HPLC Fluorescence Method for Eugenols in Basil Products Derivatized with DIBI

Makoto Takada; Masashi Sakamoto; Haruna Yamada; Naoya Kishikawa; Junpei Mutoh; Yukihide Shiraishi; Naotaka Kuroda; Mitsuhiro Wada

doi:10.1248/cpb.c21-00575

Abstract

Eugenols (Eugs) such as eugenol (Eug), methyleugenol (MeEug), and linalool (Lin) in basil product are the main bioactive components of basil products and have a terminal double-bond. A sensitive HPLC-fluorescence method for Eugs derivatized with 4-(4,5-diphenyl-1H-imidazol-2-yl)iodobenzene (DIBI) was developed. Good separation of DIB-Eugs was achieved within 20 min on an Atlantis T3 column (50 × 2.1 mm i.d., 3 µm) with a mobile phase of methanol-water. The calibration curves obtained with Eug standards showed good linearities in the range of 0.1–50 µM (r ≥ 0.999). The limits of detection at a signal-to-noise ratio (S/N) = 3 for Eug, MeEug, and Lin were 1.0, 6.0, and 4.8 nM, respectively. The limits of quantitation (S/N = 10) of the Eugs were lower than 19.9 nM. The accuracies for the Eugs were within 96.8–104.6%. The intra- and inter-day precisions as relative standard deviations for the Eugs were less than 1.2 and 9.6% (n = 3). The recoveries of Eug, MeEug, and Lin were 99.0 ± 0.1, 98.0 ± 0.2, and 96.0 ± 0.4% (n = 3), respectively. The DIB-Eugs were confirmed to be stable for 2 h (>90%) at room temperature and 24 h (>95%) at 4 °C. These parameters of the proposed method were useful for the simultaneous determination of Eugs in basil products. Therefore, the developed method may be a powerful tool for the quality evaluation of dried commercially available basil products.

1. Introduction

Ocimum basilicum L. (basil) has been used for food preservation, flavoring, and as a medicine since antiquity. For example, the plant is widely used as a food and in oral care products, and its essential oil is used in perfumery.¹⁾ Eugenols (Eugs) such as eugenol (Eug), linalool (Lin), and methyleugenol (MeEug) are the main bioactive components of basil products.²⁾ Eug and Lin have antioxidant and antimicrobial activities against both Gram-positive and Gram-negative bacteria, and are generally recognized as safe by the U. S. Food and Drug Administration (FDA).^3–5) MeEug is a phenylpropanoid synthesized by the shikimic acid pathway and is reported to increase the risk of liver cancer.⁶⁾ In addition, it is well-known that Eugs contents in basil product exceedingly vary due to its cultivating or processing conditions. It would therefore be useful to evaluate the quality of basil product by sensitive analyzing these Eugs.

Several analytical methods have been reported for the determination of Eugs, including GC/MS,^7–9) HPLC-UV,^10–12) liquid chromatography-tandem mass spectrometry (LC-MS/MS),^13–15) and enzyme-linked immunosorbent assay (ELISA)¹⁶⁾ methods. Although GC/MS, LC-MS/MS, and ELISA methods are sensitive, sample pretreatment is complicated such as solid-phase extraction for clean-up or the required equipment is expensive. HPLC-UV methods are simple and easy but insufficiently sensitive for the quantitation of Eugs in food samples.

HPLC-fluorescence (FL) is more sensitive than UV methods but requires derivatization of non-FL compounds such as Eugs.^17,18) FL derivatization necessitates the reaction of suitable functional groups with a reagent to generate an FL derivative. Higashi reported an HPLC-FL method for the determination of Eug by pre-column derivatization with 4-(N-chloroformylmethyl-N-methylamino)-7-nitro-2,1,3-benzoxadiazole (NBD-COCl).¹⁹⁾ This reagent reacts with the phenolic component of Eug under mild conditions to allow high sensitivity detection. However, MeEug could not be detected, because it does not react with NBD-COCl.

Kishikawa et al. developed 4-(4,5-diphenyl-1H-imidazole-2-yl)iodobenzene (DIBI) as an FL derivatization reagent based on the Mizoroki-Heck coupling reaction in the presence of palladium acetate (Pd(OAc)₂) as a catalyst, and used this reaction for HPLC assays of pharmaceutical products with a terminal double bond.²⁰⁾ In this study, we established an assay for Eugs having terminal double bond using this DIBI/HPLC-FL detection method to quantify Eugs for the quality evaluation of dried commercially available basil products (Fig. 1).

Fig. 1. Estimated Derivatization Reaction of Eug with DIBI, and Chemical Structures of the Eugs

Experimental

Chemicals and Reagents

Eug (>99%) was obtained from Tokyo Kasei Kogyo Co. (Tokyo, Japan). MeEug, Lin (Wako 1st Grade), Pd(OAc)₂, dimethylacetamide (DMAc), ethanol (EtOH) (Guaranteed Reagent), and methanol (HPLC grade) were purchased from Wako Pure Chemical Corporation (Osaka, Japan). DIBI was synthesized according to a previous method.²⁰⁾ Other reagents used were of analytical reagent grade. Ultra-pure water was prepared using a PURELAB flex-UV from Organo (Tokyo, Japan).

Instruments

The HPLC system consisted of two chromatographic pumps (LC-20AD, Shimadzu, Kyoto, Japan), a degasser (DGU-20A5R, Shimadzu), an autosampler (SIL-20AC, Shimadzu), a system controller (CBM-20 A, Shimadzu), an Atlantis T3 column (50 × 2.1 mm i.d., 3 µm, Waters, Milford, MA, U.S.A.), a column oven (CTO-6AS, Shimadzu), an RF20AXS FL detector (Shimadzu), and an LC workstation (LabSolutions/LC solution version 1.21, Shimadzu). A mixture of water (solvent A) and methanol (solvent B) was used as the mobile phase and the total flow rate was set at 0.5 mL/min. The gradient program was as follows: the ratio of solvent B was initialized at 62% (0–13.7 min), changed to 85% (13.7–20 min) to elute the components, then returned to 62% for equilibration of the analytical column. The column temperature was set at 40 °C, and eluates were monitored at 330 nm (λex) and 400 nm (λem).

Preparation of Standard Samples

To 100 µL of 1 µM Eugs in EtOH, 500 µM DIBI in EtOH (50 µL), 40 µM Pd(OAc)₂ in aq. solution (100 µL), and 500 mM DMAc in EtOH (50 µL) were successively added and mixed well for 10 s. After deoxygenation by N₂ purge for 10 s, the reaction mixture was heated at 100 °C for 70 min. Following filtration through a membrane filter (0.45 µm, TORAST™ Disc, Shimadzu GLC), a 5-µL aliquot was injected into the HPLC system.

Pretreatment of Basil Product Samples

The triturated basil product (0.1 g) was added to EtOH (10 mL), sonicated for 5 min, then the sample solution was centrifuged at 700 × g for 5 min. To the supernatant (100 µL), 5000 µM DIBI in EtOH (50 µL), 1000 µM Pd(OAc)₂ aq. solution (100 µL) and 500 mM DMAc in EtOH (50 µL) were successively added and mixed well for 10 s. After deoxygenation by N₂ purge for 10 s, the reaction mixture was heated at 100 °C for 70 min. After filtration through a membrane filter, a 5-µL aliquot was injected into the HPLC system.

Methods Validation

The validation parameters were linearity, limit of detection (LOD), and the limit of quantitation (LOQ), determined experimentally using standard Eugs, and accuracy, precision, and recovery were determined using standard Eugs spiked with basil product. Linearity was determined by analyzing standard Eugs (0.1–100 µM). The LOD and LOQ were defined as the concentration at a signal-to-noise ratio of 3 (S/N = 3) and S/N = 10, respectively. The accuracy and precision of the method were determined by triplicate measurements of three concentrations for each compound. Recovery was calculated as the ratio of the slope of the spiked calibration curve to that of the standard. The concentration of Eugs in each basil product was determined by a standard addition method (n = 3).

Results and Discussion

Separation of DIB-Derivatives of Eugs

The separation of DIB-Eugs was examined using Cadenza HS-C18, Scherzo SM-C18, TSKgel NH₂-100,ZIC-HILIC, Develosil, Gemini C6-Phenyl, XBridge BEH C18, CORTECS C18, XSELECT HSS C18, and Atlantis T3 columns. Eugs had lower molecular weights than DIBI (the FL derivatization reagent), making it difficult to separate DIB-Eugs from excess DIBI chromatographically. The Atlantis T3 column gave the best separation of the Eugs. The separation of DIB-Eug and -MeEug was not sufficiently completed using other columns under the examined condition. The high surface coverage of the Atlantis T3 column by endcapping provided higher retention times of the analytes and superior peak shapes. Figure 2 shows typical chromatograms obtained for (a) Eug standards each at 1 µM and (b) basil product G. DIB-Eugs were acceptably separated from DIBI and other interfering peaks within 20 min. The retention times of Eug, MeEug, and Lin were 15.6, 16.0, and 16.6 min, respectively. It is considered that a peak observed at 18.3 min in each chromatogram of Fig. 2 is derived from DIBI.

Fig. 2. Chromatograms of (a) Standard Eugs (1 µM), and (b) the Basil Product G

Extraction Conditions

The extraction conditions, such as extracting solution (EtOH, 80% EtOH and DMAc, Fig. 3(a)) and extraction time (from 0 to 30 min, Fig. 3(b)), were examined to increase the yield of Eugs from the basil products. Based on the results, extraction with EtOH for 5 min was used in following experiments.

Fig. 3. Effects of (a) Extracting Solution, and (b) Extraction Time on the Relative Peak Area of DIB-Eugs in the Basil Product

The highest relative peak area was taken as 100.

Derivatization Conditions

Using the derivatization conditions for Eug standards shown in Experimental, small peaks were detected from the basil products. It was therefore necessary to re-optimize the derivatization conditions, such as the DIBI, DMAc, and Pd(OAc)₂ concentrations, and the heating temperature and time, using the basil products. First, we examined the DIBI concentration in the range of 500–10000 µM. The highest relative peak area of DIB-Eugs was obtained when 5000 µM of DIBI was used (Fig. 4(a)). Using DMAc concentrations ranging from 100 to 1000 mM gave the highest and constant relative peak area of DIB-Eugs in the basil products (data not shown). Next, the effects of Pd(OAc)₂ concentration in the range of 10–1500 µM on relative peak area were examined (Fig. 4(b)). The highest and constant relative peak area was obtained when more than 750 µM of Pd(OAc)₂ was used and thus 1000 µM of Pd(OAc)₂ was selected in the following experiments. We also investigated the heating temperature in the range of 70–120 °C and found that 100 °C gave the highest relative peak area for each Eug. Finally, the heating time in the range of 20–90 min was examined. The highest relative peak area was observed following heating for 70 min (Fig. 4(c)). The decrease of relative peak area of DIB-Lin with 90 min of heating time was observed due to its stability. Using these optimized conditions, we found that high concentrations of DIBI and Pd(OAc)₂ were necessary using the basil products, suggesting that components in addition to Eugs in the basil products were also consuming the reagents.

Fig. 4. Effects of (a) DIBI and (b) Pd(OAc)₂ Concentrations, and (c) Heating Time on the Relative Peak Area of DIB-Eugs in the Basil Product

The highest relative peak area was taken as 100.

The stabilities of the DIB-Eugs from the basil products in the autosampler maintained at room temperature and 4 °C were determined. As shown in Fig. 5, DIB-Eug and DIB-MeEug were confirmed to be stable for 6 h (>90%) at room temperature, whereas DIB-Lin was stable for only 2 h (>90%). The DIB-Eugs were confirmed to be stable for 24 h (>95%) at 4 °C.

Fig. 5. Stability of DIB-Eugs in the Basil Product in the Autosampler at (a) Room Temperature, and (b) 4 °C

The initial relative peak area was taken as 100.

Method Validation

Calibration curves obtained with standard Eugs showed good linearities in the ranges of 0.1–50 µM (r = 0.999). The LODs of Eug, MeEug, and Lin were 1.0, 6.0, and 4.8 nM, respectively. The LOQs were lower than 19.9 nM. These parameters for the proposed method are summarized in Table 1.

Table 1. Calibration Curve, Limit of Detection (LOD), and Limit of Quantitation (LOQ) of Standard Eugs

Compound	Dynamic range, µM	r	LOD, nM (S/N = 3)	LOQ, nM (S/N = l0)
Eug	0.1–50	0.999	1.0	3.4
MeEug	0.1–50	0.999	6.0	19.9
Lin	0.1–50	0.999	4.8	16.1

Accuracy, intra- and inter-day precisions, and recovery by the proposed method were evaluated by using basil product spiked with known concentrations of Eugs (0.1, 2, and 50 µM) and the results are summarized in Table 2. The assay accuracies for Eug, MeEug, and Lin were 100.2 ± 0.0 to 104.6 ± 1.0%, 98.8 ± 1.6 to 100.4 ± 0.3%, and 96.8 ± 0.1 to 98.9 ± 0.1%, respectively. The intra-day precisions (relative standard deviations, RSDs) for Eug, MeEug, and Lin were <0.4, <1.2, and <0.5%, respectively. The inter-day precisions (RSDs) were less than 9.6%. The recovery was calculated by the ratio of the slope of the spiked calibration curve to that of the standard. Those for Eug, MeEug, and Lin were 99.0 ± 0.1, 98.0 ± 0.2, and 96.0 ± 0.4%, respectively. These parameters of the proposed method were acceptable for the reliable analyses of these compounds in the basil products.

Table 2. Method Validation of the Proposed Method

Compound	Spiked concentration, µM	Accuracy^a⁾ %	Precision^a⁾ RSD%		Recovery^a,b⁾ %
Compound	Spiked concentration, µM	Accuracy^a⁾ %	Intra-day	Inter-day	Recovery^a,b⁾ %
Eug	0.1	101.7 ± 1.1	0.1	1.4	99.0 ± 0.1
	2	104.6 ± 1.0	0.4	1.0
	50	100.2 ± 0.0	0.1	1.7
	0.1	100.4 ± 0.3	0.3	4.9
MeEug	2	98.8 ± 1.6	1.2	6.0	98.0 ± 0.2
	50	100.0 ± 0.0	0.2	6.4
	0.1	96.8 ± 0.1	0.5	5.6
Lin	2	98.9 ± 0.1	0.3	9.6	96.0 ± 0.4
Lin	50	97.9 ± 0.2	0.3	2.5	96.0 ± 0.4

a) n = 3, b) Ratio of the slope of the spiked calibration curve to that of the standard.

The sensitivity of the proposed method was compared with those of the previous reports and the results are shown in Table 3. The proposed method is considered to be more sensitive than GC/MS methods,^8,9) HPLC-UV methods,^10–12) another HPLC-FL method using derivatization with NBD-COCl,¹⁹⁾ and an ELISA method.¹⁶⁾ Our method is comparable to the LC-MS/MS method for Eug,¹⁵⁾ but is less sensitive for MeEug.¹⁴⁾ However, the proposed method does not require expensive instruments and laboratory equipment for sample pretreatment. Additionally, the HPLC-FL method, following derivatization with DIBI, could analyze Eug and MeEug simultaneously, in contrast to the derivatization method with NBD-COCl. Therefore, the proposed method is more useful than conventional methods.

Table 3. Comparison of the LOD of the Proposed Method with Previous Reports

Compound	LOD, nM
Compound	Our method	GC/MS	HPLC-UV	HPLC-FL¹⁹⁾	LC-MS/MS	ELISA¹⁶⁾
Eug	1.0	3.8 × 10⁴ ⁸⁾	3.1 × 10³ ¹¹⁾	36.5	9.0¹⁵⁾	73.1
MeEug	6.0	8.4 × 10³ ⁹⁾	5.6 × 10² ¹⁰⁾	—	0.2¹⁴⁾	22.4
Lin	4.8	1.4 × 10⁴ ⁸⁾	9.7 × 10² ¹²⁾	—	—	—

Analysis of Basil Products

We assessed the applicability of the proposed method by analyzing seven dried commercially available basil products (Table 4). Eugs in all samples could be quantified by the standard addition method and were in the range of 29.3–274.1, 9.5–164.2, and 85.0–584.8 µg/g for Eug, MeEug, and Lin, respectively. The concentration of Eugs differed by a factor of 10 between the products.

Table 4. Concentrations of Eugs in Basil Products by the Standard Addition Method

Subject	Concentration, µg/g^a⁾
Subject	Eug	MeEug	Lin
A (from Egypt)	38.6 ± 0.3	54.9 ± 0.7	214.4 ± 1.9
B (from U.S.A.)	150.7 ± 2.3	9.5 ± 0.2	584.8 ± 6.2
C (from Egypt)	81.3 ± 0.3	164.2 ± 2.8	85.0 ± 0.2
D (from Egypt)	29.3 ± 0.6	52.2 ± 1.9	116.3 ± 4.6
E (from Spain)	274.1 ± 3.8	57.3 ± 1.1	238.9 ± 2.4
F (from U.S.A.)	65.5 ± 0.4	51.0 ± 0.4	192.5 ± 5.8
G (from Egypt)	86.0 ± 1.5	68.4 ± 1.6	405.8 ± 4.3

a) n = 3.

Among the basil products, a higher concentration of MeEug than Eug and Lin was observed in product C. The reason for this high concentration is unknown but this concentration is consistent with those measured using conventional methods.²¹⁾ FDA food additive regulations do not authorize the use of MeEug as a synthetic flavoring compound for use in food, in response to a food additive petition in 2018.²²⁾ However, it was reported that hepatocellular adenomas occurred only in rodents exposed to high doses of MeEug.²³⁾ The concentration of MeEug in basil product C investigated in the current study was much lower than that found in food additives.

Conclusion

A sensitive HPLC-FL method for the determination of Eugs in basil products derivatized with DIBI was developed. The specificity of DIBI for compounds with a terminal double bond allowed the simultaneous determination of Eug and MeEug. Well-defined DIB-Eug peaks were obtained by separation, without any interfering peaks, using the proposed method. The obtained validation parameters for Eug, MeEug, and Lin in the basil products were acceptable. Furthermore, this method was successfully used to analyze these compounds in actual basil products.

The developed method showed sufficient sensitivity, reproducibility, and applicability for quantifying the Eugs in dried basil products and may be a powerful tool for the quality evaluation of basil products.

Acknowledgments

This work was supported by the Smoking Research Foundation in Japan (No. 2020G009).

Conflict of Interest

The authors declare no conflict of interest.

References

1) Seyed E. S., DARU, 14, 128–130 (2006).
2) Ruiz-Rico M., Pérez-Esteve É., Bernardos A., Sancenón F., Martínez-Máñez R., Marcos M. D., Barat J. M., Food Chem., 233, 228–236 (2017).
3) FDA, Database of Select Committee on GRAS Substances (SCOGS) Reviews (2006).
4) Liu K., Chen Q., Liu Y., Zhou X., Wang X., J. Food Sci., 77, 1156–1161 (2012).
5) Gang D. R., Wang J., Dudareva N., Nam K. H., Simon J. E., Lewinsohn E., Pichersky E., Plant Physiol., 125, 539–555 (2001).
6) Abtahi Froushani S. M., Zarei L., Esmaeili Gouvarchin Ghaleh H., Mansori Motlagh B., Aloyan S., Trchounian A., Avicenna J. Phytomed., 6, 526–534 (2016).
7) Di Cesare L. F., Forni E., Viscardi D., Nani R. C., J. Agric. Food Chem., 51, 3575–3581 (2003).
8) Muráriková A., Ťažký A., Neugebauerová J., Planková A., Jampílek J., Mučaji P., Mikuš P., Molecules, 22, 1221 (2017).
9) Abtahi Froushani S. M., Zarei L., Esmaeili Gouvarchin Ghaleh H., Mansori Motlagh B., Avicenna J. Phytomed., 6, 526–534 (2016).
10) Gursale A., Dighe V., Parekh G., J. Chromatogr. Sci., 48, 59–62 (2010).
11) Li Y. H., Sun Z. H., Zheng P., Chromatographia, 60, 709–713 (2004).
12) Ródenas-Montano J., Carrasco-Correa E. J., Beneito-Cambra M., Ramis-Ramos G., Herrero-Martínez J. M., J. Chromatogr. A, 1296, 157–163 (2013).
13) Guénette S. A., Ross A., Marier J. F., Beaudry F., Vachon P., Eur. J. Pharmacol., 562, 60–67 (2007).
14) Jeong S. H., Jang J. H., Lee G. Y., Yang S. J., Cho H. Y., Lee Y. B., J. Chromatogr. B Analyt. Technol. Biomed. Life Sci., 1126–1127, 121743 (2019).
15) Sun P., Gao Y., Lian U. Food Anal., Methods, 10, 3217–3224 (2017).
16) Shen X., Wu X., Liu L., Kuang H., Food Agric. Immunol., 30, 1318–1331 (2019).
17) Wada M., Yamahara K., Ikeda R., Kikura-Hanajiri R., Kuroda N., Nakashima K., Biomed. Chromatogr., 26, 21–25 (2012).
18) Nagaoka M. H., Nagaoka H., Kondo K., Akiyama H., Maitani T., Chem. Pharm. Bull., 54, 922–924 (2006).
19) Higashi Y., J. Anal. Chem., 70, 1401–1405 (2015).
20) Kishikawa N., Ohkuma M., Wada M., Ohyama K., Ikeda R., Nakashima K., Kuroda N., J. Chromatogr. A, 1218, 3002–3006 (2011).
21) Zheljazkov V. D., Callahan A., Cantrell C. L., J. Agric. Food Chem., 56, 241–245 (2008).
22) Gooderham N. J., Cohen S. M., Eisenbrand G., Fukushima S., Guengerich F. P., Hecht S. S., Rietjens I. M. C. M., Rosol T. J., Davidsen J. M., Harman C. L., Murray I. J., Taylor S. V., Food Chem. Toxicol., 145, 111585 (2020).
23) Williams G. M., Iatropoulos M. J., Jeffrey A. M., Duan J. D., Food Chem. Toxicol., 53, 187–196 (2013).

Corresponding author

Register with J-STAGE for free!