Fluorescence Resonance Energy Transfer Between Polyphenolic Compounds and Riboflavin Indicates a Possible Accessory Photoreceptor Function for Some Polyphenolic Compounds
ABSTRACT
The photoreceptive extreme tip of the wheat coleoptile exhibits intense green-yellow fluorescence under UV light, suggesting the presence of UV-absorbing materials. Fluorescence spectra of the intact coleoptile tip and tip homogenate showed the presence of the known photoreceptor pigments flavin and carotene, and a preponderance of phenolic compounds. Absorption spectra and fluorescence spectra of various phenolic compounds showed close overlap with the absorption and fluorescence spectra of the wheat coleoptile tip homogenate. Fluorescence spectra of several phenolic compounds showed close overlap with the absorption bands of flavin, carotene and pterine, suggesting possible energy transduction from phenols to these photoreceptors. Excitation of gentisic acid and ferulic acid with 340 nm light in the presence of flavin showed enhancement of flavin fluorescence in a concentration- and viscosity-dependent fashion, indicating fluorescence resonance energy transfer between them and riboflavin. Furthermore, several phenolic compounds tested generated superoxide anion on excitation at 340 nm, suggesting that superoxide-dependent signal cascades could operate in a polyphenol-mediated pathway. Phenolic compounds thus may act as accessory photoreceptors bringing about excitation energy transfer to the reactive photoreceptor molecules, or they may take over the function of the normal photoreceptor in genetic mutations lacking the system, or both processes may occur. The responses of plants to UV-B and UV-A light in mutants may be explained in terms of various phenolics acting as energy transducers in photoreceptor functioning.
INTRODUCTION
All organisms perceive, code, transmit and integrate environmental information that directs the cellular metabolism and developmental processes. The detection of different wavelengths of light by specific mechanisms plays a key role in plant development. The reception of the light stimulus by pigments that operate in the short wavelength regions and its intersystem crossing with one or more pigment systems responsible for blue light effects and the generation of early intracellular transduction events appear to be complex (1–3). The light receptors—photoreceptors as they are called—initiate a range of important responses in plants, such as phototropism, stomatal opening, inhibition of hypocotyl growth and accompanying expression of several genes.
Identification of the pigments responsible for photochemical processes has occurred to compare the action spectrum for the process with the absorption spectra of pigments suspected of being involved. Although these spectra never matched perfectly, this approach has identified flavins, carotenoids, pterin and zeaxanthin as probable blue light receptors. Various lines of evidence indicate that several photoreceptors must absorb in the UV-B (280–320 nm), UV-A (320 to 390 nm) and blue (390 to 500 nm) regions of the spectrum (1,3–5).
To date, four genes (CRY1, CRY2, PHOT1 and PHOT2) encoding the major UV-A/blue light receptors have been identified (6–9). Although UV-B inductive pathways appear to be different from UV-A/blue light pathways, distinct blue and UV-A pathways interact synergistically with the UV-B pathway to stimulate chal-cone synthase (CHS) expression in Arabidopsis (10). Similarly, Brassica napus grown under natural conditions displays a novel phototropic response to supplementary ultraviolet (UV-B and UV-A) radiation (11). Because the action maximum for phototropic curvature of the siliquae of B. napus (319 and 324 nm) did not match the major absorption peaks of flavin (270, 380 and 450 nm) or of carotene (450 nm and 480 nm), the possibilities of having either a novel UV-A/blue light absorbing pigment system or the operation of an intermediate UV-A absorbing and blue emitting molecule (or having both systems) performing fluorescence resonance energy transfer to the blue light receptor pigment are suspected. The coleoptile tip has a heavy load of phenolic compounds that absorb heavily in the UV-B and UV-A spectra. The generation of active oxygen species by blue light activation of photoreceptor has been shown to initialize blue light signal transduction (12). In this paper, we examine the possibilities of energy transfer between a selected set of phenolic compounds and riboflavin, and its relevance in relation to photochemical signal transduction in plants.
MATERIALS AND METHODS
Reagents
Riboflavin, β-carotene (trans), pterine, 3–4-dihydroxy benzoic acid, trans-cinnamic acid, gentisic acid, syringic acid, p-coumaric acid, caffeic acid, ferrulic acid, vanillic acid, m-coumaric acid, salicylhydroxamic acid, hydrocinnamic acid, hydrocaffeic acid, umbelliferone, quercetin, aesculetin, scopoletin, coumarin, rutin, morin, chlorogenic acid and PBN were purchased from Sigma (St. Louis, MO). Ethanol, methanol and hexane were purchased from E Merck (Darmstad, Germany).
Coleoptile preparation
Approximately 150 seeds of Triticum vulgare were wet in 100 mL of distilled water and were placed on five layers of Whatman-10 discs, placed in a plant growth dish (6-inch diameter) and then in a temperature-controlled (25 ± 1°C) dark room. The seeds were wet every day with 20 mL of distilled water. The coleoptile tips (approximately 1.5 cm long) were excised on Day 5 under red light (660 nm) for all experiments.
Fluorescence microscopy
Etiolated wheat coleoptiles were harvested (after first primordial leaves were pulled out from the coleoptiles) and sectioned longitudinally (approximately 50 μm) through the apex of the coleoptile. The sections from the center of the apex of the coleoptile tip were selected and then transferred onto the stage of an inverted microscope under UV light using a UV 1-A (DM 400) Nikon filter in total darkness. Exposures were taken simultaneously through a camera port fitted with a Nikon F-301. In the same way, the intact coleoptile was placed onto the stage of an inverted microscope under a UV1-A filter and photographed.
Absorption spectroscopy
Three hundred etiolated wheat coleoptiles were harvested in the dark under safe red light. First primordial leaves were pulled out from the coleoptiles. Coleoptile tips were then cut 2 mm in length from the top and homogenized in different organic solvents; viz., 85% ethanol, methanol and hexane. After homogenization, samples were centrifuged and supernatants were collected separately. The absorption spectra were assessed on a Shimadzu UV-1601 spectrophotometer.
A 33 mM solution of the test flavonoids/polyphenols (riboflavin, β-carotene [trans], pterine, 3–4-dihydroxy benzoic acid, trans-cinnamic acid, gentisic acid, syringic acid, p-coumaric acid, caffeic acid, ferrulic acid, vanillic acid, m-coumaric acid, SHAM, hydrocinnamic acid, hydrocaffeic acid, umbelliferone, quercetin, aesculetin, scopoletin, coumarin, rutin, morin and chlorogenic acid) were prepared in 50% ethanol and their absorption spectra were assessed using a Shimadzu UV-1601 spectrophotometer.
Emission spectroscopy
The emission spectra of the samples described above were taken using a Shimadzu RF 540 spectroflurophotometer. The excitation wavelengths used were 340 nm, 390 nm and 450 nm.
Spin trapping
A 1 μM solution of each of the test compounds was prepared in 50% ethanol. An aliquot of 90 μL of the solution described above was mixed with a superoxide radical spin trap PBN at a separate 50 mM final concentration (PBN was made in 50% ethanol). The samples were incubated for 15 min and electron spin resonance spectra were recorded. Electron paramagnetic resonance (EPR) spectra were recorded for all these sets of reaction mixtures in the dark as well as with UV-A (340 nm, 15 min exposure at 0.85 mW/cm2) and blue light illumination (450 ± 10 nm, 15 min exposure at 0.3 mW/cm2) separately. All the EPR spectra were recorded at room temperature under the following settings: scan range, 100 G; field set, 3237 G; time constant, 1 s; scan time, 8 min; modulation amplitude, 2 G; modulation frequency, 100 kHz; receiver gain, 2.5 × 104× 10.
Energy transfer experiments
Riboflavin, gentisic acid, ferulic acid and 3,4-dihydroxy benzoic acid were prepared in 50% ethanol and were used to assess emission spectra at various concentrations. For fluorescence resonance energy transfer (FRET) experiments, the intensity of emission of riboflavin (1 μM, 0.1 μM and 0.01 μM final concentrations) at λex = 340 nm and λem = 518 nm was recorded as the baseline riboflavin emission (non-FRET emission). Gentisic acid (1 μM, 0.1 μM and 0.01 μM final concentrations) was introduced into the reaction and FRET was monitored at λex = 340 nm and λem = 518 nm (FRET emission). FRET emission of riboflavin was normalized with non-FRET emission, and the FRET efficiency was fit into a percent scale (a value of 100 meant that energy was not transferred in the system). Viscosity of the medium was altered by introducing 10% glycerol (final concentration). The experiments were repeated five times and the observations were tabulated.
Statistical analysis
All the experiments were repeated a minimum of five times, and the spectra presented represent the observations obtained in the experiments. For FRET experiments, the non-FRET and FRET fluorescence from riboflavin was acquired in five independent experiments in the absence and presence of 10% glycerol. The FRET efficiency was computed for each experiment and the mean ± SEM of five independent observations was calculated. Analysis of variance (ANOVA) between FRET efficiencies at varying viscosities was carried out using one-way ANOVA (Sigma Plot Packages, USA).
RESULTS
Absorption and emission properties of coleoptile tips
Extracts of wheat coleoptile tips showed very heavy absorbance in the UV-A (320–400 nm) region and moderate absorbance in the blue (450 nm) region (Fig. 1). These spectra showed overt similarities with the absorption spectra of riboflavin, carotene and pterine in the UV-A region. On the other hand, although carotene and pterine had relatively weak absorption properties in the blue region, riboflavin showed strong absorbance in blue regions of the spectrum (Fig. 1). Further, the blue light absorption pattern of riboflavin matched the absorption properties of cloeoptile tip extracts in the 400–500 nm region of the visible spectrum, which was not obvious with carotene and pterine (Fig. 1).
Absorption spectrum of wheat coleoptile tip extract superimposed over the absorption spectra of riboflavin β-carotene and pterin. Coleoptile tip extract showed strong absorbance in the ranges 300–400 and 400–500 nm, with a λmax = 450 nm in the blue region of the spectrum, which corresponded with the absorption spectrum of riboflavin (marked with a vertical arrow). Carotene and pterin had weaker absorption properties in the blue region of the absorption spectrum.
Absorption properties of polyphenolic compounds
Because the coleoptile extracts absorbed strongly in the range of 300–400 nm, we screened a number of polyphenolic compounds for their ability to absorb light in this region. The results are presented in Fig. 2. The compounds in Group 1 showed strong absorbance in the 300–400 nm region (gentisic acid, p-coumaric acid, caffeic acid, ferulic acid, quercetin, aesculetin, scopoletin, coumarin, morin and chlorogenic acid) whereas other phenols tested (Group 2, including 3,4-dihydroxy benzoic acid, trans-cinnamic acid, syringic acid, vanillic acid, m-coumaric acid, salicyl hydroxymic acid, hydrocinnamic acid, hydrocaffeic acid, umbelli-ferone and rutin) showed relatively weak absorption in this region.
Absorption spectra of a select set of polyphenolic compounds. All the polyphenolic compounds employed in this study absorbed strongly in the UV-B or UV-C region (or both), but a few showed an additional absorption band in the UV-A region, with λmax falling close to 340 nm, which are categorized in Group 1. The members of Group 1 include gentisic acid, p-coumaric acid, caffeic acid, ferulic acid, quercetin, aesculetin, scopoletin, coumarin, morin and chlorogenic acid. The remaining members comprise Group 2.
Emission properties of coleoptile tips
Evaluation of the autofluorescence of the intact wheat coleoptile tip (Fig. 3A) and a longitudinal section of the coleoptile tip (Fig. 3B,C) by fluorescence microscopy demonstrated that the blue light absorbing molecules were located at the growing tip of the coleoptile as emitting heavy green localized fluorescence. The emission spectrum of the coleoptile tip extract in ethanol showed a strong emission peak at 518 nm, which corresponded well with the emission properties of riboflavin (Fig. 3D). However, the emission of carotene was relatively weak at 518 nm, but that of pterine was substantial (Fig. 3D).
Autofluorescence images of wheat coleoptiles under 450 nm excitation optics. Whole coleoptile with 2 mm from extreme tip excised and imaged separately from the rest of the coleoptile (A), a longitudinal section of wheat coleoptile imaged under combination optics (phase contrast and epifluorescence) (B), and an image of B under epifluorescence optics (C). Blue light absorbing entities were located at a distance of approximately 0.5 mm from the extreme apex of the coleoptile.
Emission properties of polyphenolic compounds
Coleoptile tip extracts excited with 340 nm or 390 nm produced an emission maximum around 430–450 nm. When we matched these emission characteristics with the emission characteristics of the test compounds, we observed a large number of phenolic compounds and we observed that the Group 1 compounds (which showed strong absorption of 340 nm light) also showed strong emission at 450 nm (when excited with either 340 or 390 nm), whereas the Group 2 compounds did not show this capability (Fig. 4).
Emission spectra of wheat coleoptile tip, a representative of Group 1 polyphenol (gentisic acid) and a representative of Group 2 polyphenol (3,4-dihydroxy benzoic acid). Both coleoptile tip extract and gentisic acid produced emission maxima at λ= 450 nm upon excitation with either 340 nm or 390 nm light. 3,4-Dihydroxy benzoic acid failed to produce emission at λ= 450 nm under this experimental setup. Other members of the respective groups also showed similar phenomena (spectra not presented).
Energy transfer experiments
Because the emission peaks of the Group 1 polyphenolic compounds showed good overlap with the excitation peak of riboflavin (Fig. 5), it was of interest to examine whether fluorescence energy transfer could take place between members of Group 1 polyphenolic compounds and riboflavin. We chose all members of Group 1 compounds, and 3,4-dihydroxybenzoic acid as a representative of Group 2 compounds for performing fluorescence energy transfer between polyphenolic compounds and riboflavin under in vitro conditions. Figure 6 summarizes our FRET observations between the selected polyphenols and riboflavin when excited at the 340 nm wavelength of light.
Excitation spectrum of riboflavin showing excitation maxima at λ= 340 nm and λ= 450 nm. Emission spectrum of gentisic acid with emission maximum at λ= 450 nm (λex = 340 nm) poses a possibility of fluorescence resonance energy transfer between gentisic acid and riboflavin (shown by a vertical arrow). Emission spectrum of riboflavin with λex = 450 nm also is shown.
Emission spectra showing energy transfer between gentisic acid and riboflavin (left), ferrulic acid and riboflavin (middle) and 3,4-dihydroxy benzoic acid and riboflavin (right). FRET was visible in the gentisic acid + riboflavin and ferulic acid + riboflavin reaction setups, but was not visible in the 3,4-dihydroxy benzoic acid + riboflavin reaction.
Of all the compounds tested, gentisic acid and ferulic acid present an interesting picture when excited with 340 nm in the presence of riboflavin. Excitation of gentisic acid with 340 nm light produced a strong emission at 440 nm (Fig. 6). Whereas excitation of riboflavin with 340 nm light can produce emission at 518 nm, the presence of gentisic acid in the system caused an increase in the emission at 518 nm, indicating possible energy transfer between gentisic acid and riboflavin (Fig. 6). Similarly, excitation of ferulic acid with 340 nm light produced a fluorescence band at 420 nm, which caused enhancement of fluorescence from riboflavin at 518 nm. Thus, gentisic acid and ferulic acid appear to bring about an excitation energy transfer complex with riboflavin. Neither the other members of the Group 1 polyphenols nor the Group 2 polyphenols could exhibit energy transfer efficiency with riboflavin (Fig. 6). Further, FRET was observable until a critical minimum concentration of riboflavin (0.1 μM) and gentisic acid (0.01 μM) were achieved. FRET from gentisic acid to riboflavin was viscosity dependent as expected. Thus, in the presence of 10% glycerol in the medium, FRET from gentisic acid (0.1 μM and below) to riboflavin (0.1 μM and below) was totally abolished (Table 1). An extrapolation of this result suggests that some of the polyphenolic compounds in the coleoptile tips can function as an important component of a photoreceptor system with riboflavin as a reaction center involved in light signaling.
| FRET efficiency | |||
|---|---|---|---|
| Riboflavin (μM) | Gentisic acid (μM) | 0% Glycerol | 10% Glycerol |
| 1 | 1 | 109.47 ± 8.23 | 112.09 ± 5.2 |
| 0.1 | 1 | 333.89 ± 4.11 | 546.29 ± 8.22 |
| 0.1 | 0.1 | 129.46 ± 0.13 | 107.33 ± 1.72* |
| 0.1 | 0.01 | 148.79 ± 3.08 | 74.92 ± 0.79* |
- * P < 0.001 when compared with the corresponding observation in 0% glycerol medium.
Generation of O2•− by blue light excitation of various phenols
Generation of O2•− has been shown to initialize blue light signal transduction in wheat coleoptile tips (13). Because phenols absorb UV-B/near-blue wavelengths of light it was of interest to examine whether these would also generate O2•− anion radical on excitation with UV-B/blue light as do flavin, carotene and pterine.
Most of these phenolic compounds generated superoxide anion when excited with 366 nm and 450 nm wavelengths, the wavelength maxima for phototropic curvature 366 nm being the more effective compound than 450 nm. Various phenolics, benzoic acid derivatives, cinnamic acids, coumarins and flavonoids differed in their efficiency to generate O2•− radical on being excited with the two wavelengths of light (Fig. 7).
Superoxide radical production by various polyphenolic compounds in dark (left column), after illumination with λ= 340 nm (middle column) and after illumination with λ= 450 nm (right column). None of the polyphenolic compounds produced superoxide radical in the dark, but all produced superoxide radicals at varying levels upon excitation with λ= 340 nm. On excitation with λ= 450 nm, some of the polyphenolic compounds produced superoxide anion radical, although the yield was lower when compared with λ= 340 nm excitation. Polyphenolic compounds that failed to produce detectable levels of superoxide radical at λ= 450 nm excitation include 3,4-dihydroxy benzoic acid, ferulic acid, salicyl hydroxymic acid, umbelliferone, rutin, morin and chlorogenic acid.
DISCUSSION
Attempts to elucidate the nature of the photoreceptor pigment in plants by evaluating the similarities between action spectra of the photoresponse and absorption spectra of photoreceptor candidate molecules have shown that in part, this action spectra corresponds to flavin and in part to carotene (14–16). Thus the action spectrum optimum at 340 nm is believed to be characteristic of flavin, whereas the shoulders around 430 nm and 470 nm with a peak of 450 nm are characteristic of carotene (these shoulders are lacking in flavin, showing a single maxima at 450 nm); however a critical examination of the action spectra and the absorption spectra of carotene and flavin would show that the overall pattern of the action spectra is entirely different than the single absorption spectrum of any photoreceptor pigment (17).
The ratios of the maxima obtained at 340 nm and 450 nm are remarkably higher in the action spectrum compared with the absorption spectrum of flavin (1:1.25). Such an action spectrum could be imagined if the two photoreceptors worked in tandem or energy from some other blue light absorbing photoreceptor compounds was transferred to the active center pigment for phototropism.
The identification and characterization of CRY and Phot genes have strengthened the case of flavin as a photoreceptor pigment in the coleoptile tip (6–9). Spectroscopic studies on wheat coleoptile tip homogenates reported in this paper also demonstrated that apart from the blue absorbing (absorption maxima at 450 nm and fluorescence emission around 520 nm, characteristic of flavin and carotene) pigments, there was a preponderance of UV-A absorbing compounds in coleoptile tips (Fig. 1). Further, the emission spectra of the coleoptile tip homogenates obtained with excitation wave-lengths 340 nm and 390 nm clearly indicated that these UV-A absorbing material(s) could emit blue fluorescence (Fig. 4). Analysis of the absorption and emission properties of a select set of polyphenols indicated that a subset of them could possibly account for both the properties displayed by the coleoptile homogenates. It is interesting to note that the emission spectra of many of these phenolic compounds shared good overlap with the excitation spectrum of riboflavin (Fig. 5), making a case for energy transfer between two groups of molecules. Further examination of this possibility revealed that gentisic acid and ferulic acid could act as donors in a FRET pair with riboflavin as the acceptor, whereas the other molecules failed to show energy transfer (Fig. 6). Thus, energy transfer from excited phenolic compound to flavin is indicated and may be involved in the phototropic response by UV-A wavelengths of light.
We also observed that these phenolic compounds are themselves capable of generating superoxide radical by excitation with UV-A light and can possibly elicit signal transduction cascades on their own. This would explain the observation by Shinkle et al. (18), who have recently shown that shorter wavelengths of UV-B and UV-A can induce phototropic curvature and inhibition of hypocotyl elongation irrespective of blue light in cucumber and Arabidopsis seedlings. This also explains the phototropic response of siliquae of oil seed rape showing an action maximum of carotene at 319 and 324 nm, the phototropic response of nph-1 mutant defective in normal blue light phototropism of Arabidopsis (18). Evidence has been provided that distinct blue and UV-A pathways interact synergistically with UV-B pathways to stimulate chalcone synthase (CHS) expression in Arabidopsis. Thus energy transfer between various photoreceptors absorbing UV-B, UV-A and blue light could proceed under normal conditions and in mutants lacking or defective in normal blue light. In any of these photoreceptor systems the function could be taken over by the other short wave system.
It is well known that light induces activation of phenylalanine ammonialyase (PAL) in plants followed by the activation of CHS (10,19,20). PAL acting on phenylalanine synthesize as trans-cinnamic acid, which is readily converted into p-coumaric acid, thus opening the pathway for the synthesis of cinnamic-coumaric acid derivatives present in different plants. The induction of CHS activates the pathway of flavonoid synthesis in all plants, leading to the synthesis of a variety of flavonoid compounds in different plants, absorbing the blue wavelength of light.
For these compounds to participate in the signaling process and bring about energy transfer from one to the other, these would have to be present near enough to each other at the receptor site to facilitate resonance energy transfer at the site of photoinduction. Flavin and pterin binding sites have been demonstrated in CRY 1, CRY 2 and Phot 1 and Phot 2 but phenolic binding sites in the protein remains to be demonstrated. This may not be unlikely because a recent demonstration by Jiang et al. showed that in purple bacterium, Rhodospirillum, phytochrome has an additional N-terminal sequence that bears a covalently attached p-hydroxy cinnamic acid moiety that absorbs blue light instead of red (21). Further, Halorhodospira halophila also has a photoactive yellow protein with p-coumaric acid linked to Cys 69 via a thiol ester bond (22). The signaling is initiated by trans-cis isomerization of p-coumaric acid, which brings about conformational changes in the protein moiety responsible for signaling (23,24). Thus, some of the polyphenols that can act as donors in a FRET pair with riboflavin as the acceptor in vitro invoke the idea of polyphenols as energy transducers in photoreception by plants.
Acknowledgements
This work was supported by the Council of Scientific and Industrial Research, New Delhi. Part of the work was performed at the School of Life Sciences, Devi Ahilya University, Indore, India.
