Chromatographic Evaluation and Characterization of Constituents of Sunflower Seed Extract Used as Food Additives

Abstract

Sunflower seed extract, an antioxidant agent registered on the List of Existing Food Additives in Japan, was evaluated using HPLC, and three common constituents were detected. These peaks were identified as monocaffeoylquinic acids (3-O-caffeoylquinic acid, 4-O-caffeoylquinic acid, and 5-O-caffeoylquinic acid [chlorogenic acid]). Upon scrutinizing other components, dicaffeoylquinic acids (isochlorogenic acids; 3,4-di-O-caffeoylquinic, 3,5-di-O-caffeoylquinic, and 4,5-di-O-caffeoylquinic acids) were also identified. Structures of two newly isolated compounds were determined to be 3-O-(3S-2-oxo-3-hydroxy-indole-3-acetyl)-5-O-caffeoylquinic and 4-O-(3S-2-oxo-3-hydroxy-indole-3-acetyl)-5-O-caffeoylquinic acids. To identify the components that contribute to the antioxidant activity of sunflower seed extract, we fractionated the food additive sample solution and examined the active fractions for 2,2-diphenyl-1-picrylhydrazyl (DPPH) radical scavenging activity. Monocaffeoylquinic and dicaffeoylquinic acids showed high DPPH activity, including their contribution to the antioxidant activity of this food additive. DPPH radical scavenging activity of the new compounds showed almost the same value as that of the positive control, Trolox. Therefore, the contribution of these compounds was also considered.

Introduction

Most existing food additives officially defined under the Food Sanitation Act in Japan are natural product extracts containing various ingredients. However, the characteristic and effective components of the existing food additives used since before 1995, which are mostly derived from natural sources, have not always been properly defined owing to poor characterization of the ingredients in the respective raw materials and food additive products themselves. To resolve these issues, we investigated the ingredients of existing food additives and aimed to obtain the necessary chemical data to set standardized specifications.¹⁾

Sunflower seeds are widely used in the food and nutraceutical industries because of their high oil and protein contents and other valuable bioactive components.²⁾ Antioxidants, such as flavonoids and phenolic acid, are well-known components of sunflower seeds, and in addition to antioxidant activity, sunflower seeds also have antimicrobial and antidiabetic properties.^2–4)

Sunflower seed extract (SSE) using raw material is an antioxidant agent listed on the List of Existing Food Additives in Japan.⁵⁾ The origin, manufacturing method, and essence are described as “obtained by extraction with hot water or water-containing ethanol from the seeds or the oil-pressed phase of the seeds of the Asteraceae sunflower (Helianthus annuus LINNÉ), and the active ingredients are isochlorogenic and chlorogenic acids.” The standard test for SSE is listed in the Voluntary Standards for Existing Food Additives (the 5th edition published by the Japan Food Additives Association), which describes the methods for quantifying isochlorogenic and chlorogenic acids.

However, to the best of our knowledge, no scientific data on the ingredients of SSE additive products have been reported so far, and ingredients other than isochlorogenic and chlorogenic acids may be the main ingredients. For example, the raw material, sunflower seeds, have been suggested to contain other polyphenols.⁶⁾ In this study, we investigated the composition of SSE to establish its quality standards as a food additive. Additionally, the phytochemical profile and antioxidant activity of the constituents obtained from the product are described and discussed with regard to their use as antioxidants. To the best of our knowledge, this is the first study to analyze the constituents of SSE as a food additive.

Results and Discussion

Isolation and Characterization of Phenolic Constituents

To characterize the constituents of SSE, three additive products (A–C) were analyzed using reversed-phase (RP)-HPLC. Three peaks that were commonly detected as the main peaks were confirmed (Fig. 1). To elucidate these peaks, we attempted to separate and refine the components of SSE. The SSE was dissolved in H₂O and separated using Diaion HP-20 column chromatography to yield H₂O (Fraction 1; F1), 30% MeOH (Fraction 2; F2), and MeOH eluate extracts (Fraction 3; F3). Figure 2 shows the HPLC chromatograms of Fractions 1–3. Fractions 2 and 3 were further separated using Sephadex LH-20 and/or YMC GEL ODS chromatographies with aqueous MeOH to yield 3-O-caffeoylquinic acid (3), 5-O-caffeoylquinic acid (chlorogenic acid) (4), 4-O-caffeoylquinic acid (5), caffeic acid (6), 3,4-di-O-caffeoylquinic acid (7), 3,5-di-O-caffeoylquinic acid (8), 4,5-di-O-caffeoylquinic acid (9), and compounds 1 and 2 (Fig. 3). Known compounds 3–9 were identified through direct comparison with authentic specimens or via spectral comparisons with data reported in literature.^7,8)

Fig. 1. HPLC Chromatograms of Sunflower Seed Extract Products

A–C; Products A–C.

Fig. 2. HPLC Chromatograms of Extracts Sunflower Seed Extract Fractionated Using Diaion HP-20 Column

F1–3; Fractions 1–3.

Fig. 3. Chemical Structures Isolated from Sunflower Seed Extract

Compound 1 was isolated as a pale-yellow amorphous powder. C₂₆H₂₅O₁₂N was assigned as its molecular formula based on the high-resolution electrospray ionization (HR-ESI)-MS (m/z: 542.1301 [M − H]⁻, Calcd for C₂₆H₂₅O₁₂N − H: 542.1304) and ¹³C-NMR signals. The UV spectrum showed absorption maxima at 233, 298, and 329 nm. The ¹H-NMR spectrum revealed that Compound 1 composed a caffeoyl group (ABX-type aromatic proton signals: δ 6.76 [1H, d, J = 8 Hz], 6.94 [1H, dd, J = 2, 8 Hz], and 7.03 [1H, d, J = 2 Hz], and trans-olefin proton signals: δ 6.20 and 7.51 [each 1H, d, J = 16 Hz]) and a quinic acid proton signal with a coupling pattern assigned using ¹H–¹H correlation spectroscopy (COSY). Other signals were analyzed, and signals corresponding to 2-oxo-3-hydroxy-indole-3-acetic acid were observed.^9,10) The linking position of each unit was confirmed by the cross-peaks between quinic acid H-5 (δ 5.24) and caffeoyl group C-9′ (δ 168.9), including H-2′, 6′ (δ 7.03, 6.94) and C-7′ (δ 147.3), and H-7′, 8′ (δ 7.51, 6.20) and C-9′ (δ 168.9), together with quinic acid H-3 (δ 5.19) and 2-oxo-3-hydoroxy-indole-3-acetic acid unit C-11″ (δ 170.1), including H-4″ (δ 7.42) and C-3″ (δ 74.9), and H-10″ (δ 3.10) and C-11″ (δ 170.1), which were determined using heteronuclear multiple bond correlations (HMBC) (Fig. 4). The absolute configuration of the 3-hydroxyoxindole moiety was determined using the circular dichroism (CD) method. The CD spectrum of Compound 1 was recorded, and a positive Cotton effect was observed in the 250–220-nm range. Thus, the absolute configuration of Compound 1 was determined to be 3S.⁹⁾ Based on these data, the structure of Compound 1 was established to be 3-O-(3S-2-oxo-3-hydroxy-indole-3-acetyl)-5-O-caffeoylquinic acid, as shown in Fig. 3.

Fig. 4. Selected HMBC Correlations (H→C) of Compounds 1 and 2

Furthermore, the molecular formula of Compound 2 was determined to be C₂₆H₂₅O₁₂N based on HR-ESI-MS (m/z: 542.1301 [M − H]⁻, Calcd for C₂₆H₂₅O₁₂N − H: 542.1304) and ¹³C-NMR signals. The UV spectrum showed absorption maxima at 240, 299, and 330 nm, and the spectral data were similar to those of Compound 1. The ¹H-NMR data also showed signals similar to those of Compound 1, except for the chemical shifts at the H-3–5 positions of quinic acid. Therefore, Compound 2 is also suggested to have quinic acid, a caffeoyl group, and 2-oxo-3-hydroxy-indole-3-acetic acid unit. The structure was established according to the HMBC correlations between the quinic acid H-5 (δ 5.31) and caffeoyl group C-9′ (δ 168.0), including H-2′, 6′ (δ 7.03, 6.93) and C-7′ (δ 147.4), and H-7′, 8′ (δ 7.45, 6.12) and C-9′ (δ 168.0), together with the quinic acid H-4 (δ 4.91) and 2-oxo-3-hydoroxy-indole-3-acetic acid unit C-11″ (δ 169.9), including H-4″ (δ 7.37) and C-3″ (δ 74.6), and H-10″ (δ 3.08, 3.14) and C-11″ (δ 169.9) (Fig. 4). The CD spectrum of Compound 2 was recorded, and 3S was determined to be its absolute configuration based on the positive Cotton effect in the 250–220-nm range.⁹⁾ Based on these data, Compound 2 was established as 4-O-(3S-2-oxo-3-hydroxy-indole-3-acetyl)-5-O-caffeoylquinic acid, as shown in Fig. 3.

DPPH Radical Scavenging Activity

The 2,2-diphenyl-1-picrylhydrazyl (DPPH) assay is a standard method for evaluating the antioxidant capacity of food additives.¹¹⁾ Hence, to determine the components responsible for the antioxidant activity of SSE, the DPPH radical scavenging activities of three food additive products (A–C) and Fractions 1–3 that were derived from Product A, which were fractionated using Diaion HP-20 column chromatography, were evaluated. The half-maximal effective concentration (EC₅₀) values of Products A–C were found to be in the range of 136.4–229.5 µg/mL, and the EC₅₀ values of Product A-derived Fractions 1–3 were 115.6, 97.8, and 23.5 µg/mL, respectively (Table 1). Considering the respective yields of Fractions 1–3 (62.5, 20.8, and 1.4 g), Fractions 1 and 2, with their high yield, are suggested to contribute greatly to the antioxidant activity of SSE. Although the yield of Fraction 3 was low, it showed the highest activity value; therefore, it is also thought to contribute to the antioxidant activity. Monocaffeoylquinic acids (Compounds 3–5), including 5-O-caffeoylquinic acid (chlorogenic acid), were recognized amongst the Fractions 1 and 2 ingredients that were confirmed from the RP-HPLC chromatograms (Fig. 2). In contrast, the HPLC chromatogram of Fraction 3 showed dicaffeoylquinic acids (Compounds 7–9) (isochlorogenic acids). Evaluation of their DPPH radical scavenging activity showed strong activity with EC₅₀ values of 12.3–14.2 and 6.1–7.0 µM for monocaffeoylquinic and dicaffeoylquinic acids, respectively (Table 2). Therefore, the activity tended to increase with the number of caffeoyl groups. Overall, monocaffeoylquinic and dicaffeoylquinic acids exhibited potent activity, including their contribution to the antioxidant activity of SSE.

Table 1. DPPH Radical Scavenging Activities of Sunflower Seed Extract Products and Fractions-Derived Product

	EC50 (µg/mL)
Product A	136.4
Product B	152.9
Product C	229.5
Fraction 1 (F1)	115.6
Fraction 2 (F2)	97.8
Fraction 3 (F3)	23.5

Table 2. DPPH Radical Scavenging Activities of Compounds 1–9

	EC50 (µM)
Compound 1	33.5
Compound 2	28.5
3-O-Caffeoylqunic acid (3)	14.2
5-O-Caffeoylquinic acid (5)	13.8
4-O-Caffeoylquinic acid (4)	12.3
Caffeic acid (6)	32.1
3,4-Di-O-caffeoylquinic acid (7)	7.0
3,5-Di-O-caffeoylquinic acid (8)	6.1
4,5-Di-O-caffeoylquinic acid (9)	6.2
Trolox	24.1

Conclusion

The constituents of SSE, an existing food additive, were investigated, and three commonly detected constituents were identified as monocaffeoylquinic acids (3-O-caffeoylquinic acid [3], 4-O-caffeoylquinic acid [4], and 5-O-caffeoylquinic acid [chlorogenic acid] [5]). Other ingredients included dicaffeoylquinic acids (isochlorogenic acids; 3,4-di-O-caffeoylquinic acid [7], 3,5-di-O-caffeoylquinic acid [8], and 4,5-di-O-caffeoylquinic acid [9]). Moreover, we isolated two new compounds and determined their structures to be 3-O-(3S-2-oxo-3-hydroxy-indole-3-acetyl)-5-O-caffeoylquinic acid (1) and 4-O-(3S-2-oxo-3-hydroxy-indole-3-acetyl)-5-O-caffeoylquinic acid (2). We also evaluated the DPPH radical scavenging activity of the additive sample solution fractions and monocaffeoylquinic acids, which were mainly detected using HPLC were found to contribute to the antioxidant activity of SSE. Dicaffeoylquinic acids, found in the low-yield fraction, also showed high DPPH activity values, indicating their contribution to antioxidant activity. The DPPH radical scavenging activity of the new compounds showed almost the same value as that of the positive control Trolox. We believe that these data provide useful information for developing test methods for the quality control of SSE products as food additives.

Experimental

General

Optical rotations were measured using a JASCO-P-1020 digital polarimeter (JASCO Corporation, Tokyo, Japan). UV spectra were recorded using a Shimadzu UVmini-1240 (Shimadzu Corporation, Kyoto, Japan). HR-ESI-MS spectra were obtained using a micrOTOF-Q mass spectrometer (Bruker Daltonics, Billerica, MA, U.S.A.) with direct sample injection and acetonitrile as the solvent. The conditions were optimized in negative-ion mode as follows: drying gas; flow-rate, 3.0 L/min; dry temperature, 200 °C; and capillary voltage, 3.5 kV. NMR spectra were recorded using a Bruker AVANCE500 instrument (500 and 126 MHz for ¹H and ¹³C, respectively; Bruker Biospin, Billerica, MA, U.S.A.) at 300 K, and chemical shifts were expressed as parts per million (ppm) values relative to MeOH-d₄ (δ_H 3.30; δ_C 49.0) on a tetramethylsilane scale. Standard pulse sequences programmed for the instrument (AVANCE500) were used for all two dimensional measurements (COSY, heteronuclear single quantum coherence, and HMBC). J_CH was set at 10 Hz for HMBC analysis. Column chromatography was performed using a Diaion HP-20, MCI-gel CHP-20P (Mitsubishi Chemical Co., Tokyo, Japan), Sephadex LH-20 (Cytiva, Tokyo, Japan), and YMC GEL ODS (YMC Co., Ltd., Kyoto, Japan). The RP-HPLC conditions were as follows: column, L-column ODS (5 µm, 150 × 2.1 mm i.d.; Chemicals Evaluation and Research Institute, Tokyo, Japan); mobile phase, solvent A-5% acetic acid in water and solvent B-acetonitrile (0–30 min, 0–40% B in A; 30–35 min, 40–60% B in A; 35–40 min, 60–85% B in A; and 40–50 min, 85–100% B in A); injection volume, 2 µL; column temperature, 40 °C; flow-rate, 0.3 mL/min; and detection wavelength, 200–400 nm.

Samples and Reagents

Commercial SSEs (Reference numbers: Product A, C2241; B, A1089; and C, A1090) were obtained from the Japan Food Additives Association (JAFA; Tokyo, Japan). Chlorogenic acid (5-O-caffeoylquinic acid), 4-O-caffeoylquinic acid, and 3-O-caffeoylquinic acid were purchased from Nagara Science (Gifu, Japan); 3,4-di-O-caffeoylquinic, 3,5-di-O-caffeoylquinic, and 4,5-di-O-caffeoylquinic acids were purchased from MedChemExpress (NJ, U.S.A.); and caffeic acid was purchased from Tokyo Kasai (Tokyo, Japan). DPPH radical scavenging activity was determined using a DPPH Antioxidant Assay Kit (Dojin Laboratories, Kumamoto, Japan). All the other chemicals were of analytical grade.

Extraction and Isolation

The SSE (100 g) was dissolved in H₂O (500 mL), and then separated by column chromatography using Diaion HP-20 (40 × 5.0 cm i.d.) with aqueous MeOH solvent (0 : 100→30 : 70→100 : 0, 2 L each) in stepwise gradient mode to yield H₂O eluate extract (Fraction 1, 62.5 g), 30% MeOH eluate extract (Fraction 2, 20.8 g), and MeOH eluate extract (Fraction 3, 1.4 g). Fraction 2 (2.0 g) was separated by column chromatography using YMC GEL ODS (40 × 1.1 cm i.d.) with aqueous MeOH to obtain 4-O-caffeoylquinic acid (5, 21.4 mg), and 5-O-caffeoylquinic acid (4, 27.0 mg). Fraction 3 (1.3 g) was separated by column chromatography using Sephadex LH-20 (40 × 1.1 cm i.d.) with aqueous MeOH solvent, and then using YMC GEL ODS (40 × 1.1 cm i.d.) with aqueous MeOH to obtain 3-O-caffeoylquinic acid (3, 2.6 mg), 5-O-caffeoylquinic acid (3, 7.4 mg), caffeic acid (6, 2.7 mg), 3,4-di-O-caffeoylquinic acid (7, 31.0 mg), 3,5-di-O-caffeoylquinic acid (8, 5.8 mg), and 4,5-di-O-caffeoylquinic acid (9, 8.5 mg), along with Compounds 1 (6.6 mg) and 2 (27.6 mg). Compounds 3–9 were identified by directly comparing the authentic specimens, or by comparing their spectral data with those reported in the literature. The physical and spectral data for the new Compounds 1 and 2 are as follows:

Compound 1: Pale yellow amorphous powder. HR-ESI-MS m/z: 542.1301 ([M − H]⁻, Calcd for C₂₆H₂₅O₁₂N − H: 542.1304). UV λ_max (MeOH) nm (log ε): 233 (4.23), 298 (4.13), and 329 (4.20). [α]_D²¹ −23° (c = 0.2, MeOH). CD (MeOH) [θ] (nm): −5.3 × 10⁴ (210), +3.9 × 10⁵ (237), and −1.1 × 10⁴ (262). ¹H-NMR (500 MHz, MeOH-d₄) δ: 7.51 (1H, d, J = 16.0 Hz, H-7′), 7.42 (1H, d, J = 7.5 Hz, H-4″), 7.24 (1H, td, J = 1.5, 7.5 Hz, H-6″), 7.03 (2H, m, H-2′, H-5″), 6.94 (1H, dd, J = 2.0, 8.0 Hz, H-6′), 6.87 (1H, d, J = 7.5 Hz, H-7″), 6.76 (1H, d, J = 8.0 Hz, H-5′), 6.20 (1H, d, J = 16.0 Hz, H-8′), 5.24 (1H, m, H-5), 5.19 (1H, m, H-3), 3.82 (1H, dd, J = 3.5, 7.5 Hz, H-4), 3.10 (2H, d, J = 3.5 Hz, H-10″), and 2.12 and 1.90 (each 2H, m, H-2, H-6). ¹³C-NMR (126 MHz, MeOH-d₄) δ: 180.0 (C-2″), 177.2 (C-7), 170.1 (C-11″), 168.2 (C-9′), 149.6 (C-4′), 147.3 (C-7′), 146.7 (C-3′), 143.2 (C-8″), 131.8 (C-9″), 130.9 (C-6″), 127.8 (C-1′), 125.5 (C-4″), 123.6 (C-5″), 123.1 (C-6′), 116.4 (C-5′), 115.2 (C-2′), 115.0 (C-8′), 111.4 (C-7″), 74.9 (C-3″), 72.9 (C-3), 72.0 (C-5), 71.2 (C-1), 70.0 (C-4), 42.8 (C-10″), 37.4, 35.5 (C-2, 6).

Compound 2: Pale yellow amorphous powder. HR-ESI-MS m/z: 542.1312 ([M − H]⁻, Calcd for C₂₆H₂₅O₁₂N − H: 542.1304). UV λ_max (MeOH) nm (log ε): 245 (4.10), 299 (4.09), and 330 (4.18). [α]_D¹⁹ −19° (c = 0.2, MeOH). CD (MeOH) [θ] (nm): −5.0 × 10⁴ (210), +2.9 × 10⁴ (239), and −0.78 × 10⁴ (265). ¹H-NMR (500 MHz, MeOH-d₄) δ: 7.45 (1H, d, J = 16.0 Hz, H-7′), 7.37 (1H, d, J = 8.0 Hz, H-4″), 7.19 (1H, td, J = 1.5, 8.0 Hz, H-6″), 7.03 (1H, d, J = 2.0 Hz, H-2′), 6.96 (1H, td, J = 1.5, 8.0 Hz, H-5″), 6.93 (1H, dd, J = 2.0, 8.5 Hz, H-6′), 6.84 (1H, d, J = 8.0 Hz, H-7″), 6.77 (1H, d, J = 8.5 Hz, H-5′), 6.12 (1H, d, J = 16.0 Hz, H-8′), 5.31 (1H, m, H-5), 4.91 (1H, dd, J = 3.0, 9.0 Hz, H-4), 4.07 (1H, m, H-3), 3.14 and 3.08 (each 1H, d, J = 16.0 Hz, H-10″), and 1.90–2.20 (4H, m, H-2, 6). ¹³C-NMR (126 MHz, MeOH-d₄) δ: 180.7 (C-2″), 177.0 (C-7), 169.9 (C-11″), 168.0 (C-9′), 149.6 (C-4′), 147.4 (C-7′), 146.7 (C-3′), 143.3 (C-8″), 131.6 (C-9″), 131.0 (C-6″), 127.7 (C-1′), 125.2 (C-4″), 123.6 (C-5″), 123.2 (C-6′), 116.5 (C-5′), 115.3 (C-2′), 114.8 (C-8′), 111.4 (C-7″), 75.6 (C-4), 74.6 (C-3″), 69.0, 68.9 (C-1, 3, 5), 42.5 (C-10″), 39.1, 38.2 (C-2, 6).

DPPH Radical Scavenging Activities

The DPPH radical scavenging activity was evaluated using the DPPH Antioxidant Assay Kit (Dojin Laboratories, Kumamoto, Japan), following the manufacturer’s instructions.^11,12) Briefly, the sample solution, assay buffer, and DPPH working solution were mixed in 96-well plates and incubated in a dark place at 25 °C for 30 min. Absorbance was measured at a wavelength of 517 nm using an Infinite F200 microplate reader (Tecan Group Ltd., Mannedorf, Switzerland). The EC₅₀ was determined by regression line analysis, and Trolox was used as the positive control. All experiments were performed in triplicates.

Acknowledgments

This work was supported in part by a Health Sciences Research Grant (23KA1012) from the Ministry of Health, Labour and Welfare, Japan.

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

This article contains supplementary materials.

References

• 1) Amakura Y., Yoshimura M., Sugimoto N., Akiyama H., Chem. Pharm. Bull., 69, 11–17 (2021).
• 2) Guo S., Ge Y., Na Jom K., Chem. Cent. J., 11, 95 (2017).
• 3) Zoumpoulakis P., Sinanoglou V. J., Siapi E., Heropoulos G., Proestos C., Antioxidants, 6, 46 (2017).
• 4) Karamać M., Kosińska A., Estrella I., Hernández T., Dueñas M., Eur. Food Res. Technol., 235, 221–230 (2012).
• 5) Ministry of Health, Labour and Welfare of Japan, “Notification No. 120” Apr. 16, 1996.
• 6) Grasso S., Pintado T., Pérez-Jiménez J., Ruiz-Capillas C., Herrero A. M., Foods, 9, 445 (2020).
• 7) Iwai K., Kishimoto N., Kakino Y., Mochida K., Fujita T., J. Agric. Food Chem., 52, 4893–4898 (2004).
• 8) Islam M. S., Yoshimoto M., Yahara S., Okuno S., Ishiguro K., Yamakawa O., J. Agric. Food Chem., 50, 3718–3722 (2002).
• 9) Li M., Wang Y., Tsoi B., Jin X.-J., He R.-R., Yao X.-J., Dai Y., Kurihara H., Yao X.-S., Fitoterapia, 99, 48–55 (2014).
• 10) Shataer D., Li J., Duan X.-M., Liu L., Xin X.-L., Aisa H. A., J. Agric. Food Chem., 69, 4111–4119 (2021).
• 11) Shimamura T., Sumikura Y., Yamazaki T., Tada A., Kashiwagi T., Ishikawa H., Matsui T., Sugimoto N., Akiyama H., Ukeda H., Anal. Sci., 30, 717–721 (2014).
• 12) Uchikura T., Kitano T., Yoshimura M., Amakura Y., Biosci. Biotechnol. Biochem., 87, 688–695 (2023).