Food Chemistry 169 (2015) 127–133
Contents lists available at ScienceDirect
Food Chemistry journal homepage: www.elsevier.com/locate/foodchem
Analytical Methods
Simultaneous determination of soy isoflavone glycosides, daidzin and genistin by monoclonal antibody-based highly sensitive indirect competitive enzyme-linked immunosorbent assay Seiichi Sakamoto 1, Gorawit Yusakul 1, Benyakan Pongkitwitoon, Madan Kumar Paudel, Hiroyuki Tanaka ⇑, Satoshi Morimoto Department of Pharmacognosy, Graduate School of Pharmaceutical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan
a r t i c l e
i n f o
Article history: Received 5 May 2014 Received in revised form 30 July 2014 Accepted 1 August 2014 Available online 9 August 2014 Keywords: Daidzin Enzyme-linked immunosorbent assay Genistin Monoclonal antibody Soy isoflavones
a b s t r a c t Soy isoflavones are known as major bioactive compounds in soybean (Glycine max), which is an indispensable food. Despite their utility, the consumption of isoflavones has recently been limited because they exhibit oestrogenic and topoisomerase II inhibitory effects. To assess their intake limitation, accurate, sensitive, and effective quantitative analyses are necessary. In this study, we produced the monoclonal antibody (MAb) against daidzin (DZ) and applied it to an indirect competitive enzyme-linked immunosorbent assay (icELISA) for the simultaneous determination of DZ and genistin (GEN), which are known as two major soy isoflavone glycosides in soy products. Using the DZ-MAb, we developed a sensitive icELISA method, where the limit of detection for DZ and GEN was 1.95 ng/ml. Several validation analyses revealed that the icELISA is sufficiently accurate and sensitive to be used to assess the overconsumption of soy isoflavones, which would lead to the safe dietary intake of soy products. Ó 2014 Published by Elsevier Ltd.
1. Introduction Soy isoflavones are major compounds isolated from soybean (Glycine max), which is an indispensable food material. Soy isoflavones are mainly composed of three glycosides (daidzin (DZ), genistin (GEN), and glycitin) and their aglycones (daidzein, genistein, and glycitein) (Fig. 1). They have recently been the subject of much attention because of their ability to decrease the morbidity rates related to age-related cardiovascular and osteoporosis, hormonedependent breast and prostate cancers, and menopausal symptoms (Cano, García-Pérez, & Tarín, 2010; Mourouti & Panagiotakos, 2013; Park & Weaver, 2012). These effects are mainly attributed to the structural resemblance between isoflavones and estradiol. Contrary to these beneficial effects, adverse effects by overconsumption have also been proposed (Setchell & Cassidy, 1999). In addition, one of the soy isoflavones, genistein, has been reported to induce a DNA-topoisomerase II inhibitory effect by stabilising the covalent DNA cleavage complex, which is an event predicted to cause DNA damage via the activation of p53 and checkpoint kinase 2 (Ye et al., 2001) by the same mechanism as that of topoisomerase II inhibitor, etoposide (Lu, Taya, Ikeda, & Levine, 1998). ⇑ Corresponding author. Tel./fax: +81 92 642 6581. 1
E-mail address: [email protected] (H. Tanaka). These authors contributed equally to this work.
http://dx.doi.org/10.1016/j.foodchem.2014.08.004 0308-8146/Ó 2014 Published by Elsevier Ltd.
Therefore, the relationships between the consumption of dietary isoflavones and infant leukaemia have been clinically studied (Azarova et al., 2010). In Japan, opportunities to intake soy isoflavones are abundant because soy products have been traditionally and generally used in Japanese foods including tofu (soybean curd), kinako (soybean flour), shoyu (soy sauce), miso (soybean paste), and natto (fermented soybeans). In addition, prescription Japanese Kampo medicines mainly derived from kudzu (Pueraria lobata Ohwi) also contain substantial amounts of isoflavones (Boué et al., 2003). As such, the intake of isoflavones has been limited to 75 mg/day (30 mg/day as supplements) by the Food Safety Commission of Japanese Government since 2006 (Food Safety Commission Novel Foods Expert Committee, 2006). In addition to the Japanese case, The Nutrient Data Laboratory of the Agricultural Research Service of the United States Department of Agriculture, in collaboration with Iowa State University, released a Special Interest Database on isoflavones in foods in 1999 and updated it several times until 2008 by increasing the number of soy products (U.S. Department of Agriculture, 2008). In both cases, intake amounts of isoflavones were evaluated on the basis of aglycone forms because the isoflavone glucosides are converted into aglycones by intestinal flora in vivo (Tsuchihashi, Okawa, Nohara, Okabe, & Kinjo, 2004; Yasuda, Mizunuma, Kano, Saito, & Ohsawa, 1996; Yasuda, Ueda, & Ohsawa, 2001) and in vitro (Tsuchihashi, Sakamoto, Kodera, Nohara, & Kinjo, 2008). Therefore, an analytical
128
S. Sakamoto et al. / Food Chemistry 169 (2015) 127–133
(A)
(D)
(B)
(E)
(C)
(F)
3 3
Fig. 1. Structures of major soy isoflavones (A) daidzin (DZ), (B) genistin (GEN), and (C) glycitin and their corresponding aglycones (D) daidzein, (E) genistein, and (F) glycitein.
method to determine isoflavone aglycones has been mainly developed using ultra-high-pressure liquid chromatography (UHPLC) (Fiechter, Opacak, Raba, & Mayer, 2013), HPLC separation (Zeng et al., 2012) coupled with diode array detection (Wang et al., 2013), flow injection chemiluminescence (Wang & Han, 2012), and enzyme-linked immunosorbent assays (ELISAs) (BennetauPelissero et al., 2003; Pongkitwitoon et al., 2011). However, accurate, sensitive, and effective qualitative methods are necessary for the determination of isoflavone glycosides, especially in the case of DZ and GEN because they are two major isoflavone glycosides in soy products before fermentation and in medicinal plants (Wang & Murphy, 1994). To date, few studies have addressed the quantitative analysis of isoflavone glycosides by methods other than HPLC (Hosoda et al., 2008), HPLC/UV-MS (Klejdus et al., 2004), or UHPLC/UV (Toro-Funes et al., 2012). In this study, monoclonal antibody (MAb) against the simplest structure among soy isoflavone glycosides, DZ (DZ-MAb), was produced by several injections of DZ-bovine serum albumin (BSA) conjugates into BALB/c mice, followed by a hybridoma method. Characterisation of DZ-MAb revealed that it exhibits cross-reactivity against DZ and GEN, thereby enabling the development of an effective indirect competitive ELISA (icELISA) for the simultaneous determination of both glycosides of DZ and GEN. Their limit of determination in this system is 1.95 ng/ml. Several validation analyses revealed that the icELISA is sufficiently accurate and sensitive to be used for the determination of soy isoflavones contained in the samples. The preparation of DZ-MAb and its application to the icELISA system are demonstrated in this paper.
2. Materials and methods 2.1. Chemicals and reagents Daidzin (DZ, P99%), daidzein (P97%), genistin (GEN, P99%), and genistein (P99%) were purchased from Fujicco Co. (Kobe, Japan). Bovine serum albumin (BSA, P97%) and albumin from
chicken egg (OVA, P98%) were acquired from Sigma–Aldrich (St. Louis, MO, USA). Freund’s complete and incomplete adjuvants were obtained from Difco (Detroit, MI, USA). Peroxidase-labelled anti-mouse IgG was purchased from Organon Teknika Cappel Products (West Chester, PA, USA). All other chemicals were standard commercial products of analytical-reagent grade. 2.2. Animals Five-week-old male BALB/c mice were purchased from KBT Oriental Co. (Saga, Japan). A standard diet (MF; Oriental Yeast Co., Tokyo, Japan) and water were provided ad libitum. All of the procedures and care related to the animals were approved by the Committee on the Ethics of Animal Experiments, Graduate School of Pharmaceutical Sciences, Kyushu University, and were conducted according to the Guidelines for Animal Experiments of the Graduate School of Pharmaceutical Sciences, Kyushu University (approval no: A24-004-0). 2.3. Sample preparation from soy products Eight kinds of soy supplements indicated to contain soy isoflavones and two soybean products (tofu and boiled soybean) were commercially obtained. The soybean products were lyophilised. Tablet supplements and lyophilised tofu and boiled soybeans were ground into powders using mortar and pestle, and oily supplements were directly measured without pretreatment. Constant amounts (30 mg) of powder were measured, and soy isoflavones were extracted with methanol (1 ml) with sonication for 30 min. The extract was collected in micro test tubes after centrifugation at 1000 rpm for 5 min. This extraction was performed five times, and the obtained extract (5 ml) was dried at 60 °C until dryness. The residues were then dissolved in 80% methanol (3 ml), and diluted with 20% methanol and absolute methanol for the icELISA and HPLC analyses, respectively. The dilution rate, which is fallen within the determination range of DZ and GEN (1.9562.5 ng/ml) was adopted in this study.
S. Sakamoto et al. / Food Chemistry 169 (2015) 127–133
2.4. Synthesis of antigen conjugates DZ-BSA conjugates were synthesised by the periodate oxidation method (Chao et al., 2013). DZ (3.0 mg) in 0.4 ml of dimethyl sulfoxide (DMSO) was added dropwise to 0.4 ml of a solution containing sodium periodate (NaIO4) (3.1 mg); the resulting mixture was stirred at room temperature for 1 h. The reaction mixture (0.8 ml) was then added into 50 mM carbonate buffer solution (pH 9.6, 2 ml) containing BSA (6.0 mg) and stirred at room temperature for 5 h. The reaction mixture was dialysed five times against distilled water at 4 °C and then lyophilised to obtain DZ-BSA conjugates (5.6 mg). DZ-OVA conjugates (5.0 mg) were also synthesised using the OVA (5.0 mg) in the same manner as DZ-BSA conjugates. Determination of hapten number of DZ-BSA conjugates by matrix-assisted laser desorption/ionisation time-of-flight mass spectrometry (MALDI-TOF-MS) The hapten number of the DZ-BSA conjugates was determined by MALDI-TOF-MS according to the method described in the literature (Xuan et al., 2000). A small amount (110 pmol) of an antigen conjugate was mixed with a 103-fold molar excess of sinapinic acid in an aqueous solution containing 0.15% (v/v) trifluoroacetic acid and acetonitrile in the ratio of 2–1. The mixture (2 ll) was spotted onto the stainless steel target, dried at room temperature for 1 h, and subjected to analysis by MALDI-TOF-MS (BRUKER Autoflex III, Bruker Daltonics, Bremen, Germany) using a nitrogen laser (337 nm, 200 Hz maximum firing rate). The spectra were recorded in linear positive high-mass mode with a mass range of 10,000100,000 Da, and were subsequently analysed using the flexControl software (Bruker Daltonics, Bremen, Germany).
2.5. Immunisation, hybridisation, and purification of DZ-MAb Five-week-old BALB/c male mice (KBT Oriental; Saga, Japan) were immunised with the DZ-BSA conjugate, and MAbs were generated as described in previous reported protocols (Paudel et al., 2013; Sakamoto et al., 2008; Sakata, Shoyama, & Murakami, 1994), with some modifications. The first immunisation (50 lg as DZ-BSA conjugates) was administered into the abdominal cavity of the BALB/c mice as a 1:1 emulsion in Freund’s complete adjuvant. The second immunisation (50 lg as DZ-BSA conjugates) was administered as a 1:1 emulsion in Freund’s incomplete adjuvant. On the third day after the final booster (100 lg as DZ-BSA conjugates), the mice were sacrificed by cervical spine fracture dislocation, and the spleens were removed to fuse the splenocytes with mice myeloma cell line, SP2/0, using the polyethylene glycol (PEG) method. They were then cultured in hypoxanthine–aminopterin–thymidine selective medium, and hybridomas producing MAb reactive to DZ were cloned using the limited dilution method. Selected hybridomas were cultured in enriched RPMI 1640-Dulbecco’s-Ham’s F12 (eRDF; Kyokuto Pharmaceutical Industrial Co.; Tokyo, Japan) medium supplemented with RD-1 additives containing 5 lg/ml of insulin, 10 lg/ml of transferrin, 25 lM of ethanolamine, and 25 nM of sodium selenite, where 10% (v/v) of fetal calf serum (FCS; GibcoInvitrogen, Carlsbad, CA, USA) and 1% (v/ v) of Zap-Hybridoma solution (InVitria, Fort Collins, CO, USA) were added to the eRDF medium. The hybridomas were finally scaled up using the same medium without FCS until their volume became 500 ml. DZ-MAb was purified using a Protein G FF column (0.46 11 cm, Pharmacia Biotech; Uppsala, Sweden). The cultured medium (500 ml) containing IgG was adjusted to pH 7.0 with 1 M Tris–HCl solution (pH 9.0) and was applied to the column; the column was then washed with 10 mM phosphate buffer (pH 7.0).
129
Adsorbed IgG was eluted with 100 mM citrate buffer (pH 3.0). The eluted IgG was neutralised with 1 M Tris–HCl (pH 9.0), dialysed three times against distilled water, and lyophilised to give 5.2 mg MAb. 2.6. Direct ELISA and icELISA using DZ-MAb Both direct ELISA and indirect competitive ELISA (icELISA) were performed to investigate the reactivity of the MAbs against DZ-OVA conjugates and free DZ molecules, respectively (Sakamoto, Tanizaki, Pongkitwitoon, Tanaka, & Morimoto, 2011; Sakamoto et al., 2012). In direct ELISA, a 96-well immunoplate (Nunc, Maxisorb, Roskilde, Denmark) was coated with DZ-OVA (1 lg/ml, 100 ll/well) conjugates in 50 mM carbonate buffer (pH 9.6) and was subsequently incubated for 1 h. The plate was then washed three times with phosphate buffer saline (PBS) containing 0.05% (v/v) Tween 20 (PBS-T) before being treated with 300 ll of PBS containing 10% (w/v) skimmed milk (PBS-sm) for 1 h to reduce non-specific adsorption. Subsequently, various concentrations of DZ-MAb (100 ll/well) were incubated for 1 h. The plate was then washed three times with PBS-T and incubated with a 1000-fold diluted solution of peroxidase-labelled anti-mouse IgG (100 ll/ well; Organon Teknika Cappel Products) for 1 h. After the plate was washed three times with PBS-T, 100 ll of 0.3 mg/ml 2, 20 -azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (ABTS) substrate solution in 0.1 M citrate buffer (pH 4.0) supplemented with 0.003% (v/v) H2O2 was added to each well and the wells were incubated for 15 min. In the case of icELISA, the same procedures used in the indirect ELISA were used until the blocking step. After the blocked-plate was washed three times with PBS-T, 50 ll of various concentrations of DZ in 20% (v/v) methanol were incubated with 50 ll of DZ-MAb solution (500 ng/ml) for 1 h. The plate was then washed three times with PBS-T, and the DZ-MAb bound to DZ-OVA was combined with 100 ll of a 1000-fold diluted solution of peroxidase-labelled anti-mouse IgG for 1 h. After the plate was washed three times with PBS-T, 100 ll of ABTS substrate solution was added to each well and the wells were incubated for 15 min. All ELISA incubation steps were performed at 37 °C. Absorbance was measured at 405 nm using a microplate reader (Maltiskan FC, Thermo Scientific). The cross-reactivities (CRs) of the MAbs against various compounds were evaluated and calculated using the method of Weiler and Zenk (1976), as follows:
CRs ð%Þ ¼
IC50 for DZ IC50 for compound under investigation
2.7. Recovery of DZ and GEN from spiked samples Various amounts of DZ and GEN (5, 10, 15, 20, and 30 lg) were individually spiked into dried powders of boiled soybean samples (30 mg). The total amounts of DZ and GEN in the unspiked sample was determined to be 35.67 lg per 30 mg dry wt. by icELISA. After the samples were dried, they were extracted via the same procedure described in Section 2.3, and the total amount of DZ and GEN in each sample were determined by the developed icELISA. The recovery was calculated from the measured total amount of DZ and GEN and the added amounts of DZ or GEN in the same concentration ranges as follows:
Recovery ð%Þ ¼
measured total amount of DZ and GEN 35:67 lg added amount of DZ or GE 100
130
S. Sakamoto et al. / Food Chemistry 169 (2015) 127–133
(BRUKER Autoflex III). A broad peak in the MALDI-TOF-MS spectrum of DZ-BSA appeared at approximately m/z 70,348. Because the molecular weights of BSA and DZ were 66,431 and 416, respectively, at least 9 DZ molecules were conjugated with the BSA. This result reveals that adequate numbers of DZ molecules were coupled with BSA, which means that the DZ-BSA conjugates could be used as an immunogen to initiate anti-DZ antibodies in mice. With respect to the DZ-OVA conjugates that were used as a coated antigen in ELISA, the hapten number was not evaluated because the obtained peak was too broad for the molecular weights of the conjugates to be accurately determined. 3.2. Production and typing of MAbs against DZ
Fig. 2. Reactivity of DZ-MAb to coated antigen, DZ-OVA (1 lg/ml) conjugates by indirect ELISA. In indirect ELISA, various concentrations of DZ-MAb were prepared by serial dilution with PBS-T and applied as primary antibody.
2.8. HPLC analysis The HPLC analysis was performed using a Gilson 805 Manometric Module pump connected to an SPD-20A Shimadzu Prominence UV/VIS detector (254 nm) and an HP ProBook 4230S computer. A COSMOSIL-packed 5C18ARII column (4.6 mm 150 mm, 5 lm particle sizes, Nacalai Tesque, Kyoto, Japan) was used. The mobile phase consisted of 15% acetonitrile. The flow rate was 1.0 ml/ min. Calibration curves for DZ and GEN were constructed in the concentration range of 1.56–50 lg/ml. Analyses of the samples were performed in triplicate. 3. Results and discussion 3.1. Determination of hapten numbers in conjugates by MALDITOF-MS The binding numbers of DZ molecules to the carrier BSA and OVA proteins can be directly determined using MALDI-TOF-MS
The BALB/c mice hyperimmunised by DZ-BSA conjugates, which were used to derive the above-described cell clones, yielded splenocytes that were fused with SP2/0 myeloma cells via the procedure developed by our laboratory (Sakata et al., 1994). One hybridoma cell line producing MAbs reactive to DZ was obtained and was classified as IgG1, which have j light chains, using the IsoQuick Strips and Kits for Mouse Monoclonal Isotyping (Sigma–Aldrich, St. Louis, MO, USA). 3.3. Indirect ELISA and icELISA The reactivity of the DZ-MAb to DZ-OVA conjugate (1 lg/ml) was analysed by indirect ELISA. We constructed the reactivity response curve by plotting absorbance against the logarithm of DZ-MAb concentration in indirect ELISA. The concentration of each DZ-MAb positively correlated with the absorbance value in a logarithmic manner (Fig. 2). These results suggested that the optimal concentration of DZ-MAb for further icELISA was 500 ng/ml. Subsequently, an icELISA was conducted to analyse the inhibitory activity of DZ-MAb against DZ. DZ-MAb (500 ng/ml) was incubated with serially double-diluted concentrations of free DZ on an immunoplate. Any DZ-MAb binding to free DZ was washed out. The DZ-MAb that bound to the immobilised DZ-OVA conjugates were incubated with peroxidase-labelled anti-mouse IgG and were then treated with ABTS solution to develop colour. In this icELISA, the detectable range of DZ concentrations ranged from 1.95 to 62.5 ng/ml (Fig. 3(A)). When their limit of detection of this icELISA
Fig. 3. Calibration curves for the determination of (A) DZ and (B) GEN by icELISA. Concentrations of DZ-MAb and DZ-OVA were 500 ng/ml and 1 lg/ml, respectively. A0 and A corresponds the absorbance in the absence and presence of (A) DZ and (B) GEN, respectively. An inset indicates the enlarged linearised curve for (A) DZ and (B) GEN with the determination range from 1.95 to 62.5 ng/ml.
131
S. Sakamoto et al. / Food Chemistry 169 (2015) 127–133 Table 1 Cross-reactivities (CRs) of the DZ-MAb against naturally occurring compounds. Class
Compound
Cross-reactivity (%)
Isoflavonoids
Daidzin Daidzein Genistin Genistein Glycitin Glycitein Puerarin Miroestrol Deoxymiroestrol Isomiroestrol b-Sitosterol Sennoside A Sennoside B Rhein Rutin Baicalin (+) Catechin Luteolin-7-glucoside Saikosaponin A Ginsenoside Rg1 Swertiamarin Capsaicin Quinine Curcumin Plumbagin Berberine
100.00 16.16 82.35 24.42 11.92 1.02 3.37
Contents lists available at ScienceDirect
Food Chemistry journal homepage: www.elsevier.com/locate/foodchem
Analytical Methods
Simultaneous determination of soy isoflavone glycosides, daidzin and genistin by monoclonal antibody-based highly sensitive indirect competitive enzyme-linked immunosorbent assay Seiichi Sakamoto 1, Gorawit Yusakul 1, Benyakan Pongkitwitoon, Madan Kumar Paudel, Hiroyuki Tanaka ⇑, Satoshi Morimoto Department of Pharmacognosy, Graduate School of Pharmaceutical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan
a r t i c l e
i n f o
Article history: Received 5 May 2014 Received in revised form 30 July 2014 Accepted 1 August 2014 Available online 9 August 2014 Keywords: Daidzin Enzyme-linked immunosorbent assay Genistin Monoclonal antibody Soy isoflavones
a b s t r a c t Soy isoflavones are known as major bioactive compounds in soybean (Glycine max), which is an indispensable food. Despite their utility, the consumption of isoflavones has recently been limited because they exhibit oestrogenic and topoisomerase II inhibitory effects. To assess their intake limitation, accurate, sensitive, and effective quantitative analyses are necessary. In this study, we produced the monoclonal antibody (MAb) against daidzin (DZ) and applied it to an indirect competitive enzyme-linked immunosorbent assay (icELISA) for the simultaneous determination of DZ and genistin (GEN), which are known as two major soy isoflavone glycosides in soy products. Using the DZ-MAb, we developed a sensitive icELISA method, where the limit of detection for DZ and GEN was 1.95 ng/ml. Several validation analyses revealed that the icELISA is sufficiently accurate and sensitive to be used to assess the overconsumption of soy isoflavones, which would lead to the safe dietary intake of soy products. Ó 2014 Published by Elsevier Ltd.
1. Introduction Soy isoflavones are major compounds isolated from soybean (Glycine max), which is an indispensable food material. Soy isoflavones are mainly composed of three glycosides (daidzin (DZ), genistin (GEN), and glycitin) and their aglycones (daidzein, genistein, and glycitein) (Fig. 1). They have recently been the subject of much attention because of their ability to decrease the morbidity rates related to age-related cardiovascular and osteoporosis, hormonedependent breast and prostate cancers, and menopausal symptoms (Cano, García-Pérez, & Tarín, 2010; Mourouti & Panagiotakos, 2013; Park & Weaver, 2012). These effects are mainly attributed to the structural resemblance between isoflavones and estradiol. Contrary to these beneficial effects, adverse effects by overconsumption have also been proposed (Setchell & Cassidy, 1999). In addition, one of the soy isoflavones, genistein, has been reported to induce a DNA-topoisomerase II inhibitory effect by stabilising the covalent DNA cleavage complex, which is an event predicted to cause DNA damage via the activation of p53 and checkpoint kinase 2 (Ye et al., 2001) by the same mechanism as that of topoisomerase II inhibitor, etoposide (Lu, Taya, Ikeda, & Levine, 1998). ⇑ Corresponding author. Tel./fax: +81 92 642 6581. 1
E-mail address: [email protected] (H. Tanaka). These authors contributed equally to this work.
http://dx.doi.org/10.1016/j.foodchem.2014.08.004 0308-8146/Ó 2014 Published by Elsevier Ltd.
Therefore, the relationships between the consumption of dietary isoflavones and infant leukaemia have been clinically studied (Azarova et al., 2010). In Japan, opportunities to intake soy isoflavones are abundant because soy products have been traditionally and generally used in Japanese foods including tofu (soybean curd), kinako (soybean flour), shoyu (soy sauce), miso (soybean paste), and natto (fermented soybeans). In addition, prescription Japanese Kampo medicines mainly derived from kudzu (Pueraria lobata Ohwi) also contain substantial amounts of isoflavones (Boué et al., 2003). As such, the intake of isoflavones has been limited to 75 mg/day (30 mg/day as supplements) by the Food Safety Commission of Japanese Government since 2006 (Food Safety Commission Novel Foods Expert Committee, 2006). In addition to the Japanese case, The Nutrient Data Laboratory of the Agricultural Research Service of the United States Department of Agriculture, in collaboration with Iowa State University, released a Special Interest Database on isoflavones in foods in 1999 and updated it several times until 2008 by increasing the number of soy products (U.S. Department of Agriculture, 2008). In both cases, intake amounts of isoflavones were evaluated on the basis of aglycone forms because the isoflavone glucosides are converted into aglycones by intestinal flora in vivo (Tsuchihashi, Okawa, Nohara, Okabe, & Kinjo, 2004; Yasuda, Mizunuma, Kano, Saito, & Ohsawa, 1996; Yasuda, Ueda, & Ohsawa, 2001) and in vitro (Tsuchihashi, Sakamoto, Kodera, Nohara, & Kinjo, 2008). Therefore, an analytical
128
S. Sakamoto et al. / Food Chemistry 169 (2015) 127–133
(A)
(D)
(B)
(E)
(C)
(F)
3 3
Fig. 1. Structures of major soy isoflavones (A) daidzin (DZ), (B) genistin (GEN), and (C) glycitin and their corresponding aglycones (D) daidzein, (E) genistein, and (F) glycitein.
method to determine isoflavone aglycones has been mainly developed using ultra-high-pressure liquid chromatography (UHPLC) (Fiechter, Opacak, Raba, & Mayer, 2013), HPLC separation (Zeng et al., 2012) coupled with diode array detection (Wang et al., 2013), flow injection chemiluminescence (Wang & Han, 2012), and enzyme-linked immunosorbent assays (ELISAs) (BennetauPelissero et al., 2003; Pongkitwitoon et al., 2011). However, accurate, sensitive, and effective qualitative methods are necessary for the determination of isoflavone glycosides, especially in the case of DZ and GEN because they are two major isoflavone glycosides in soy products before fermentation and in medicinal plants (Wang & Murphy, 1994). To date, few studies have addressed the quantitative analysis of isoflavone glycosides by methods other than HPLC (Hosoda et al., 2008), HPLC/UV-MS (Klejdus et al., 2004), or UHPLC/UV (Toro-Funes et al., 2012). In this study, monoclonal antibody (MAb) against the simplest structure among soy isoflavone glycosides, DZ (DZ-MAb), was produced by several injections of DZ-bovine serum albumin (BSA) conjugates into BALB/c mice, followed by a hybridoma method. Characterisation of DZ-MAb revealed that it exhibits cross-reactivity against DZ and GEN, thereby enabling the development of an effective indirect competitive ELISA (icELISA) for the simultaneous determination of both glycosides of DZ and GEN. Their limit of determination in this system is 1.95 ng/ml. Several validation analyses revealed that the icELISA is sufficiently accurate and sensitive to be used for the determination of soy isoflavones contained in the samples. The preparation of DZ-MAb and its application to the icELISA system are demonstrated in this paper.
2. Materials and methods 2.1. Chemicals and reagents Daidzin (DZ, P99%), daidzein (P97%), genistin (GEN, P99%), and genistein (P99%) were purchased from Fujicco Co. (Kobe, Japan). Bovine serum albumin (BSA, P97%) and albumin from
chicken egg (OVA, P98%) were acquired from Sigma–Aldrich (St. Louis, MO, USA). Freund’s complete and incomplete adjuvants were obtained from Difco (Detroit, MI, USA). Peroxidase-labelled anti-mouse IgG was purchased from Organon Teknika Cappel Products (West Chester, PA, USA). All other chemicals were standard commercial products of analytical-reagent grade. 2.2. Animals Five-week-old male BALB/c mice were purchased from KBT Oriental Co. (Saga, Japan). A standard diet (MF; Oriental Yeast Co., Tokyo, Japan) and water were provided ad libitum. All of the procedures and care related to the animals were approved by the Committee on the Ethics of Animal Experiments, Graduate School of Pharmaceutical Sciences, Kyushu University, and were conducted according to the Guidelines for Animal Experiments of the Graduate School of Pharmaceutical Sciences, Kyushu University (approval no: A24-004-0). 2.3. Sample preparation from soy products Eight kinds of soy supplements indicated to contain soy isoflavones and two soybean products (tofu and boiled soybean) were commercially obtained. The soybean products were lyophilised. Tablet supplements and lyophilised tofu and boiled soybeans were ground into powders using mortar and pestle, and oily supplements were directly measured without pretreatment. Constant amounts (30 mg) of powder were measured, and soy isoflavones were extracted with methanol (1 ml) with sonication for 30 min. The extract was collected in micro test tubes after centrifugation at 1000 rpm for 5 min. This extraction was performed five times, and the obtained extract (5 ml) was dried at 60 °C until dryness. The residues were then dissolved in 80% methanol (3 ml), and diluted with 20% methanol and absolute methanol for the icELISA and HPLC analyses, respectively. The dilution rate, which is fallen within the determination range of DZ and GEN (1.9562.5 ng/ml) was adopted in this study.
S. Sakamoto et al. / Food Chemistry 169 (2015) 127–133
2.4. Synthesis of antigen conjugates DZ-BSA conjugates were synthesised by the periodate oxidation method (Chao et al., 2013). DZ (3.0 mg) in 0.4 ml of dimethyl sulfoxide (DMSO) was added dropwise to 0.4 ml of a solution containing sodium periodate (NaIO4) (3.1 mg); the resulting mixture was stirred at room temperature for 1 h. The reaction mixture (0.8 ml) was then added into 50 mM carbonate buffer solution (pH 9.6, 2 ml) containing BSA (6.0 mg) and stirred at room temperature for 5 h. The reaction mixture was dialysed five times against distilled water at 4 °C and then lyophilised to obtain DZ-BSA conjugates (5.6 mg). DZ-OVA conjugates (5.0 mg) were also synthesised using the OVA (5.0 mg) in the same manner as DZ-BSA conjugates. Determination of hapten number of DZ-BSA conjugates by matrix-assisted laser desorption/ionisation time-of-flight mass spectrometry (MALDI-TOF-MS) The hapten number of the DZ-BSA conjugates was determined by MALDI-TOF-MS according to the method described in the literature (Xuan et al., 2000). A small amount (110 pmol) of an antigen conjugate was mixed with a 103-fold molar excess of sinapinic acid in an aqueous solution containing 0.15% (v/v) trifluoroacetic acid and acetonitrile in the ratio of 2–1. The mixture (2 ll) was spotted onto the stainless steel target, dried at room temperature for 1 h, and subjected to analysis by MALDI-TOF-MS (BRUKER Autoflex III, Bruker Daltonics, Bremen, Germany) using a nitrogen laser (337 nm, 200 Hz maximum firing rate). The spectra were recorded in linear positive high-mass mode with a mass range of 10,000100,000 Da, and were subsequently analysed using the flexControl software (Bruker Daltonics, Bremen, Germany).
2.5. Immunisation, hybridisation, and purification of DZ-MAb Five-week-old BALB/c male mice (KBT Oriental; Saga, Japan) were immunised with the DZ-BSA conjugate, and MAbs were generated as described in previous reported protocols (Paudel et al., 2013; Sakamoto et al., 2008; Sakata, Shoyama, & Murakami, 1994), with some modifications. The first immunisation (50 lg as DZ-BSA conjugates) was administered into the abdominal cavity of the BALB/c mice as a 1:1 emulsion in Freund’s complete adjuvant. The second immunisation (50 lg as DZ-BSA conjugates) was administered as a 1:1 emulsion in Freund’s incomplete adjuvant. On the third day after the final booster (100 lg as DZ-BSA conjugates), the mice were sacrificed by cervical spine fracture dislocation, and the spleens were removed to fuse the splenocytes with mice myeloma cell line, SP2/0, using the polyethylene glycol (PEG) method. They were then cultured in hypoxanthine–aminopterin–thymidine selective medium, and hybridomas producing MAb reactive to DZ were cloned using the limited dilution method. Selected hybridomas were cultured in enriched RPMI 1640-Dulbecco’s-Ham’s F12 (eRDF; Kyokuto Pharmaceutical Industrial Co.; Tokyo, Japan) medium supplemented with RD-1 additives containing 5 lg/ml of insulin, 10 lg/ml of transferrin, 25 lM of ethanolamine, and 25 nM of sodium selenite, where 10% (v/v) of fetal calf serum (FCS; GibcoInvitrogen, Carlsbad, CA, USA) and 1% (v/ v) of Zap-Hybridoma solution (InVitria, Fort Collins, CO, USA) were added to the eRDF medium. The hybridomas were finally scaled up using the same medium without FCS until their volume became 500 ml. DZ-MAb was purified using a Protein G FF column (0.46 11 cm, Pharmacia Biotech; Uppsala, Sweden). The cultured medium (500 ml) containing IgG was adjusted to pH 7.0 with 1 M Tris–HCl solution (pH 9.0) and was applied to the column; the column was then washed with 10 mM phosphate buffer (pH 7.0).
129
Adsorbed IgG was eluted with 100 mM citrate buffer (pH 3.0). The eluted IgG was neutralised with 1 M Tris–HCl (pH 9.0), dialysed three times against distilled water, and lyophilised to give 5.2 mg MAb. 2.6. Direct ELISA and icELISA using DZ-MAb Both direct ELISA and indirect competitive ELISA (icELISA) were performed to investigate the reactivity of the MAbs against DZ-OVA conjugates and free DZ molecules, respectively (Sakamoto, Tanizaki, Pongkitwitoon, Tanaka, & Morimoto, 2011; Sakamoto et al., 2012). In direct ELISA, a 96-well immunoplate (Nunc, Maxisorb, Roskilde, Denmark) was coated with DZ-OVA (1 lg/ml, 100 ll/well) conjugates in 50 mM carbonate buffer (pH 9.6) and was subsequently incubated for 1 h. The plate was then washed three times with phosphate buffer saline (PBS) containing 0.05% (v/v) Tween 20 (PBS-T) before being treated with 300 ll of PBS containing 10% (w/v) skimmed milk (PBS-sm) for 1 h to reduce non-specific adsorption. Subsequently, various concentrations of DZ-MAb (100 ll/well) were incubated for 1 h. The plate was then washed three times with PBS-T and incubated with a 1000-fold diluted solution of peroxidase-labelled anti-mouse IgG (100 ll/ well; Organon Teknika Cappel Products) for 1 h. After the plate was washed three times with PBS-T, 100 ll of 0.3 mg/ml 2, 20 -azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (ABTS) substrate solution in 0.1 M citrate buffer (pH 4.0) supplemented with 0.003% (v/v) H2O2 was added to each well and the wells were incubated for 15 min. In the case of icELISA, the same procedures used in the indirect ELISA were used until the blocking step. After the blocked-plate was washed three times with PBS-T, 50 ll of various concentrations of DZ in 20% (v/v) methanol were incubated with 50 ll of DZ-MAb solution (500 ng/ml) for 1 h. The plate was then washed three times with PBS-T, and the DZ-MAb bound to DZ-OVA was combined with 100 ll of a 1000-fold diluted solution of peroxidase-labelled anti-mouse IgG for 1 h. After the plate was washed three times with PBS-T, 100 ll of ABTS substrate solution was added to each well and the wells were incubated for 15 min. All ELISA incubation steps were performed at 37 °C. Absorbance was measured at 405 nm using a microplate reader (Maltiskan FC, Thermo Scientific). The cross-reactivities (CRs) of the MAbs against various compounds were evaluated and calculated using the method of Weiler and Zenk (1976), as follows:
CRs ð%Þ ¼
IC50 for DZ IC50 for compound under investigation
2.7. Recovery of DZ and GEN from spiked samples Various amounts of DZ and GEN (5, 10, 15, 20, and 30 lg) were individually spiked into dried powders of boiled soybean samples (30 mg). The total amounts of DZ and GEN in the unspiked sample was determined to be 35.67 lg per 30 mg dry wt. by icELISA. After the samples were dried, they were extracted via the same procedure described in Section 2.3, and the total amount of DZ and GEN in each sample were determined by the developed icELISA. The recovery was calculated from the measured total amount of DZ and GEN and the added amounts of DZ or GEN in the same concentration ranges as follows:
Recovery ð%Þ ¼
measured total amount of DZ and GEN 35:67 lg added amount of DZ or GE 100
130
S. Sakamoto et al. / Food Chemistry 169 (2015) 127–133
(BRUKER Autoflex III). A broad peak in the MALDI-TOF-MS spectrum of DZ-BSA appeared at approximately m/z 70,348. Because the molecular weights of BSA and DZ were 66,431 and 416, respectively, at least 9 DZ molecules were conjugated with the BSA. This result reveals that adequate numbers of DZ molecules were coupled with BSA, which means that the DZ-BSA conjugates could be used as an immunogen to initiate anti-DZ antibodies in mice. With respect to the DZ-OVA conjugates that were used as a coated antigen in ELISA, the hapten number was not evaluated because the obtained peak was too broad for the molecular weights of the conjugates to be accurately determined. 3.2. Production and typing of MAbs against DZ
Fig. 2. Reactivity of DZ-MAb to coated antigen, DZ-OVA (1 lg/ml) conjugates by indirect ELISA. In indirect ELISA, various concentrations of DZ-MAb were prepared by serial dilution with PBS-T and applied as primary antibody.
2.8. HPLC analysis The HPLC analysis was performed using a Gilson 805 Manometric Module pump connected to an SPD-20A Shimadzu Prominence UV/VIS detector (254 nm) and an HP ProBook 4230S computer. A COSMOSIL-packed 5C18ARII column (4.6 mm 150 mm, 5 lm particle sizes, Nacalai Tesque, Kyoto, Japan) was used. The mobile phase consisted of 15% acetonitrile. The flow rate was 1.0 ml/ min. Calibration curves for DZ and GEN were constructed in the concentration range of 1.56–50 lg/ml. Analyses of the samples were performed in triplicate. 3. Results and discussion 3.1. Determination of hapten numbers in conjugates by MALDITOF-MS The binding numbers of DZ molecules to the carrier BSA and OVA proteins can be directly determined using MALDI-TOF-MS
The BALB/c mice hyperimmunised by DZ-BSA conjugates, which were used to derive the above-described cell clones, yielded splenocytes that were fused with SP2/0 myeloma cells via the procedure developed by our laboratory (Sakata et al., 1994). One hybridoma cell line producing MAbs reactive to DZ was obtained and was classified as IgG1, which have j light chains, using the IsoQuick Strips and Kits for Mouse Monoclonal Isotyping (Sigma–Aldrich, St. Louis, MO, USA). 3.3. Indirect ELISA and icELISA The reactivity of the DZ-MAb to DZ-OVA conjugate (1 lg/ml) was analysed by indirect ELISA. We constructed the reactivity response curve by plotting absorbance against the logarithm of DZ-MAb concentration in indirect ELISA. The concentration of each DZ-MAb positively correlated with the absorbance value in a logarithmic manner (Fig. 2). These results suggested that the optimal concentration of DZ-MAb for further icELISA was 500 ng/ml. Subsequently, an icELISA was conducted to analyse the inhibitory activity of DZ-MAb against DZ. DZ-MAb (500 ng/ml) was incubated with serially double-diluted concentrations of free DZ on an immunoplate. Any DZ-MAb binding to free DZ was washed out. The DZ-MAb that bound to the immobilised DZ-OVA conjugates were incubated with peroxidase-labelled anti-mouse IgG and were then treated with ABTS solution to develop colour. In this icELISA, the detectable range of DZ concentrations ranged from 1.95 to 62.5 ng/ml (Fig. 3(A)). When their limit of detection of this icELISA
Fig. 3. Calibration curves for the determination of (A) DZ and (B) GEN by icELISA. Concentrations of DZ-MAb and DZ-OVA were 500 ng/ml and 1 lg/ml, respectively. A0 and A corresponds the absorbance in the absence and presence of (A) DZ and (B) GEN, respectively. An inset indicates the enlarged linearised curve for (A) DZ and (B) GEN with the determination range from 1.95 to 62.5 ng/ml.
131
S. Sakamoto et al. / Food Chemistry 169 (2015) 127–133 Table 1 Cross-reactivities (CRs) of the DZ-MAb against naturally occurring compounds. Class
Compound
Cross-reactivity (%)
Isoflavonoids
Daidzin Daidzein Genistin Genistein Glycitin Glycitein Puerarin Miroestrol Deoxymiroestrol Isomiroestrol b-Sitosterol Sennoside A Sennoside B Rhein Rutin Baicalin (+) Catechin Luteolin-7-glucoside Saikosaponin A Ginsenoside Rg1 Swertiamarin Capsaicin Quinine Curcumin Plumbagin Berberine
100.00 16.16 82.35 24.42 11.92 1.02 3.37
