcript expression and CHS (CHALCONE SYNTHASE) protein accumulation in response to UV-B. Notably, HY5 regulates the expression of MYB12 and MYB111 genes, known as PRODUCTION OF FLAVONOL GLYCOSIDES (PFG), in response to not only UV-B, but also to high white light.

12There is scarce evidence of PIN-flavonoid modulation of plant shape in bryophytes. However, recent findings show that the ‘ancestral’ PIN6 protein in Arabidopsis and PINA in P. patens have dual ER and PM localization. The finding that PINs proteins in P. patens are highly responsive to naringenin and strongly controls the shoot architecture is of significance, and opens new perspectives on similar regulation of plant shape by PIN/flavonoids in bryophytes and angiosperms.

13There is recent compelling evidence that flavonols, especially derivatives of quercetin may greatly affect the ABA-signaling pathway, by antagonizing the ABA-induced stomatal closure in both Arabidopsis and tomato. Arabidopsis guard cells rich in quercetin have reduced levels of H2O2 and, consequently, greater stomatal aperture than stomata deprived of quercetin. Similarly, the anthocyanin reduced (are) tomato mutant, which has low flavonol levels, displays higher ROS content and lower stomatal aperture as compared to the anthocyanin without (aw) mutant, which is rich in quercetin. H2O2 is a key second messenger in the ABA-signaling pathway, and it is necessary to promote stomatal closure (reviewed in ). It is conceivable that the antagonistic effect of quercetin on ABA-signaling might result not only from its ability to quench H2O2, but also through the inhibition of MAPKs activities, which act downstream of H2O2 in the regulation of guard cell movements (;; Figure 1). Our hypothesis is strongly supported by the observation that flavonols in Arabidopsis guard cells have cytoplasmic and especially nuclear distribution.

14Notably, ABA has also been reported to promote the biosynthesis of flavonols, other than the biosynthesis of anthocyanins. This is not surprising, as relatively recent experiments have shown a strong integration between ABA and light signaling, which may also help to explain the steeply enhanced flavonol biosynthesis in leaves exposed to high light irradiance, even in the absence of UV-radiation. There is intriguing evidence that increases in the levels of foliar ABA upon high light irradiance likely results from enhanced de-glucosylation of the ‘inactive’ ABA-glucoside (ABA-GE), and not from de novo synthesis through the plastidial MEP-pathway. The notion that β-glucosidase1 (BG1), which promotes the release of free-ABA from ABA-GE, is located at the ER, the site of flavonoid biosynthesis, is of the greatest interest. We speculate that the quick production of free ABA from ABA-GE might sustain steep increases in local ABA concentrations that are required for initiating early signaling responses to excess light stress, and these may well-include the biosynthesis of flavonols (Figure 1). The flavonol regulation of the ABA-signaling pathway may offer an additional chemical control to fine tune stomata movement. Nonetheless, how and how much the flavonols–ABA relationship may affect the acclimation/adaptive responses of plants to environmental pressures associated to habitats with contrasting soil water availability and solar irradiance is an interesting matter that needs further investigation. It is, however, plausible that the flavonol-regulation of stomata movements may have allowed early land plants to radiate toward habitats of increasingly excess of solar irradiance and contrasting soil water availability, thereby contributing to their adaptation on land.

15Here, we have hypothesized that the mutual regulation of phytohormone signaling and flavonol, particularly quercetin biosynthesis may have been of great value for the adaptation of plants on the land. It may be not a mere coincidence that mosses as well as modern terrestrial plants preferentially synthesize derivatives of quercetin in response to a severe light excess, despite the evolution of flavonoid metabolism has produced 1000s of structures. This is likely for equipping plants with a versatile metabolite capable of serving multiple functions. Here we have shown that quercetin has a superior ability to modulate ROS- and phytohormone signaling, but not a greater capacity to screen-off the most energetic (UV-B) solar wavelengths as compared to other flavonoids. Our reasoning supports the view that modulation of ROS and phytohormone signaling is likely an ancestral, robust function of quercetin (as well as of other ‘antioxidant’ flavonoids), since it has a persisted against environmental, stochastic and genetic perturbations, over an extended time-scale level.

16The observation that a wide range of abiotic stressors activates the biosynthesis of flavonoids, particularly of the di-hydroxy B-ring-substituted structures when plants suffer from severe excess of radiant energy, suggests that change in cellular ROS homeostasis is one of the main drivers for flavonoid biosynthesis. In turn, flavonoids with effective antioxidant potential (i.e., they are ROS scavengers at very low concentrations, ) may act as buffers of ROS levels, and enhance stress resistance. The ability of flavonoids to modulate ROS as well as phytohormone signaling (Figure 1) may help to explain their involvement in stress-induced morphogenic responses, a general response of plants to ‘fly away’ from adverse environmental conditions. For instance, ‘effective antioxidant’ flavonoids, distributed in the nucleus, the chloroplast, and the vacuole of mesophyll cells, constitutes more than 90% of the total flavonoid pool in plants inhabiting harsh, sunny habitats, and display bushy phenotypes, with small and thick leaves and short internodes. In contrast, monohydroxy-substituted, ‘poor antioxidant’ flavones and flavonols, mostly located in the adaxial epidermal cells, predominate in shade-adapted plants, whose phenotypes are associated to large, thin leaves and long internodes. These are, however, merely correlative observations, and further experimentation is needed to explore the significance of flavonol-phytohormone interactions in the mechanisms regulating the morphology and anatomy of individual organs and the whole plant under natural conditions.

17All authors listed have made a substantial, direct and intellectual contribution to the work, and approved it for publication.

18The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.