2024 Volume 72 Issue 7 Pages 648-657
Butin and butein are significant bioactive flavanones derived from plants, existing as tautomers of each other. However, their physicochemical attributes, such as their spectral profiles under varying experimental conditions in aqueous solutions and established chromatographic methods for distinguishing between them, remain undetermined. In this study, we determined the basic properties of butin and butein using conventional spectroscopic, reversed-phase, and chiral HPLC analyses. The spectra of the synthesized butin and butein were analyzed using a UV-Vis spectrophotometer in several solvents with different polarities as well as in aqueous solutions at various pH values. Furthermore, the behavior of the measured spectra was reproduced by calculations to reveal the effects of the solvent and pH on the spectra of butin and butein in organic and aqueous solutions. Subsequently, we assessed the structural stability of butin and butein using reversed-phase HPLC, which revealed that butein is unstable compared with butin in a general culture medium. The synthesized butin was effectively separated into R- and S-isomers with positive and negative Cotton effects, respectively, via HPLC using a chiral column. These findings will aid in uncovering the individual properties of both butin and butein that may have been concealed by their tautomerism and enable the synthesis of S-butin, which is typically challenging and time-consuming to isolate.
Plants contain several types of flavonoids and chalcones, which are polyphenols. Extensive research has been conducted to elucidate their biological activities and apply them to pharmaceutical drugs and health food products.1,2) Butin, a type of flavonoid possessing three hydroxy groups at the 7, 3′, and 4′ positions of the flavanone skeleton, and butein, the chalcone derivative of butin (Chart 1), are isolated from some plants as glycosides or aglycones.3,4) Several studies on butein have been performed, investigating its biological activity and use in chemotherapy.5–10) In contrast to this, research focused on butin is comparatively limited, although a considerable number of studies have identified butin as a flavonoid in plants using the HPLC technique.11–16) Therefore, there is room for further research and it is anticipated that studies on the biological activities and chemotherapy applications of butin and butin-butein systems will increase.
(a) DIPEA, dry CH2Cl2, MOMCl; 93% yield. (b) DIPEA, dry CH2Cl2, MOMCl; 69% yield. (c) Dry EtOH, KOH; 33% yield. (d) Co-solvent mixture of EtOH and aqueous Na2HPO4 solution; 59% yield. (e) Dry cosolvent mixture of acetonitrile and MeOH, BiCl3; 53% yield. (f) Dry cosolvent mixture of acetonitrile and MeOH, BiCl3; 54% yield.
It is essential to obtain the physicochemical properties, including the spectral profiles, of butin and butein in aqueous solutions reflecting experimental conditions, as well as establish chromatographic techniques to differentiate between them. However, studies investigating the fundamental and physicochemical properties of butin and butein are very few compared to those on other flavonoids, and there is insufficient information on the fundamental UV-Vis absorption behavior of butin,which has been considered as a minor flavonoid, to adequately perform biochemical experiments.16–20) Butin and butein are tautomers (Chart 1) of each other. Butin undergoes decyclization to form butein, which in turn undergoes cyclization to form butin under various environmental conditions. Therefore, it is essential to avoid the effects caused by the presence of the other tautomer when determining which compound, either butin or butein, exhibits activity in biological assays. Despite the importance of evaluating the effects of the tautomers on biological activities, to the best of our knowledge, there are no studies evaluating the ratio of butin and butein. Furthermore, butin exhibits stereoisomerism, with S-butin{(−)-butin} predominantly found in natural products.2,3) The effects of this stereoisomerism should be considered when using butin, which is synthesized using a conventional technique without chiral selectivity. However, there are no reports on the stereoisomers of synthesized butin in the case of biological and chemotherapeutic experiments.
Herein, we studied the physicochemical properties of the synthesized butin and butein to obtain widely usable datasets for chemical and biological experiments such as health-food development and pharmaceutical research. Butin and butein were synthesized via a partially modified synthetic scheme involving cyclization from a precursor chalcone to a precursor flavanone, and a deprotection reaction was performed on the precursors to generate butin and butein, respectively. Additionally, their physicochemical properties were analyzed, and we measured the UV-Vis spectra of butin and butein in solvents with varying polarities and demonstrated the pH dependence of the spectral changes of butin and butein in aqueous solutions. Furthermore, we assessed the structural stabilities of butein and butin against tautomerization in a general culture medium using reversed-phase HPLC and analytically separated the R-/S-isomers of the synthetic butin using a chiral HPLC technique.
Butin and butein were synthesized as shown in Chart 1. Compounds 2 and 4 were obtained after compounds 1 and 3 underwent protection reactions, respectively, using methoxy-methyl chloride (MOMCl). Compound 5 was obtained via the condensation reaction between compounds 2 and 4 in the presence of a base (KOH) in ethanol. These syntheses were similar to those previously reported by Kagawa et al.21) Compound 7 was obtained through the cyclization of compound 5 using an aqueous solution in the presence of Na2HPO4 instead of H3PO4.22) This cyclization method offers the advantage of easier manipulation compared to using an unstable anhydrous HCl-methanol (MeOH) solution.21) Finally, the MOM groups of phenolic compounds 5 and 7 were deprotected using BiCl3 to form 6 (butein) and 8 (butin), respectively.23,24) Using this deprotection method aided in effectively suppressing the generation of a tautomer byproduct, which is typically obtained through a conventional scheme using a methanol solution with a high HCl concentration.21,24)
Spectral Profiles of Butin and Butein in Various SolventsThe solubilities of butin and butein were poor in pure water and cyclohexane. Subsequently, highly concentrated MeOH solutions of each compound were prepared. These solutions were diluted with four solvents with varying polarities: MeOH, dimethyl sulfoxide (DMSO), cyclohexane, and water, and the resulting 20 µM butin or butein solutions were used for UV-Vis spectra analysis (each sample solution except MeOH contained a negligible amount of MeOH).
Figure 1 shows the UV-Vis absorbance spectra of butin (black lines) and butein (gray lines) in four solvents: MeOH, DMSO, cyclohexane, and water (Figs. 1a–d, respectively). The spectral profiles of butin in each solvent were nearly identical, characterized by two local λmax values at approximately 280 nm (shorter wavelength) and 310 nm (longer wavelength), with the exception when water was employed as the solvent, where a longer wavelength value of 318 nm was observed (Fig. 1, Table 1). Moreover, for butin, the local λmax value around the weak band located around approximately 300–320 nm in water was higher by approximately <10 nm compared to those in the other solutions, indicating that the polarities of the solvents had a minimal effect on the spectra of butin in the solvents employed (Fig. 1d). The spectral profiles of butein were very similar, with nearly identical characteristic λmax values at 380–381 nm for MeOH, cyclohexane, and water. However, the characteristic local λmax value around 381 nm observed in these solutions shifted to a longer wavelength of 393 nm in DMSO; furthermore, a new band appeared around λ = 470–570 nm with a local λmax = 517 nm (Fig. 1b, Table 1).
A simulated spectrum in each solvent calculated using the TD-DFT/UB3LYP/6-311 + G(d) basis set is displayed in the inset. The simulated spectra of butin and butein are shown in black and gray lines, respectively.
| Solvent | Butin | Butein | ||
|---|---|---|---|---|
| MtOH | 278 | 313 | 381 | |
| DMSO | 277 | 310 | 393 | 517 |
| Cyclohexane | 278 | 311 | 380 | |
| Water | 279 | 318 | 381 | |
The values are derived from the spectra displayed in Fig. 1.
We also simulated a spectrum in each solvent, calculated using the TD-DFT/UB3LYP/6-311 + G(d) level in the Gaussian program25) (Fig. 1 inset, Supplementary Figs. S1, S2). The absorption behaviors of butin and butein in our experiment were accurately reproduced by the calculations, with the exception of butein in DMSO (Fig. 1b, its inset: gray lines). We, therefore, speculated that the experimental absorption band of butein at λ = 470–570 nm in DMSO that was not observed when simulating the absorption behavior of the butein molecule might arise owing to the solvation of a complex molecule, such as butein, with DMSO molecules.
The spectra shown in Fig. 1 for butin are evidently different from those of butein, although a recent study demonstrated that the two compounds had similar spectra in a methanolic solution.20) This unexpected result may likely be attributed to the fact that a different compound from butin was synthesized by the authors and provided as butin for the UV-Vis spectroscopic analysis, based on the 1H-NMR spectrum presented in Supplementary Fig. S2 of the abovementioned study.20) Conversely, our spectral results are consistent with the fact that the absorbance peak of an open-ring chalcone is generally observed in the longer wavelength region compared to that of a closed-ring flavanone.17) Furthermore, our spectral profile of butin in methanol well matched the predicted spectral profile based on the date set of local λmax values of metabolic butin from M. ramannianus.26)
pH Dependency of the Spectral Profiles of Butin and Butein in Aqueous SolutionsFigure 2a-1 shows the spectra of butin in aqueous solutions at pH 2.0, 3.0, 4.0, and 5.0. These spectra had nearly identical profiles, with characteristic local maximum wavelengths at λmax = approx. 279 nm and approx. 311 nm in the λ = 250–350 nm region. Figure 2a-2 presents the spectra of butin in aqueous solutions at pH 5.0, 6.0, 7.0, 8.0, and 9.0. At pH = 6.0 to 9.0, the spectral shapes of butin changed dramatically as the pH values increased, with decreased absorbance at λ = approx. 260–290 nm and increased absorbance at λ = approx. 300–350 nm. Figure 2a-3 displays the spectra of butin in aqueous solutions at pH 9.0, 10.0, 11.0, and 12.0; these spectra had nearly identical profiles, with a maximum wavelength at λmax = 335 nm.
(a) Butin: (a-1) The pH values are 2.0 (gray dotted line), 3.0 (gray solid line), 4.0 (black dotted line), and 5.0 (black solid line). (a-2) The pH values are 5.0 (gray dotted line), 6.0 (gray solid line), 7.0 (black dotted line), 8.0 (black solid line), and 9.0 (black dashed line). (a-3) The pH values are 9.0 (gray dotted line), 10.0 (gray solid line), 11.0 (black dotted line), and 12.0 (black solid line). (b) Butein: (b-1) The pH values are 2.0 (gray dotted line), 3.0 (gray solid line), 4.0 (black dotted line), 5.0 (black solid line), and 6.0 (black dashed line). (b-2) The pH values are 6.0 (gray dotted line), 7.0 (gray solid line), 8.0 (gray dashed line), 9.0 (black dotted line), 10.0 (black solid line), 11.0 (black dashed line), and 12.0 (black dotted-dashed line). (c) The calculated spectral profiles of butin and butein in states where protons (-nH+) have been subtracted from each molecule (Chart 2). (c-1) Calculated spectra of butin (gray line), butin-H+ (black line), butin-2H+ (black dashed line), and butin-3H+ (black dotted line). (c-2) Calculated spectra of butein (gray line), butein-H+ (black line), butein-2H+ (black dashed line), butein-3H+ (black dotted line), and butein-4H+ (black dotted dash line).
The spectra of butein in aqueous solutions at pH 2.0, 3.0, 4.0, 5.0, and 6.0 are presented in Fig. 2b-1. These spectra have almost identical spectral profiles, with a maximum wavelength at λmax = 377–378 nm. Figure 2b-2 displays the spectra of butein in aqueous solutions at pH 6.0, 7.0, 8.0, 9.0, 10.0, 11.0, and 12.0. The spectra of butein in the basic region were red-shifted (from λmax = 378 to λmax = 451 nm) as the pH values of the aqueous solutions increased. This red-shift phenomenon is consistent with the results for xanthohumol, a prenylated chalcone, as well as those for some recently reported hydroxychalcone molecules.27,28)
In the abovementioned studies, the red shift in the spectra and coloring of chalcones in aqueous solutions was attributed to the ionization of the chalcone molecules.27,28) Therefore, we calculated the vertical transition energy and oscillator strength using the TD-DFT/UB3LYP/6-311G(d) level to suggest a possible explanation for both the large absorption change in the butin (flavanone) spectra (Fig. 2a-2) and the red shift in the butein (chalcone) spectra (Fig. 2b-2). Initially, the geometry optimizations in the water environment were performed by substituting the O− group for an OH group on the non-ionized butin and butein, as shown in Chart 2 and Supplementary Figs. S3 and S4. The simulated spectra of butin and its ionized forms in water are shown in Fig. 2c, and Supplementary Figs. S3 and S4, following calculations utilizing the optimized structure of the ionized molecules.
In the calculated spectrum of butin-H+ (Fig. 2c-1, black line, Supplementary Fig. S3), the peak (local λmax = approx. 270 nm) of the band (λ = 250–280 nm) was dramatically smaller compared to that of butin (Fig. 2c-1, gray line), and a notable band (λ = approx. 300 to approx. 330 nm) with a local λmax of approximately 315 nm appeared. Similar behaviors were observed in the calculated spectra of butin-2H+ and butin-3H+ (Fig. 2c-1, black dashed and dotted lines, respectively, Supplementary Fig. S3), although each calculated absorption peak (local λmax = approx. 270 nm) was not smaller than that of butin-H+. Furthermore, no significant changes in the main absorption bands (λ = approx. 300–330 nm) were observed in the calculated spectra of butin-H+, butin-2H+, and butin-3H+; therefore, the experimental spectra of butin for pH 9 or higher (Fig. 2a-3) represent the spectra of ionized butin molecules with one or more protons subtracted. Moreover, the change in the calculated spectra shown in Fig. 2c-1 closely mirrors that observed in the experimental spectra in Fig. 2a-2, although the calculated band (λ = approx. 300–330 nm) (Fig. 2c-1) is shifted towards a slightly shorter wavelength compared to the band in the experimental spectrum (λ = approx. 310–350 nm) (Figs. 2a-2, -3). The large shift in the absorption band of butin in the experimental spectra in the pH 8.0–9.0 range (Fig. 2a-2) can be explained by the ionization of a butin molecule with one proton subtracted.
The band (λ = approx. 370–410 nm) in the calculated spectrum of butein shifted towards longer wavelengths ranging from λ = approx. 440–480 nm, approx. 430–470 nm, and approx. 500–540 nm for butein-H+, butein-2H+, and butein-4H+, respectively (Fig. 2c-2; gray solid, black solid, black dashed, and black dotted dash lines, respectively, Supplementary Fig. S4). In the calculated spectrum of butein-3H+, two bands (λ = approx. 380–430 nm and approx. 540–600 nm) emerged towards longer wavelengths compared to the band (λ = approx. 370–410 nm) in the calculated spectrum of butein (Fig. 2c-2, black dotted and gray solid lines, respectively). Therefore, the red shift observed in the experimental absorption spectra of butein (Fig. 2b-2) can be qualitatively explained by the ionization of a butein molecule with one or more protons subtracted. Moreover, the experimental results obtained by measuring the absorption spectra of butein in aqueous solutions with pH values up to 12.0 (Fig. 2b-2) were consistent with the calculated results obtained from calculating the spectra of butein with up to two protons subtracted.
Assessment of the Structural Stability of Butin and Butein during Tautomerization in a Culture Medium Using Reversed-Phase HPLCAs mentioned earlier, it is essential to assess the stability of butin and butein against their tautomerizations when investigating their individual biological activities, as these compounds undergo tautomerization owing to various conditions in biological solutions such as pH and temperature, among others.29) Therefore, we assessed the structural stability of butin and butein by monitoring the time course of their tautomerization using reversed-phase HPLC.
Currently, HPLC studies aimed at separating only butin and butein are limited, and in most studies, butin and butein have been separated as components of several flavonoids or chalcones in plant extracts by HPLC(-MS) in gradient mode.14–16) In this study, first, butin and butein were separated using HPLC in an isocratic mode. The retention times of butin and butein obtained from each pure solution were approximately 3.5–5 (Fig. 3a) and 8–10 min (Fig. 3b), respectively, meaning that they could be separated in a co-existing solution (Fig. 3c).
The reversed-phase HPLC conditions are described in Experimental.
Second, we tested the structural stability of butin and butein in a generally used medium, Dulbecco’s Modified Eagle Medium (D-MEM) buffer, for cell culture under standard culture conditions (pH 7.2, 37 °C). Each sample (butin or butein) was prepared at a 500 µM concentration in the medium, and the stability of each compound was assessed by monitoring the time-course changes in the respective HPLC profiles, as shown in Fig. 4. When the incubation of the butin medium solution was started, a butin peak was observed at approximately 4 min (Fig. 4a-1, black arrow) without any butein peak. The butin peak was clearly retained 3 h later (Fig. 4a-2, black arrow). Furthermore, butin predominantly did not tautomerize to butein, as evidenced by the fact that the negligible peak of butein at 8–10 min did not grow during the 12 h of incubation (Figs. 4a-2–5, white arrows). When incubation of the butein medium solution was started, the butein peak was observed at 8–10 min, while the butin peak was not observed (Fig. 4b-1, white arrow). A distinct peak at approx. 4 min was observed, indicating that butin had undergone tautomerization from butein, accompanied by a decrease in the peak area of butein (Figs. 4b-2–5, white and black arrows) compared to the beginning (Fig. 4b-1, white arrow) as time progressed.
(a) Butin (500 µM) in D-MEM buffer (initial preparation) was incubated at 37 °C for 0 h (a-1), 3 h (a-2), 6 h (a-3), 9 h (a-4), and 12 h (a-5). (b) Butein (500 µM) in D-MEM buffer (initial preparation) was incubated at 37 °C for 0 h (b-1), 3 h (b-2), 6 h (b-3), 9 h (b-4), and 12 h (b-5). The reversed-phase HPLC conditions are described in Experimental.
Figure 5 presents the time–course changes in the tautomerization ratios of butin (filled rectangles) and butein (open rectangles) obtained from the results of the reversed-phase HPLC shown in Fig. 4 (see Experimental for tautomerization ratio calculations). The tautomerization of butin to form butein was suppressed following a 3 h incubation and reached an equilibrium state with approximately 10% butein in the culture medium (Fig. 5, filled rectangles). Conversely, the tautomerization of butein to form butin was not suppressed and proceeded with time, with approximately 60% conversion achieved after 12 h of incubation.
The error bars represent the standard errors of the mean (n = 3).
Therefore, the results of our reversed-phase HPLC experiments showed that butein is more susceptible to tautomerization than butin in a general culture medium. Generally, flavanones are unstable, while their corresponding chalcones are stable under basic conditions.29) However, our experiment clearly showed the opposite outcome when a weak base solution near physiological experimental conditions was employed. Further precise research regarding to structural stabilities of flavonoids is needed.
Separation of the R/S Stereoisomers of the Synthesized Butin Using Chiral HPLCRacemic butin was obtained using a general synthetic method without chiral selection, as shown in Chart 1. The butin synthesized in this study possessed both R- and S-isomers, resulting from a chiral center at the 2-position on the C-ring of the flavanone skeleton (Chart 1). The R- and S-isomers were effectively separated in a 1 : 1 ratio using a chiral column under isocratic conditions using a mixed solution of ethanol and hexane (Fig. 6). The fraction peak with the shorter retention time at approx. 5.5 min {peak area: (5.76 ± 0.31) × 106 (average ± standard error of the mean (S.E.M.)) (n = 3)} and that with the longer retention time at approx. 7.0 min {peak area: (5.69 ± 0.32) × 106 (average ± S.E.M.) (n = 3)} represented the R- and S-isomers, respectively (Fig. 6a), based on a previous report on the flavanone liquiritigenin and Fig. 6b.19) Figure 6b shows that the R- {shorter retention time at approx. 5.5 min, (2.57 ± 0.12) × 105 (average ± S.E.M.) (n = 3)} and S- {longer retention time at approx. 7.0 min, peak area: (2.59 ± 0.14) × 105 (average ± S.E.M.) (n = 3)} isomers exhibit positive and negative cotton effects, respectively.
(a) Profiles obtained from a UV-Vis detector. (b) Profiles obtained from a CD detector. The chiral HPLC conditions are described in Experimental.
In this study, we demonstrated the basic properties of butin and butein for application not only in chemistry research but also in biological experiments. The spectral profile of butin notably changed in the weakly acidic to weakly basic pH range, whereas that of butein was largely red-shifted in the neutral to basic pH range. This property allows both butin and butein to serve as indicators of pH.28) The sensitivity of butin and butein to pH must be carefully considered in biochemical experiments and cell cultures. Reversed-phase HPLC studies revealed that the structural stability of butin and butein was substantially different under our experimental conditions; therefore, stability assessments against tautomerization are crucial for polyphenols for biological and pharmaceutical research. Furthermore, the R- and S-isomers of the synthesized butin were separated using chiral HPLC. This will enable us to synthetically and easily obtain S-butin, which is typically time-consuming to obtain as a pure compound through natural product purification, thereby facilitating more effective research.
All solvents and reagents used for the synthesis were obtained from commercial sources and used without further purification. The reactions were performed under an argon or nitrogen atmosphere using dry solvents. TLC was performed on silica gel precoated onto an aluminum sheet (silica gel 60 F254, Merck, Darmstadt, Germany). The silica gel used for flash column chromatography was purchased from Kanto Chemical (Tokyo, Japan, silica gel 60N, spherical, neutral, 40–50 or 100–210 µm). To identify the synthesized products, 1H-NMR and high-resolution mass spectra were obtained using an ECP-500 (500 MHz, JEOL, Tokyo, Japan) and an LC-MS-9300 (Shimadzu, Kyoto, Japan), respectively (Supplementary Figs. S5–S10).
SynthesisCompound 2To a solution of 1 (3.96 g, 26.0 mmol) and N, N-diisopropylethylamine (DIPEA) (11.2 mL) in dry CH2Cl2 (80 mL), MOMCl (3.14 g, 39.0 mmol) was added dropwise and stirred for >1 h at room temperature. Following this, water was added, and the mixture was extracted using CH2Cl2. The extract was dried over anhydrous Na2SO4 and evaporated. Subsequently, 2 (4.76 g, yield 93%) was yielded following silica gel column chromatography (hexane : acetone = 5 : 1). 1H-NMR (500 MHz, CDCl3) δ = 7.652 [d, J = 9.0 Hz, 1H], 6.594 [d, J = 2.0 Hz, 1H], 6.549 [dd, J = 9.0, 2.5 Hz, 1H], 5.206 [s, 2H], 3.477 [s, 3H], 2.565 [3H]. High resolution-electrospray ionization (HR-ESI)-MS: m/z 197.0809 [M + H+] (Calcd. 197.0814).
Compound 4To a solution of 3 (9.06 g, 65.6 mmol) and DIPEA (39.5 mL) in dry CH2Cl2 (90 mL), MOMCl (15.8 g, 197 mmol) was added dropwise and stirred at room temperature overnight. Subsequently, water was added, and the mixture was extracted using CH2Cl2. The extract was dried over anhydrous Na2SO4 and evaporated. Compound 4 (10.3 g, yield 69%) was obtained using silica gel column chromatography (hexane : acetone = 4 : 1). 1H-NMR (500 MHz, CDCl3) δ = 9.873 [s, 1H], 7.687 [d, J = 1.5 Hz, 1H], 7.516 [dd, J = 8.3, 1.8 Hz, 1H], 7.291 [d, J = 9.0 Hz, 1H], 5.334 [s, 2H], 5.302 [s, 2H], 3.534 [s, 3H], 3.528 [3H]. HR-ESI-MS: m/z = 227.0915 [M + H+] (Calcd. 227.0919).
Compound 5KOH (5.05 g, 90.0 mmol) was added in one portion to a solution of 2 (1.63 g, 8.31 mmol) and 4 (1.89 g, 8.36 mmol) in dry ethanol (EtOH) (120 mL) and stirred for 3 d at room temperature. The reaction mixture was neutralized by adding 10% acetic acid (AcOH) in EtOH and evaporated. The residue was added to ethyl acetate (AcOEt) and washed with water. The organic layer was dried over anhydrous Na2SO4 and evaporated. Compound 5 (1.11 g, yield 33%) was yielded following silica gel column chromatography (hexane : acetone = 3 : 1). 1H-NMR (500 MHz, CDCl3) δ = 7.854 [d, 1H, J = 8.5 Hz], 7.834 [d, 1H, J = 16 Hz], 7.483 [d, 1H, J = 2.5 Hz], 7.448 [d, 1H, J = 15.5 Hz], 7.285 [dd, 1H, J = 8.5, 2.3 Hz], 7.218 [d, 1H, J = 8.5 Hz], 6.647 [d, J = 2.0 Hz 1H], 6.599 [dd, 1H, J = 8.5, 2.3 Hz], 5.305 [s, 2H], 5.299 [s, 2H], 5.231 [s, 2H], 3.563 [s, 3H], 3.531 [s, 3H], 3.494 [3H]. ESI-MS: m/z = 405.1553 [M + H+] (Calcd. 405.1549).
Compound 6 (butein)BiCl3 (0.45 g) was added in one portion to a solution of 5 (0.51 g, 1.26 mmol) in a dry cosolvent mixture of acetonitrile (12.25 mL) and MeOH (0.25 mL) and stirred overnight at 50 °C. The reaction mixture evaporated. Compound 6 (0.18 g, yield 53%) was subsequently obtained via silica gel column chromatography (benzene : AcOEt = 55 : 45). 1H-NMR (500 MHz, DMSO-d6) δ = 10.644 [s, 1H], 9.757 [s, 1H], 9.095 [s, 1H], 8.14 [d, 1H, J = 9.0 Hz], 7.657 [s, 1H × 2], 7.271 [d, 1H, J = 2.0 Hz], 7.203 [dd, 1H, J = 8.0, 1.8 Hz], 6.808 [d, 1H, J = 7.5 Hz], 6.396 [dd, 1H, J = 9.0, 1.8 Hz], 6.271 [d, 1H, J = 2.5 Hz]. HR-ESI-MS: m/z = 271.0494 [M − H+] (Calcd. 271.0607).
Compound 7An aqueous solution of Na2HPO4 (20 mL) was added to a solution of 5 (0.80 g, 1.98 mmol) in EtOH (20 mL) and stirred for approximately 24 h at 115 °C. The mixture was then extracted using AcOEt. The organic layer was dried over anhydrous Na2SO4 and evaporated. Following this, compound 7 was obtained (0.47 g, yield 59%) using silica gel column chromatography (benzene : AcOEt = 12 : 1). NMR (500 MHz, CDCl3) δ = 7.876 [d, 1H, J = 8.5 Hz], 7.293 [d, 1H, J = 1.5 Hz], 7.211 [d, 1H, J = 8.5 Hz], 7.071 [dd, 1H, J = 8.5, 2.3 Hz], 6.712 [dd, 1H, J = 8.5, 2.5 Hz], 6.685 [d, 1H, J = 2.5 Hz], 5.399 [dd, 1H, J = 13.5, 3.0 Hz], 5.272–5260 [m, 4H], 5.206 [s, 2H], 3.533 [s, 3H], 3.525 [s, 3H], 3.479 [s, 3H], 3.044 [dd, 1H, J = 17.0, 13.5 Hz], 2.814 [dd, 1H, J = 17.0, 3.3 Hz]. HR-ESI-MS: m/z = 405.1553 [M + H+] (Calcd. 405.1549).
Compound 8 (butin)BiCl3 (0.16 g) was added in one portion to a solution of 7 (0.52 g, 1.29 mmol) in a dry cosolvent mixture of acetonitrile (12.5 mL) and MeOH (0.25 mL) and stirred for 6 h at 50 °C. The reaction mixture evaporated. Compound 8 (0.19 g, yield 54%) was yielded following two rounds of silica gel column chromatography (benzene : AcOEt = 55 : 45 and CHCl3:MeOH = 90 : 10). 1H-NMR (500 MHz, CD3OD) δ = 7.723 [d, 1H, J = 8.5 Hz], 6.924 [d, 1H, J = 2.0 Hz], 6.812–6.769 [m, 2H], 6.493 [dd, 1H, J = 9.0, 2.3 Hz], 6.354 [d, 1H, J = 2.0 Hz], 5.321 [dd, 1H, J = 13.0, 3.0 Hz], 3.011 [dd, 1H, J = 17.0, 13.0 Hz], 2.690 [dd, 1H, J = 16.5, 2.8 Hz]. HR-ESI-MS: m/z = 271.0609 [M − H+] (Calcd. 271.0607).
Computational MethodDensity functional theory (DFT) calculations were performed to confirm the UV-Vis spectral profiles of butin, butein, and their ionized species in the solutions. The geometries of butin and butein were fully optimized with the Gaussian16 set of programs using the DFT method with the B3LYP/6-311 + G(d) basis set.25) When optimizing their geometries, solvent effects were included using the Polarized Continuum Model (PCM). The optimized structures exhibited no imaginary frequencies. The vertical excitation energies of the optimized species in solution were calculated at the same DFT level using the time-dependent (TD) DFT approach.
UV-Vis SpectroscopyThe spectra of the synthesized butin and butein in the solvents were measured using a V-760 spectrophotometer (JASCO, Tokyo, Japan). Methanol solutions of butin or butein were prepared at a concentration of 4.0 mM, and the highly concentrated samples were diluted 200-fold to achieve a final concentration of 20 µM using spectroscopic-grade organic solvents (FUJIFILM Wako, Osaka, Japan) or ion-exchanged RO-water (≥18.2 MΩ·cm; Merck). The pH values of the aqueous solvents were adjusted by adding HCl or NaOH solutions, except for aqueous solvents with pH values ranging from 6.0 to 9.0. Phosphate buffers (20 mM) (pH = 6.0, 7.0, and 8.0) and Tris–Cl buffer (20 mM) (pH = 9.0) were used as aqueous solvents. Each measurement was performed using a quartz cell with a 10 mm light path at 20.0 ± 0.1 °C.
HPLC AnalysisReversed-Phase HPLC AnalysisButin and butein, as tautomers, were separated by passing through a column (TSKgel ODS-100 V, 5 µm, 4.6 mm I.D. × 150 mm, TOSOH, Tokyo, Japan) for reversed-phase HPLC purification (using a PU-2089 Plus pump unit and UV-2075 Plus detector unit, JASCO). The monitoring wavelength employed was 275 nm. A 20 µL sample solution, initially diluted with the developing solvent consisting of methanol and 0.5% phosphate aqueous solution (v/v = 55/45) to achieve a 50 µM butin or butein concentration, was applied to the column. The developing solvent flowed in isocratic mode at a rate of 1 mL/min at room temperature (22–23 °C).
When assessing the structural stabilities of butin and butein, 10 mM methanolic solutions of both compounds were diluted to a concentration of 500 µM using D-MEM (high glucose) culture medium (FUJIFILM Wako) as the initial preparation for the time course incubation. The prepared samples were placed in a dark incubator (SLC-25A, Mitsubishi Electric Engineering, Tokyo, Japan) with the temperature maintained at 37 ± 0.5 °C. After measurement at each time point, the peak area of butein was calculated from the corresponding HPLC profile (Fig. 4). The tautomerization ratio of butin (filled rectangle in Fig. 5) converted to butein (denoted as Rbutin) was determined by calculating the fraction of the peak area of butein (Abutein 4a, xh) at each time point (xh; x = 0, 3, 6, 9, and 12) in Fig. 4a divided by the peak area at 0 h (Abutein 4b, 0h) in Fig. 4b-1. The peak area at 0 h was assumed to be the area when butin was fully converted to butein (Fig. 4a-1). Therefore, Rbutin was obtained using the formula: Rbutin = Abutein 4a, xh/Abutein 4b, 0h. Similarly, the tautomerization ratio of butein (open rectangle in Fig. 5) converted to butin (denoted as Rbutein) was obtained by subtracting the remaining fraction (Rbutein, re), obtained by dividing the peak area (Abutein 4b, xh) at each time point (xh; x = 0, 3, 6, 9, and 12) in Fig. 4b with the peak area (Abutein 4b, 0h) at 0 h, as shown in Fig. 4b-1, from 1. Therefore, Rbutein was obtained using the formula: Rbutein = 1−Rbutein, re = 1−Abutein 4b, xh/Abutein 4b, 0h.
Chiral HPLC AnalysisThe stereoisomers of the synthesized butin were separated using a column (CHIRALPAK IA, 5 µm, 4.6 mm I.D. × 250 mm, DAICEL, Osaka, Japan ) for chiral HPLC purification (with an LC-20AD pump unit and SPD-M20A detector unit, Shimadzu) coupled with a CD detector (CD-1595, JASCO). The monitoring wavelength was set at 300 nm. The sample solution comprised 0.25 µg/µL of the synthesized butin in the developing solvent, which consisted of the eluent hexane and ethanol (v/v = 70/30). Subsequently, 20 µL of the sample solution was injected for separation using the chiral HPLC column. The developing solvent flowed in isocratic mode at a rate of 1 mL/min at room temperature (22–23 °C).
This work was supported by the Hokuriku University Special Research Grants (T. K., T. T., M. M., and H. Suzuki). Grants-in-Aid for Scientific Research C (JP21K12694; T. T. and H. Suzuki, JP19K05378; H. Saito), JST Spring Grant (JPMJSP2102; R. H.), Center for Advanced Computing Infrastructure and Nano Materials and Technology, JAIST (R. H. and K. S.), and a joint research project between Tokyo Denki University and Hokuriku University (3220025; J. M. and T. K.).
T. K. conceived, designed, and performed the research, including experiments and analyses. J. M. performed the calculations. H. Suzuki, and M. M. participated in the syntheses. H. Saito contributed to the spectral analyses. T. T. contributed to the reversed-phase HPLC analysis. R. H. and K. S. participated in the chiral HPLC analysis. T. K. wrote the paper. All authors read and approved the published version of this manuscript.
The authors declare no conflict of interest.
This article contains supplementary materials.
