Study on the Synthesis, Antioxidant Properties, and Self-Assembly of Carotenoid–Flavonoid Conjugates

Flavonoids and carotenoids possess beneficial physiological effects, such as high antioxidant capacity, anticarcinogenic, immunomodulatory, and anti-inflammatory properties, as well as protective effects against UV light. The covalent coupling of hydrophobic carotenoids with hydrophilic flavonoids, such as daidzein and chrysin, was achieved, resulting in new amphipathic structures. 7-Azidohexyl ethers of daidzein and chrysin were prepared in five steps, and their azide-alkyne [4 + 2] cycloaddition with pentynoates of 8′-apo-β-carotenol, zeaxanthin, and capsanthin afforded carotenoid–flavonoid conjugates. The trolox-equivalent antioxidant capacity against ABTS•+ radical cation and self-assembly of the final products were examined. The 1:1 flavonoid–carotenoid hybrids generally showed higher antioxidant activity than their parent flavonoids but lower than that of the corresponding carotenoids. The diflavonoid hybrids of zeaxanthin and capsanthin, however, were found to exhibit a synergistic enhancement in antioxidant capacities. ECD (electronic circular dichroism) and UV-vis analysis of zeaxanthin–flavonoid conjugates revealed that they form different optically active J-aggregates in acetone/water and tetrahydrofuran/water mixtures depending on the solvent ratio and type of the applied aprotic polar solvent, while the capsanthin derivatives showed no self-assembly. The zeaxanthin bis-triazole conjugates with daidzein and with chrysin, differing only in the position of a phenolic hydroxyl group, showed significantly different aggregation profile upon the addition of water.


Introduction
Countless natural bioactive compounds have been isolated from food products and their effects on human health were studied. However, an increasing amount of evidence suggests that a nutrient may have a specific action, the combination of nutrients displays different (even stronger) effects than a sole component [1,2]. As foods are complex matrices, the bioaccessibility and bioavailability of a component are obviously influenced by the presence or absence of other constituents [3,4]. At the same time, significant synergy was found in the physiological effects of many small molecules used in combination [5].
Both carotenoids and flavonoids are sensitive compounds, and for the coupling of these molecules, a very mild method was needed. Previously, azide-alkyne [4 + 2] cycloaddition (clickreaction) was successfully applied in the synthesis of carotenoid conjugates with polyethylene glycols [27] or with carbohydrates [28]. Since the pentynoate esters of carotenoids can be easily prepared [28], we worked out a procedure for the synthesis of azido-derivatives of flavonoids.
The isoflavone daidzein (1) and the flavone chrysin (7) were the molecules of choice for the preparation of 7-azidohexyl ether derivatives in five steps (Figure 1). The flavonoids (1,7) were acetylated by acetic anhydride in pyridine, and their position 7 were selectively deprotected using imidazole [29]. A bromohexyl moiety was introduced in position 7 by the Mitsonobu reaction, and after complete deacetylation, an azide substitution was applied. The 7-azidohexyl ether of daidzein and chrysin (6,12) was reacted with various carotenoid pentynoates (8'-apo-β-carotenol-pentynoate 13, zeaxanthin dipentynoate 14, capsanthin dipentynoate 15) in the presence of bis-triphenylphosphano-copper(I)-butyrate complex [30] as a catalyst, in dichloromethane. (Figure 2). The azido-flavonoids were applied in small excess (1.3-1.5 eq. per triple bond), providing the corresponding triazoles in satisfactory or good yields (60%-80%). In the case of the 8′-apo-β-carotenol-pentynoate 13, only monotriazoles can be formed. The other carotenoids were used as dipentynoate esters, and during the click-reaction the formation of monotriazoles was also observed. When the reactions were followed by TLC, the monotriazoles appeared first, and in a longer reaction time, they were further converted to bistriazole derivatives. However, small amounts of monotriazoles always remained at the end of the reactions, even with a higher excess of azidohexyl ethers. Yields for the monotriazoles were calculated from the starting The 7-azidohexyl ether of daidzein and chrysin (6,12) was reacted with various carotenoid pentynoates (8 -apo-β-carotenol-pentynoate 13, zeaxanthin dipentynoate 14, capsanthin dipentynoate 15) in the presence of bis-triphenylphosphano-copper(I)-butyrate complex [30] as a catalyst, in dichloromethane. (Figure 2). The azido-flavonoids were applied in small excess (1.3-1.5 eq. per triple bond), providing the corresponding triazoles in satisfactory or good yields (60%-80%). In the case of the 8 -apo-β-carotenol-pentynoate 13, only monotriazoles can be formed. The other carotenoids were used as dipentynoate esters, and during the click-reaction the formation of monotriazoles was also observed. When the reactions were followed by TLC, the monotriazoles appeared first, and in a longer reaction time, they were further converted to bistriazole derivatives. However, small amounts of monotriazoles always remained at the end of the reactions, even with a higher excess of azidohexyl ethers. Yields for the monotriazoles were calculated from the starting carotenoid pentynoates, as if they were the sole products. The capsanthin pentynoate (15) did not give a complete conversion, as approximately 10% of the starting pentynoate was recovered.
Molecules 2020, 24, x FOR PEER REVIEW 4 of 22 carotenoid pentynoates, as if they were the sole products. The capsanthin pentynoate (15) did not give a complete conversion, as approximately 10% of the starting pentynoate was recovered. All the prepared compounds were fully characterized by their NMR, MS, UV, and IR spectra. As capsanthin is a non-symmetric carotenoid, it can provide two regioisomeric monotriazole derivatives. In the case of the daidzein-capsanthin conjugates (20a and 20b), these regioisomers could be separated and identified. In the 1 H-NMR spectra of 20a and 20b, the chemical shifts of H-3' are slightly but characteristically different (Table 1), and in comparison with the parent capsanthin dipentynoate (15), as well as with the bistriazole 19, the substituents on C-3' can be distinguished. Similar differences in the chemical shifts of H-3 can also be observed. All the prepared compounds were fully characterized by their nmR, MS, UV, and IR spectra. As capsanthin is a non-symmetric carotenoid, it can provide two regioisomeric monotriazole derivatives. In the case of the daidzein-capsanthin conjugates (20a and 20b), these regioisomers could be separated and identified. In the 1 H-NMR spectra of 20a and 20b, the chemical shifts of H-3 are slightly but characteristically different (Table 1), and in comparison with the parent capsanthin dipentynoate (15), as well as with the bistriazole 19, the substituents on C-3 can be distinguished. Similar differences in the chemical shifts of H-3 can also be observed.
The two regioisomeric monotriazoles of 25 were also observed on TLC, but since they were formed in low yields and their separation seemed to be extremely challenging, we used them as a mixture for further investigations.

Measurement of Antioxidant Capacity
The trolox-equivalent antioxidant capacity (TEAC) of the prepared flavonoid-carotenoid conjugates (16)(17)(18)(19)(20)(21)(22)(23)(24)(25) was investigated towards ABTS •+ radical cation in aqueous ethanolic solution (see the experimental section) [31]. The synthetized compounds showed significantly higher TEAC values than their parent flavonoids ( Figure 3, Table 2); the exceptions were the conjugates of 8′-apo-β-  Zeaxanthin conjugated with one daidzein (18) or chrysin (23) was found to be weaker antioxidants than zeaxanthin; however, its bistriazole derivatives (17 and 22) exceeded the parent zeaxanthin. The antioxidant capacity of capsanthin did not significantly change on conjugation with one daidzein, and no differences in the TEAC values of monotriazoles 20a and 20b were found. The bistriazole derivative 19, however, surpassed capsanthin in antioxidant capacity. The conjugation with chrysin lowered the TEAC value in small extents. The 8 -apo-β-carotenol was found to be a better antioxidant than its flavonoid derivatives.
The daidzein-bistriazole derivatives of zeaxanthin and capsanthin were the best antioxidants under the examined conditions. The TEAC value of a simple zeaxanthin:daidzein 1:2 mixture is much lower than that of the bistriazole conjugate 17, which indicates a synergistic enhancement in antioxidant capacity.
The above results were obtained in ethanolic solution of ABTS •+ (see the experimental section). The antioxidant property of compound 17 was also determined in aqueous ABTS •+ solution, where the TEAC value was found to drop by ca. 20%. The reason for this phenomenon can be explained by the water-induced aggregation of the carotenoid derivative, which is an accordance with literature data [32].

Supramolecular Assembly
When the coupling click-reactions were followed by TLC, and samples were taken directly from the dichloromethane reaction mixtures, the formation of an intense orange-red spot on the start-line was observed in the hexane-acetone eluent while the spots of the products remained faint. This phenomenon generally suggests a salt formation; however, it was hardly possible in these reactions. After the disappearance of the starting materials, the dichloromethane solvent was evaporated from the reaction mixture of 8 -apo-β-carotenol-pentynoate and 7-azidohexyl daidzein, then the reaction mixture was taken up in acetone. Interestingly, when the acetonic solution was applied on the TLC plate and eluted by the same eluent (hexane-acetone), no start-point substance remained, but the whole amount was eluted with the R f value of the product. This finding was rather unusual, as the retention properties of a substance depend on the quality/polarity of the eluent but independent of the solvent in which it is applied on the plate, since the plate is dried before elution. We repeated the TLC experiment using different solvents for the application of the sample, but the same hexane-acetone eluent for the development. The product triazole 16 in methanolic, chloroformic, or ethyl acetate solutions was found to behave similarly as in dichloromethane, i.e., it gave a start-spot, while it was eluted when it was applied in acetonic solution. From these observations, we suspected the formation of some aggregates, which formed in the presence of certain solvents on drying and did not moved on the chromatographic plate.
Carotenoids are known to form self-assemblies among certain conditions even in nature. These supramolecular organizations frequently have ordered structures (H-or J-type aggregates), which results in significant changes in the UV-Vis spectra of carotenoids [33][34][35]. Moreover, the aggregates can exhibit a so-called supramolecular exciton chirality due to the optically active helical structures, which can be studied by electronic circular dichroism (ECD) spectroscopy [36,37]. Aggregation of amphipathic carotenoid derivatives was also examined thoroughly [38].
As 16 daidzein-8 -apo-β-carotenol hybrid is not chiral, its aggregation can only be examined by the change in its UV-Vis properties. The absorption spectrum of the ethanolic solution of 16 upon aqueous dilution showed an intense hypsochromic and a small bathochromic shift that suggest self-assembly of 16 with the formation of J-aggregate ( Figure S1).
The aggregation properties of the optically active bistriazoles 17, 19, 22, and 24, as well as monotriazole 23 were investigated by ECD spectroscopy. The acetone or tetrahydrofuran (THF) solution of these derivatives were diluted with water, and the ECD spectra were recorded with the different dilutions. Upon dilution with water, the optically active capsanthin derivatives (19,24) did not show any significant increase in the ECD transitions that would have suggested supramolecular organization. In contrast, intense ECD bands appeared in the spectra of the bistriazole daidzein-zeaxanthin conjugate 17 in acetone when the water content was increased above 25% (Figure 4a). the change in its UV-Vis properties. The absorption spectrum of the ethanolic solution of 16 upon aqueous dilution showed an intense hypsochromic and a small bathochromic shift that suggest selfassembly of 16 with the formation of J-aggregate ( Figure S1).
The aggregation properties of the optically active bistriazoles 17, 19, 22, and 24, as well as monotriazole 23 were investigated by ECD spectroscopy. The acetone or tetrahydrofuran (THF) solution of these derivatives were diluted with water, and the ECD spectra were recorded with the different dilutions. Upon dilution with water, the optically active capsanthin derivatives (19, 24) did not show any significant increase in the ECD transitions that would have suggested supramolecular organization. In contrast, intense ECD bands appeared in the spectra of the bistriazole daidzeinzeaxanthin conjugate 17 in acetone when the water content was increased above 25% (Figure 4a).  The absorption spectra showed a characteristic bathochromic shift with a new band at 519 nm, which indicated the formation of a weakly-coupled J-aggregate ( Figure S2). Zeaxanthin and zeaxanthin derivatives modified at the sec-hydroxyl group can form both J-and H-aggregates depending on the applied pH, initial concentration, and solvent/water ratio [39,40]. The intense 480 nm positive cotton effect (CE) with shoulders at 456 and 519 nm, the broad and intense negative CE at 412 nm, and the weak negative CE at 547 nm are characteristic of a J-aggregate also observed in the EtOH/water solution of zeaxanthin under certain conditions [39,40]. Interestingly, a similar ECD profile was observed in sergeant-soldier aggregate of 5% αand 95% β-carotene [37], in which weak van der Waals interactions also prevail. The ECD spectra recorded in the range of 1:0-1:3 acetone/water ratio showed strong characteristic ECD bands from a 2:1 ratio and the largest amplitudes were observed at a 1:2 ratio. A further increase of the water content to a 1:3 ratio decreased both the ECD and absorption intensity. The intensities of the five CEs did not show a monotonous change with the increase of the water content; the 547 nm CE had maximum intensity at the 1:1 ratio, while the 480 and 412 CEs had the largest amplitude at the 1:2 ratio.
In THF, a similar ECD feature appeared with the increase of the water ratio, and the maximum ECD intensity was reached at the 1:5 ratio, while a further increase of the water content decreased the amplitude of the CEs (Figure 4b). The absorption and ECD spectra of 17 at the 2:3 and 1:2 ratio differed from those at the 1:3, 1:5, and 1:10 ratios, which indicated the presence of at least two different J-aggregates. The absorption spectra showed a characteristic bathochromic shift with a new band at 519 nm, which indicated the formation of a weakly-coupled J-aggregate ( Figure S2). Zeaxanthin and zeaxanthin derivatives modified at the sec-hydroxyl group can form both J-and H-aggregates depending on the applied pH, initial concentration, and solvent/water ratio [39,40]. The intense 480 nm positive cotton effect (CE) with shoulders at 456 and 519 nm, the broad and intense negative CE at 412 nm, and the weak negative CE at 547 nm are characteristic of a J-aggregate also observed in the EtOH/water solution of zeaxanthin under certain conditions [39,40]. Interestingly, a similar ECD profile was observed in sergeant-soldier aggregate of 5% αand 95% β-carotene [37], in which weak van der Waals interactions also prevail. The ECD spectra recorded in the range of 1:0-1:3 acetone/water ratio showed strong characteristic ECD bands from a 2:1 ratio and the largest amplitudes were observed at a 1:2 ratio. A further increase of the water content to a 1:3 ratio decreased both the ECD and absorption intensity. The intensities of the five CEs did not show a monotonous change with the increase of the water content; the 547 nm CE had maximum intensity at the 1:1 ratio, while the 480 and 412 CEs had the largest amplitude at the 1:2 ratio.
In THF, a similar ECD feature appeared with the increase of the water ratio, and the maximum ECD intensity was reached at the 1:5 ratio, while a further increase of the water content decreased the amplitude of the CEs (Figure 4b). The absorption and ECD spectra of 17 at the 2:3 and 1:2 ratio differed from those at the 1:3, 1:5, and 1:10 ratios, which indicated the presence of at least two different J-aggregates.
The bistriazole chrysin-zeaxanthin conjugate (22) also formed a J-aggregate with the increasing amount of water in acetone, which was, however, different from that of 17. The conjugate 22 had a bisignate ECD curve even in pure acetone, indicating aggregate formation in the absence of water (Figure 5a,b). In acetone/water 5:1, the high-wavelength positive CE shifted to the red with increased intensity. The 545 nm positive CE with shoulders at 508 and 464 nm and the negative CE at 397 nm represented an ECD pattern similar to that of 17, with a considerable red shift. In acetone/water 4:1, the intensity increased significantly and reached the maximum, while further addition of water flattened the ECD curve and changed the absorption maxima. The diminishing ECD intensity in acetone/water ratio 2:1-1:3 suggested the deterioration of the J-aggregate.
In the THF solution, 22 had a low-intensity ECD spectrum (Figure 5c,d), different from that in acetone, which did not change much up to the 2:3 THF/water ratio. At the 1:2 THF/water ratio, an intense negative ECD couplet centered at 425 nm appeared, which was red shifted with the 1:3 ratio. A further significant change occurred at the 1:5 and 1:10 ratios, the ECD curves of which differed only in the amplitudes and they had a similar ECD pattern to that of 17, with opposite signs of CEs. The ECD spectra of 22 at the high water ratio were very different from those in acetone, and the mirror-image relationship with the ECD spectrum of 17 acetone/water 1:2 indicated their opposite sense of supramolecular chirality. The three different ECD and absorption spectra suggested the presence of at least three different types of J-aggregate.
intensity. The 545 nm positive CE with shoulders at 508 and 464 nm and the negative CE at 397 nm represented an ECD pattern similar to that of 17, with a considerable red shift. In acetone/water 4:1, the intensity increased significantly and reached the maximum, while further addition of water flattened the ECD curve and changed the absorption maxima. The diminishing ECD intensity in acetone/water ratio 2:1-1:3 suggested the deterioration of the J-aggregate.
In the THF solution, 22 had a low-intensity ECD spectrum (Figure 5c,d), different from that in acetone, which did not change much up to the 2:3 THF/water ratio. At the 1:2 THF/water ratio, an intense negative ECD couplet centered at 425 nm appeared, which was red shifted with the 1:3 ratio. A further significant change occurred at the 1:5 and 1:10 ratios, the ECD curves of which differed only in the amplitudes and they had a similar ECD pattern to that of 17, with opposite signs of CEs. The ECD spectra of 22 at the high water ratio were very different from those in acetone, and the mirrorimage relationship with the ECD spectrum of 17 acetone/water 1:2 indicated their opposite sense of supramolecular chirality. The three different ECD and absorption spectra suggested the presence of at least three different types of J-aggregate. The zeaxanthin conjugates 17 and 22 differ only in the attached daidzein/chrysin flavonoid moiety, but they produced markedly different aggregation profiles upon the addition of water in both acetone and THF. The aggregate formation was also sensitive to the type of non-protic solvent, acetone versus THF. The zeaxanthin conjugates 17 and 22 differ only in the attached daidzein/chrysin flavonoid moiety, but they produced markedly different aggregation profiles upon the addition of water in both acetone and THF. The aggregate formation was also sensitive to the type of non-protic solvent, acetone versus THF.
In the acetone/water mixtures, the monotriazole chrysin-zeaxanthin conjugate (23) showed the same characteristic changes in both the ECD and absorption spectra as the bistriazole daidzein-zeaxanthin conjugate 17, indicating the formation of the same J-aggregate (Figure 6a,b).
The ECD transitions above 400 nm reached the maximum intensities at the 1:1 ratio and there was no significant decrease in the intensity up to the 1:3 ratio.
In the THF/water mixtures, a completely different aggregation process could be observed (Figure 7a,b). Aggregation started at the 1:1 ratio and the negative ECD couplet reached the maximum intensity at the 1:2 ratio. The intensity of the ECD bands was so high at the initial concentration that a sixfold dilution had to be applied to record the ECD bands properly. According to the characteristic 530 nm absorption band, a J-aggregate formed at the 2:3 and 1:2 ratios.
By increasing the ratio of water further, the negative couplet was inverted to a positive one at the 1:5 and 1:10 ratios. At these ratios, both hypsochromic and bathochromic shifts of the absorption bands took place compared to the bands at the 1:1 ratio (Figure 8). The inversion of the ECD couplet with the increasing amount of water in THF suggests the inversion of chirality in the aggregate, which could not be observed in the acetone/water mixtures. In the acetone/water mixtures, the monotriazole chrysin-zeaxanthin conjugate (23) showed the same characteristic changes in both the ECD and absorption spectra as the bistriazole daidzeinzeaxanthin conjugate 17, indicating the formation of the same J-aggregate (Figure 6a,b).
The ECD transitions above 400 nm reached the maximum intensities at the 1:1 ratio and there was no significant decrease in the intensity up to the 1:3 ratio.
In the THF/water mixtures, a completely different aggregation process could be observed (Figure 7a,b). Aggregation started at the 1:1 ratio and the negative ECD couplet reached the maximum intensity at the 1:2 ratio. The intensity of the ECD bands was so high at the initial concentration that a sixfold dilution had to be applied to record the ECD bands properly. According to the characteristic 530 nm absorption band, a J-aggregate formed at the 2:3 and 1:2 ratios. By increasing the ratio of water further, the negative couplet was inverted to a positive one at the 1:5 and 1:10 ratios. At these ratios, both hypsochromic and bathochromic shifts of the absorption bands took place compared to the bands at the 1:1 ratio (Figure 8). The inversion of the ECD couplet with the increasing amount of water in THF suggests the inversion of chirality in the aggregate, which could not be observed in the acetone/water mixtures.  The ECD transitions above 400 nm reached the maximum intensities at the 1:1 ratio and there was no significant decrease in the intensity up to the 1:3 ratio.
In the THF/water mixtures, a completely different aggregation process could be observed (Figure 7a,b). Aggregation started at the 1:1 ratio and the negative ECD couplet reached the maximum intensity at the 1:2 ratio. The intensity of the ECD bands was so high at the initial concentration that a sixfold dilution had to be applied to record the ECD bands properly. According to the characteristic 530 nm absorption band, a J-aggregate formed at the 2:3 and 1:2 ratios. By increasing the ratio of water further, the negative couplet was inverted to a positive one at the 1:5 and 1:10 ratios. At these ratios, both hypsochromic and bathochromic shifts of the absorption bands took place compared to the bands at the 1:1 ratio ( Figure 8). The inversion of the ECD couplet with the increasing amount of water in THF suggests the inversion of chirality in the aggregate, which could not be observed in the acetone/water mixtures.

Conclusions
A straightforward method was worked out for the synthesis of novel carotenoid-flavonoid conjugates and their antioxidant capacities were measured. Some of these amphipathic compounds containing a flexible linker were found to show self-aggregation in the presence of water. The antioxidant capacity was measured in ethanolic solutions. Determination of the TEAC values in

Conclusions
A straightforward method was worked out for the synthesis of novel carotenoid-flavonoid conjugates and their antioxidant capacities were measured. Some of these amphipathic compounds containing a flexible linker were found to show self-aggregation in the presence of water. The antioxidant capacity was measured in ethanolic solutions. Determination of the TEAC values in aqueous medium seems to be more relevant biologically, but most probably, the antioxidant property under such conditions rather belongs to the aggregates than to the individual molecules. A further study is needed to clarify how the aggregation influences the antioxidant capacity. In the future, the solvent dependence of the antioxidant activity, and the structure of the aggregates will be examined with atomic force microscopy.

Materials and Methods
Melting points were measured on a Stuart SMP30 apparatus (Cole-Parmer, Staffordshire, UK) and are uncorrected. The UV spectra were implemented on a Jasco spectrophotometer model V-550 UV/Vis (JASCO, Tokyo, Japan). ECD spectra were recorded at room temperature with a J-810 spectropolarimeter (JASCO, Tokyo, Japan). The IR spectra were run on an Impact 400 FT-IR spectrophotometer (Nicolet, Germany) in KBr pellets using a KBr pellet as the background reference spectrum. nmR spectra were recorded with a Bruker Avance III Ascend 500 spectrometer (500/125 MHz for 1 H/ 13 C) (Bruker BioSpin GmbH, Rheinstetten, Germany). The 13 C and 1 H-NMR assignments for 6, 12, 16, 17, 19, and 23 were made on the basis of 1D ( 1 H, 13 C APT) and 2D (HSQC, HMBC) experiments, and the assignments for the other compounds were based on structural similarities with the above molecules. Chemical shifts were referenced to the residual solvent signals, or to Me 4 Si ( 1 H). A special numbering of H and C atoms, indicated in the formulas (Figure 9), for the flavonoid moieties was applied, since the structure of the final coupled products was rather complex. experiments, and the assignments for the other compounds were based on structural similarities with the above molecules. Chemical shifts were referenced to the residual solvent signals, or to Me4Si ( 1 H). A special numbering of H and C atoms, indicated in the formulas (Figure 9.), for the flavonoid moieties was applied, since the structure of the final coupled products was rather complex. Molar masses were obtained by an Autoflex II MALDI instrument (Bruker Daltonics, Bremen, Germany). 2,5-Dihydroxy-benzoic acid (DHB) was used for the ionization of the samples. Mass spectra were monitored either in positive or in negative mode (depending on the chemical structure) with pulsed ionization (λ = 337 nm; nitrogen laser). Spectra were measured in reflectron mode using a delayed extraction of 120 ns. The elemental analysis measurements were performed on a Fisons EA 1110 CHNS apparatus (Thermo Scientific, Waltham, MA, US).
Thin layer chromatography was performed on TLC Silica gel 60 F254 on Al sheets (Merck, Darmstadt, Germany), and the spots were visualized under UV light. Preparative layer chromatography was executed on PLC Silica gel 60 F254 1 mm on glass plate (Merck). For column chromatography, a Kieselgel 60 (Merck, particle size 0.063-0.200 mm) was used. All reagents used for synthesis were commercial and of analytically pure quality and all organic solvents were of HPLC grade. Organic solutions were dried over anhydrous Na2SO4 and concentrated under reduced pressure at 40 °C (bath temperature). Carotenoid pentynoates were prepared by a published method [28].

Assay for the Antioxidant Activity of the Carotenoids/TEAC Assay
A literature procedure [31] was applied with modifications. The 2,2′-azino-bis(3ethylbenzothiazoline-6-sulfonic acid) radical cation (ABTS •+ ) was produced in aqueous solution by reacting ABTS •+ in 7 mM and potassium persulfate in 2.45 mM concentrations. The stock ABTS •+ solution was prepared 12-16 h before the experiments, and stored at room temperature in dark. The Molar masses were obtained by an Autoflex II MALDI instrument (Bruker Daltonics, Bremen, Germany). 2,5-Dihydroxy-benzoic acid (DHB) was used for the ionization of the samples. Mass spectra were monitored either in positive or in negative mode (depending on the chemical structure) with pulsed ionization (λ = 337 nm; nitrogen laser). Spectra were measured in reflectron mode using a delayed extraction of 120 ns. The elemental analysis measurements were performed on a Fisons EA 1110 CHNS apparatus (Thermo Scientific, Waltham, MA, US).
Thin layer chromatography was performed on TLC Silica gel 60 F 254 on Al sheets (Merck, Darmstadt, Germany), and the spots were visualized under UV light. Preparative layer chromatography was executed on PLC Silica gel 60 F 254 1 mm on glass plate (Merck). For column chromatography, a Kieselgel 60 (Merck, particle size 0.063-0.200 mm) was used. All reagents used for synthesis were commercial and of analytically pure quality and all organic solvents were of HPLC grade. Organic solutions were dried over anhydrous Na 2 SO 4 and concentrated under reduced pressure at 40 • C (bath temperature). Carotenoid pentynoates were prepared by a published method [28].

Assay for the Antioxidant Activity of the Carotenoids/TEAC Assay
A literature procedure [31] was applied with modifications. The 2,2 -azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) radical cation (ABTS •+ ) was produced in aqueous solution by reacting ABTS •+ in 7 mM and potassium persulfate in 2.45 mM concentrations. The stock ABTS •+ solution was prepared 12-16 h before the experiments, and stored at room temperature in dark. The absorbance of the ABTS •+ solution was set to 0.70 ± 0.05 at 734 nm by dilution with ethanol. Trolox was dissolved in ethanol, and the carotenoids and the carotenoid conjugates in tetrahydrofuran (THF). The antioxidants (in final concentrations of 1.25, 2.5, 3.75, and 5 µM) were incubated with ABTS •+ solution at 37 • C for 6 min, respectively. For the statistical analysis, the measurements of absorbance were carried out at each concentration in triplicate (n = 3), and each measurement was repeated 2 times. SD% was lower than 5% in each case. The percentage inhibition of absorbance at 734 nm was calculated as (A 0 − A antioxidant )/A 0 , where A 0 is the absorbance of the ABTS •+ solution and A antioxidant is the absorbance measured after the addition of the antioxidant. The calculated values were plotted against the final concentration of the antioxidants. The slopes of the curves were compared with that for trolox, and the trolox equivalent antioxidant capacity (TEAC) value is the ratio of the slopes for the antioxidant and for trolox.
Statistical analysis of the TEAC values was performed by the ANOVA method (Microsoft Excel 2016) ( Table 2). To examine the relationship among the TEAC values of the carotenoids, flavonoids, and those of their conjugates, two-tailed t tests were calculated. An F test was performed in each case to determine the type of the t test to be used. Less than 5% in the t test was regarded as a significant difference (Table 3). For the calculations, the functions of Microsoft Excel (2016) were used.

Acetylation of Flavonoids
Daidzein (1) or chrysin (7) (8 mmol) was dissolved in acetic anhydride (30 mL), and was stirred at 60 • C for 10 min. Dry pyridine (8 mL) was added dropwise to the stirred mixture at 60 • C. After 5 min, a white precipitate appeared. After 2 h, the reaction mixture was poured onto ice and stirred for an hour. The formed white precipitate was filtered out, and washed with water thoroughly. After drying above potassium hydroxide for 2 days, the crude products were recrystallized from ethanol.  (2) (0.412 g, 1.21 mmol) was dissolved in 20 mL CH 2 Cl 2 , imidazole (0.248 g, 3.6 mmol, 3 eq.) in 10 mL CH 2 Cl 2 was added dropwise at −15 • C in ice/salt bath. The reaction mixture was stirred at room temperature for 2 h under N 2 . A white precipitate appeared. The reaction mixture was diluted with CH 2 Cl 2 (100 mL), was washed with 3M HCl (3 × 100 mL) and water (100 mL), dried (Na 2 SO 4 ), and the solvent was evaporated under reduced pressure. The crude products (0.41 g) were crystallized from EtOH giving white crystals 5-Acetoxy-7-hydroxy-2-phenyl-4H-chromen-4-one (9): Chrysin diacetate (8) (1.0 g, 2.96 mmol) was dissolved in 50 mL CH 2 Cl 2 , imidazole (0.603g, 8.87 mmol, 3 eq.) in 40 mL CH 2 Cl 2 was added dropwise at −15 • C in ice/salt bath. The reaction mixture was stirred at room temperature for 2 hours under N 2 . A white precipitate appeared, which was filtered out and washed with cold CH 2 Cl 2 . The crude product (0.79 g) was recrystallized from EtOH and yielded white crystals (0.42 g, 48%).

Deprotection
The acetylated 7-bromohexyl ether of daidzein (4) or chrysin (10) (5.0 mmol) was dissolved in dry MeOH (80 mL), and freshly prepared sodium methylate solution (10 mL) was added dropwise to the mixture, which was stirred at room temperature for 2 h. The pH of the mixture was controlled and the amount of NaOMe was replaced due to the formation of phenolates. After the complete disappearance of the starting material, the solution was neutralized by Amberlite IRC 86 acidic resin, and was evaporated in vacuum.

Azide Substitutions
The 7-bromohexyl ether of daidzein (5) or chrysin (11) (4.0 mmol) and sodium azide (2.6 g, 40 mmol, 10 eq.) were dissolved in dry DMF. The reaction mixture was stirred at 60 • C for 2 days under N 2 . When the TLC showed the disappearance of the starting material, the excess sodium-azide and byproducts were filtered out. The mother liquor was diluted with CHCl 3 (600 mL) and washed with water (3 × 200 mL). The organic phase was dried on Na 2 SO 4 and filtered. The solvent was evaporated under reduced pressure, and the crude product was recrystallized from EtOH.

Azide-Alkyne Click-Reactions
The carotenoid pentynoates (13,14,15) (0.1 mmol) were dissolved in dry dichloromethane (4 mL), and the 7-azidohexyl ether derivatives of flavonoids (6,12) (1.5 or 2.5 eq.) and bistriphenylphosphanocopper(I)-butyrate (C 3 H 7 COOCu(PPh 3 ) 2 ) (6 mg) were added. The reaction mixtures were stirred under nitrogen, in darkness. When TLC showed the disappearance of the starting materials (typically in 24 h), the solvent was evaporated in vacuum, and the residue was purified by column chromatography. After crystallization, the products were stored in closed ampoules under argon. Yields are given for the bistriazoles and monotriazoles separately. The analytical samples were further purified by preparative TLC.  13