Food Chemistry 146 (2014) 270277

Contents lists available at ScienceDirect

Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem

Analytical Methods

Rapid characterisation and comparison of saponin proles in the seeds of Korean Leguminous species using ultra performance liquid chromatography with photodiode array detector and electrospray ionisation/mass spectrometry (UPLCPDAESI/MS) analysis
Tae Joung Ha a, Byong Won Lee a, Ki Hun Park d, Seong Hun Jeong e, Hyun-Tae Kim a, Jong-Min Ko a, In-Youl Baek a, Jin Hwan Lee b,c,
a

Department of Functional Crop, National Institute of Crop Science (NICS), Rural Development Administration, Miryang 627-803, Republic of Korea National Institute of Biological Resources, Ministry of Environment, Incheon 404-708, Republic of Korea c Department of Monitoring and Analysis, NAKDONG River Basin Environmental Ofce, Ministry of Environment, Changwon 641-722, Republic of Korea d Division of Applied Life Science (BK21 plus), IALS, GyeongSang National University, Jinju 660-751, Republic of Korea e Namhae Garlic Research Institute, Namhae, 668-812, Republic of Korea
b

a r t i c l e

i n f o

a b s t r a c t
The present work was reported on investigation of saponin proles in nine different legume seeds, including soybean, adzuki bean, cowpea, common bean, scarlet runner bean, lentil, chick pea, hyacinth bean, and broad bean using ultra performance liquid chromatography with photodiode array detector and electrospray ionisation/mass spectrometry (UPLCPDAESI/MS) technique. A total of twenty saponins were characterised under rapid and simple conditions within 15 min by the 80% methanol extracts of all species. Their chemical structures were elucidated as soyasaponin Ab (1), soyasaponin Ba (2), soyasaponin Bb (3), soyasaponin Bc (4), soyasaponin Bd (5), soyasaponin ag (6), soyasaponin bg (7), soyasaponin ba (8), soyasaponin cg (9), soyasaponin ca (10), azukisaponin VI (11), azukisaponin IV (12), azukisaponin II (13), AzII (14), AzIV (15), lablaboside E (16), lablaboside F (17), lablaboside D (18), chikusetusaponin IVa (19), and lablab saponin I (20). The individual and total saponin compositions exhibited remarkable differences in all legume seeds. In particular, soyasaponin ba (8) was detected the predominant composition in soybean, cowpea, and lentil with various concentrations. Interestingly, soybean, adzuki bean, common bean, and scarlet runner bean had high saponin contents, while chick pea and broad bean showed low contents. 2013 Elsevier Ltd. All rights reserved.

Article history: Received 29 January 2013 Received in revised form 11 August 2013 Accepted 8 September 2013 Available online 18 September 2013 Keywords: Legumes Seeds Saponin UPLCPDAESI/MS Soyasaponin bg Methanol extract

1. Introduction Legumes have been widely used in commercial and agricultural crops for decades from many countries (Iqbal, Khalil, Ateeq, & Khan, 2006). These species are excellent sources of many nutrients such as tocopherols, proteins, minerals, dietary bers, carbohydrates, vitamins, and phytochemicals (Almeida Costa, QueirozMonici, Reis, & Oliveira, 2006; Amarowicz & Pegg, 2008; Boschin & Arnoldi, 2011). In recent years, legumes are increasingly of interest to food and medicinal industries due to benecial properties for human health. For instance, European countries have the highest consumption, between 8 and 23 g/capita per day (Aguilera et al., 2009) and the consumption of legumes in USA has signicantly increased because of functional and nutraceutical foods (Luthria & Pastor-Corrales, 2006). In addition, legumes are considered to be one of the most popular food sources in Korea for many years.
Corresponding author. Tel.: +82 55 211 1684; fax: +82 55 211 1608.
E-mail addresses: hwanletter08@hanmail.net, schem72@korea.kr (J.H. Lee). 0308-8146/$ - see front matter 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.foodchem.2013.09.051

The health benecial activities of this species are attributed to the presence of various phytochemicals, namely, avonoids, phenolic acids, tannins, triterpenic acid, and saponins (Amarowicz & Pegg, 2008; Kalogeropoulos et al., 2010; Shahidi & Naczk, 2004). For these reasons, many researchers have mainly focused on the secondary metabolites and their pharmacological effects from various legume species. Moreover, food and medicinal industries are increasingly interested in Leguminous crops with high phytochemical contents for the manufacture of supplements with therapeutic and preventive properties. In several phytochemicals, saponins are a diverse group of secondary metabolites with a structure constituting of steroid or triterpenoid aglycones and sugar moieties (Decroos et al., 2005; Hu, Lee, Hendrich, & Murphy, 2002; Kinjo et al., 1998). These compounds are widely distributed in crops, vegetables, fruits, and edible plants, and known to exhibit health benecial effects on various illnesses. Numerous studies showed that saponins are of great interest in pharmaceutical aspects due to their biological effects, such as anticarcinogenic, anticancer, antifungal, anti-inammatory, and antimicrobial

T.J. Ha et al. / Food Chemistry 146 (2014) 270277

271

activities (Gurnkel & Rao, 2003; Konoshima, 1996; Nagata et al., 1985; Sparg, Light, & van Staden, 2004). Generally, it is well documented that analytical methods of saponins have been described in various food sources using HPLCUV, HPLC/MS, HPLCELSD, ELISA, and HPTLC (Applebaum, Marco, & Birk, 1969; Chang, Han, Joh, & Kim, 2010; Guajardo-Flores, Garca-Patio, Serna-Guerrero, Gutirrez-Uribe, & Serna-Saldvar, 2012; Hu et al., 2002; Mbagwu, Okafor, & Ekeanyanwu, 2011; Price, Curl, & Fenwick, 1986). Unfortunately, these procedures have some inconvenience, namely, time-consuming, expensive apparatus, lack of reproducibility, and relative complicate. Also, there are few published researches on the simultaneous determination of the saponin compositions due to the low contents. Interestingly, the analytical time of most of the previous works was more than 30 min on the determination of saponins (Decroos et al., 2005; Hu et al., 2002; Kinjo et al., 1998). Although several scientists have focused on various foods and crops, including soybean, alfalfa, and ginseng (Chang et al., 2010; Guajardo-Flores et al., 2012; Lai et al., 2006), there is very limited information available on the saponin compositions of various Leguminous crops. Furthermore, little intelligence has been reported on the determination and comparison of saponin proles on the basis of rapid and simple analysis. Up to data, UPLC analysis has the advantage of the high chromatographic peak capacity, fast analysis, and good sensitivity as well as resolution in comparison with other isolation methods (Guan, Lai, & Li, 2007). In addition, the UPLC coupled with mass spectrometry performs adequate structural information of multiple compounds (Han et al., 2011). Based on the above considerations, our work was designed to get available information regarding saponins in legume seeds using UPLCPDAESI/MS analysis. The purpose of the current research was the characterisation of saponin proles in legume seeds. Moreover, this study was evaluated the comparison of saponin compositions with the application of the rapid and simple analytical method through UPLC coupled with mass technique from various Korean legume species. 2. Materials and methods 2.1. Plant materials and chemicals Nine legume species, soybean (Glycine max (L.) Merr, Leguminosae), adzuki bean (Vigna angularis W. F. WIGHT, Leguminosae), cowpea (Vigna unguiculata (L.) Walp, Leguminosae), common bean (Phaseolus vulgaris (L.) Leguminosae), scarlet runner bean (Phaseolus coccineus, Leguminosae), lentil (Lens culinaris Medic, Leguminosae), chick pea (Cicer arietimum, Leguminosae), hyacinth bean (Lablab purpureus, Leguminosae), and broad bean (Vicia fava, Leguminosae) were grown in the experimental eld of the National Institute of Crop Science (NICS), Rural Development Administration (RDA), Milyang, Gyeongnam, during 2010. After harvesting, legume seeds were washed with sterile water, and then air-dried for 3 days at 25 C. The dried seeds were immediately freeze-dried and stored at 40 C until analysis. Analytical grade methanol, water, and acetonitrile were purchased from J.T. Baker (Phillipsburg, NJ, USA). 2.2. Extraction of sample and UPLC conditions The dried seeds of each legume were nely grounded for 3 min using a HR 2860 coffee grinder (Philips, Drachten, Netherlands). The pulverised seeds (60 mesh) were extracted with 10 ml of the 80% methanol at 25 C for 24 h. The extracts of all samples were centrifuged at 3000g for 3 min, and ltered through a 0.45 lm syringe lter (Whtatman Inc., Maidstone, UK). Saponins in legume

seeds were characterised by the Acquity UPLC system coupled on-line with a TQ (triple quadrupole) mass spectrometer detector (Waters, MA, USA) with an acquity UPLC HSS T3 column (2.1 100 mm, 1.8 lm) at a ow rate of 0.4 ml/min. The mobile phase consisted of 0.1% formic acid in water (solvent A) and 0.1% formic acid in acetonitrile (solvent B). The gradient elution program was as follows: 01 min, 5% B; 2 min, 10% B; 10 min, 30% B; 12 min, 40% B; 15 min, 50% B; 20 min, 70% B; held at 100% B for 3 min, and then held for 3 min before returning to the initial conditions. The injection volume was 10 ll and the detection was recorded at 254 nm. The column temperature was set at 30 C and the total running time was 20 min. 2.3. Mass spectrometry conditions Mass spectra were scanned between m/z 400 and m/z 2000 in the ionisation mode (ESI) at a scan rate of 1.5 s/cycle and were monitored by a photo diode array (PDA) detector. The mass source temperature was maintained at 150 C, and the desolvation temperature was set at 400 C. The other mass spectrometric conditions were programmed as follows: capillary voltage 3.0 kV; cone voltage 60 V; collision energy 40 eV; desolvation gas ow of 850 l/h (N2); cone gas ow of 50 l/h (N2). Collision-induced fragmentation experiments were measured in the ion trap using helium as the collision gas. 3. Results and discussion It is well demonstrated that solvents contribute to the highest efciency of the extraction. Solvents with high polarity do afford good extraction results in comparison with those of low polarity (Liu, Ang, & Springer, 2000). In previously published data, ethanol and methanol were found to be the appropriate solvents regarding saponin (Guajardo-Flores et al., 2012; Iida et al., 1999). Moreover, the combination of ethanol and methanol with water offers a grateful effect for saponin extraction (Guajardo-Flores et al., 2012; Iida et al., 1999). In particular, the extraction conditions including 5075% ethanol at high temperature or reux and 50 80% methanol at room temperature have been widely used for saponins (Guajardo-Flores et al., 2012; Han et al., 2011). Although extraction temperature may cause an increase in the rate of phytochemical extraction, their contents tended to decrease due to the degradation of the molecule by extraction through high temperature (Harbourne, Marete, Christophe, Jacquier, & ORiordan, 2013). Thus, we have chosen methanol solvent for saponin extraction at room temperature from various legume seeds. Also, the extraction solvent for saponins in the present work was measured by 80% methanol as supported by published data (Guajardo-Flores et al., 2012; Hu et al., 2002; Kinjo et al., 1998). It is well-known that the mass spectrometry is widely used as one of the most important analytical methods for determination of molecular weights regarding phytochemicals (Lee & Cho, 2012; Li et al., 2012). Specically, the UPLC system coupled with mass spectrometry shows an excellent method for simultaneously identifying complicated compounds in edible sources due to the product ions produced from the fragmentation of a selected precursor ion (Dartora et al., 2011). Based on the above reasons, saponins in the seeds of nine different legume species were tentatively determined by UPLCPDAESI/MS analysis. As presented in Fig. 1, a complete chromatographic separation of various saponins including major and minor peaks reached within 15 min at a wavelength of 254 nm. In other words, this method exhibited superior sensitivity and resolution as well as rapid separation. Table 1 shows retention times, molecular ions, and elementary compositions of the identied saponins by comparison with those

272

F
T.J. Ha et al. / Food Chemistry 146 (2014) 270277

Fig. 1. Typical UPLCPDA chromatograms of saponins in the 80% methanol extracts of nine legume seeds detected at 254 nm. (A) soybean (G. max), (B) adzuki bean (V. angularis), (C) cowpea (V. unguiculata), (D) common bean (P. vulgaris), (E) scarlet runner bean (P. coccineus), (F) lentil (L. culinaris), (G) chick pea (C. arietimum), (H) hyacinth bean (L. purpureus) and (I) broad bean (V. fava).

T.J. Ha et al. / Food Chemistry 146 (2014) 270277

273

Table 1 Peak assignments, retention times, mass spectral data, elemental compositions, and cited references of saponins observed in the seeds of nine Korean Leguminous crops using UPLCPDAESI/MS analysis. Peak 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 tR (min) at 254 nm 5.7 6.5 9.4 10.1 11.2 12.1 12.9 13.6 14.3 14.7 6.2 6.3 10.7 11.8 12.4 1.5 1.9 2.6 7.1 7.6 Molecular ion [M]+ (m/z) 1436 958 942 912 956 1084 1068 1038 922 892 1134 972 796 1098 1084 1410 1394 1262 794 1082 Elemental composition C67H104O33 C48H78O19 C48H78O18 C47H76O17 C48H76O19 C54H84O22 C54H84O21 C53H82O20 C48H74O17 C47H72O16 C54H86O25 C48H76O20 C42H68O14 C54H82O23 C54H84O22 C66H106O32 C66H106O31 C66H94O28 C42H66O14 C54H82O22 Identity Soyasaponin Ab Soyasaponin Ba (V) Soyasaponin Bb (I) Soyasaponin Bc (II) Soyasaponin Bd Soyasaponin ag Soyasaponin bg Soyasaponin ba Soyasaponin cg Soyasaponin ca Azukisaponin VI Azukisaponin IV Azukisaponin II AzII AzIV Lablaboside E Lablaboside F Lablaboside D Chikusetsusaponin IVa Lablab saponin I References Chang et al. (2010) and Decroos et al. (2005) Decroos et al. (2005) Decroos et al. (2005) Decroos et al. (2005) Decroos et al. (2005) Decroos et al. (2005) and Hu et al. (2002) Decroos et al. (2005) and Hu et al. (2002) Decroos et al. (2005) and Hu et al. (2002) Decroos et al. (2005) Decroos et al. (2005) Kinjo et al. (1998) Kitagawa et al. (1983) Kinjo et al. (1998) and Kitagawa et al. (1983) Iida et al. (1999) Iida et al. (1999) Komatsu et al. (1998) Komatsu et al. (1998) Komatsu et al. (1998) Nie et al. (1984) Yoshiki et al. (1995)

of previous works. Firstly, we identied ten saponin peaks from soybean (Fig. 1A). The abundant saponins (peaks 7: tR = 12.9 min and 8: tR = 13.6 min) were detected with identical molecular ions [M]+ at m/z 1068 and m/z 1038 (Fig. 2G and H), representing

approximately 90% of the total peak area. These peaks were conrmed with soyasaponin bg (7) and soyasaponin ba (8) by comparison of the data reported in previous studies (Decroos et al., 2005; Hu et al., 2002). The MS analyses of other peaks possessed identical

Fig. 2. Mass fragmentation patterns of identied saponins. (A) Soyasaponin Ab, (B) soyasaponin Ba, (C) soyasaponin Bb, (D) soyasaponin Bc, (E) soyasaponin Bd, (F) soyasaponin ag, (G) soyasaponin bg, (H) soyasaponin ba, (I) soyasaponin cg, (J) soyasaponin ca, (K) azukisaponin VI, (L) azukisaponin IV, (M) azukisaponin II, (N) AzII, (O) AzIV, (P) lablaboside E, (Q) lablaboside F, (R) lablaboside D, (S) chikusetusaponin Iva and (T) lablab saponin I.

274

T.J. Ha et al. / Food Chemistry 146 (2014) 270277

Fig. 2 (continued)

molecular ions [M]+ at m/z 1436 (1), m/z 958 (2), m/z 942 (3), m/z 912 (4), m/z 956 (5), m/z 1084 (6), m/z 992 (9), and m/z 892 (10) (Fig. 2AF, I, and J). Additionally, their retention times exhibited 5.7, 6.5, 9.4, 10.1, 11.2, 12.1, 14.3, and 14.7 min (Table 1). On the basis of the above results and the earlier literatures, their chemical structures were tentatively assigned as soyasaponin Ab (1), soyasaponin Ba (2), soyasaponin Bb (3), soyasaponin Bc (4), soyasaponin Bd (5), soyasaponin ag (6), soyasaponin cg (9), and soyasaponin ca (10) (Fig. 3) (Chang et al., 2010; Decroos et al., 2005; Hu et al., 2002). Although the distribution of saponins showed slight differences, our results are similar to those of previous studies determining the saponin compositions in soybeans and soy food products (Guajardo-Flores et al., 2012; Hubert, Berger, & Dayd, 2005). Secondly, we investigated the saponin prole from the 80% methanol extract of adzuki bean using UPLCPDAESI/MS technique. In the earlier report, only two saponins, namely azukisaponin VI (11) and azukisaponin II (13) were identied by the HPLC analysis from the reuxed methanol extract at 205 nm (Kinjo et al., 1998). However, our work observed many peaks in the UPLCPDA chromatogram as shown in Fig. 1B. We demonstrated ten saponins according to the molecular weights of mass spectral data with previously reported studies (Decroos et al., 2005; Kinjo et al., 1998; Kitagawa, Wang, Saito, & Yoshikawa, 1983). Even though the most abundant saponin (tR = 14.4 min) was not conrmed by this method, the second major peak (tR = 14.3 min) was assigned as soyasaponin cg (9) with the molecular ion [M]+ at m/z 922 (Fig. 2I), representing approximately 42% of the total

peak area (Decroos et al., 2005). In particular, the ESI/MS spectra of four peaks 3, 6, 7, and 9 (tR = 9.4, 12.1, 12.8, and 14.3 min) exhibited the molecular ions [M]+ at m/z 942, m/z 1084, m/z 1068, and m/z 922 (Fig. 2C, F, G, and I), which were same to the patterns of saponins 3, 6, 7, and 9 in soybean chromatogram. Therefore, we revealed that soyasaponin Bb (3), soyasaponin ag (6), soyasaponin bg (7), and soyasaponin cg (9) were present in the seeds of adzuki bean. The remaining saponins were observed with identical molecular ions [M]+ at m/z 1134, 972, 796, 1098, 1084, and 1262 (Fig. 2KO, and R) and their structures were conrmed with azukisaponin VI (11) (tR = 6.2 min), azukisaponin IV (12) (tR = 6.3 min), azukisaponin II (13) (tR = 10.7 min), AzII (14) (tR = 11.8 min), AzIV (15) (tR = 12.4 min), and lablaboside D (18) (tR = 2.6 min) by comparison with published data (Iida et al., 1999; Kinjo et al., 1998; Kitagawa et al., 1983; Komatsu, Murakami, Matsuda, & Yoshikawa, 1998). To gather more information on saponins in Korean Leguminous crops, we examined saponin compositions in cowpea. It is well established that cowpea contains mainly two saponins, such as soyasaponin Ba (2) and soyasaponin Bb (3) (Kinjo et al., 1998). Although the 80% methanol extract of this species exhibited more than ve minor peaks in UPLC chromatogram (Kinjo et al., 1998), the saponin prole was not fully characterised. In the current study, saponin compositions showed signicant differences in comparison with the earlier report (Kinjo et al., 1998). These differences suggest that the types and contents of saponin in cowpea may vary depending on varieties, sources, and environmental factors as the results obtained on adzuki beans were similar to

T.J. Ha et al. / Food Chemistry 146 (2014) 270277

275

OH O HO O OH O O OAc AcO OAc R1

R1 2: OH 3: OH 4: OH 5: O 6: O- DDMP 7: O- DDMP 8: O- DDMP 9: O- DDMP 10:O-DDMP

R2 CH2OH CH2OH CH2OH CH2OH CH2OH CH2OH H CH2OH H

R3 O- -L-Rhamnose O- - D-Glucose O- -L-Rhamnose O-- L-Rhamnose O --L-Rhamnose O --D-Glucose O- -L-Rhamnose OH OH

HOOC

O OH HO HO O O OH HO HO O O OH OH

CH 2OH

HOOC

AcO

O OH HO R2 O HO

CH 2OH O OH O CH3 DDMP

O OH HO R3

HO

HO

COO OO OH O HO O O O OH HO OH O OH HO OO OH HO OH OO OH HO OH HOOC O OH HO HO O O OH OH OH HO HO HOOC O OH HO OH COO O O HO O O R HO O OH HO O O O

HOOC

HO OH

O OH HO HO O

11

HO

O OH HO OH O

12

O OH HO OH O CHO O

HO HO

HO OH

CH 3

16: CH2OH 17: CH3

O OH

HOOC O OH HO HO O O OH HO HO O O

13
O

20

HO HO O O OH HO OH COO O O HOOC O OH CH3 CH2 OH HOOC HO O O HO O OH OH

O OH HO HO O

HOOC

HO O OH HO OH RO OH HO

O OH O O

18

O O OH OH

HO OH

HO

19

14: GlcA(21) GlcA(21) Glc 15: GlcA(21) Glc(21) Glc

HO

HO OH

Fig. 3. Chemical structures of the saponin derivatives (120) identied from the seeds of nine legume species.

those reported in the previous studies (Lopes-da-Silva, EscribanoBailon, Perez Alonso, Rivas-Gonzalo, & Santos-Buelga, 2007). As illustrated in Fig. 1C, seven saponins were characterised in the 80% methanol extract. The main peak 7 was observed at m/z 1068 ([M]+, tR = 12.9 min). It was conrmed by comparison of the UPLC retention time as well as molecular weight of the soyasaponin bg (7) in soybean and adzuki bean. Our result is in accordance with previously reported data (Decroos et al., 2005; Hu et al., 2002). The remaining six saponins were detected in minor amounts and their structures were tentatively evaluated as soyasaponin Ba (2), soyasaponin Bb (3), soyasaponin Bc (4), soyasaponin ag (6), AzII (14), and AzIV (15) with molecular ions [M]+ at m/z 958, m/z 942, m/z 912, m/z 1084, m/z 1098, and m/z 1084. These structures were also determined by comparing the data reported in the earlier literature as well as saponin proles of soybean and adzuki

bean (Nie et al., 1984). Subsequently, saponin proles of common bean and scarlet runner bean were elucidated based on the characteristics of molecular ions. As described by Kinjo et al. (1998), only three saponins, including phaseoside I, soyasaponin V, and soyasaponin I were found in the 80% methanol extracts. However, our work has evaluated for the rst time various saponin derivatives in the seeds of common bean and scarlet runner bean using UPLCPDAESI/MS analysis. As presented in Fig. 1D and E, the distributions of saponin exhibited similar patterns. The main peaks 6 (tR = 12.1 min) and 7 (tR = 12.9 min) possessed identical molecular ions [M]+ at m/z 1084 and m/z 1068. These peaks were conrmed with soyasaponin ag (6) and soyasaponin bg (7) by comparison of the data reported in soybean saponins and previous studies (Decroos et al., 2005; Hu et al., 2002). Moreover, these saponins contributed about 60% of

276

T.J. Ha et al. / Food Chemistry 146 (2014) 270277

the total peak area in common bean and scarlet runner. Soyasaponin Ab (1), soyasaponin Ba (2), soyasaponin Bb (3), soyasaponin Bc (4), azukisaponin VI (11), and azukisaponin IV (12) were also detected in two species (Fig. 1D and E). The remaining saponins were soyasaponin cg (9) at m/z 922 (Fig. 2I) in common bean and AzIV (15) at m/z 1084 (Fig. 2O) in scarlet runner bean. Additionally, peak 18 (tR = 2.6 min) was assumed to be lablaboside D, by comparing its molecular ion ([M]+, m/z 1262) (Fig. 2R) with the result of the earlier report (Komatsu et al., 1998). Although lablaboside D has been identied from the seeds of Dolichos labrab (Komatsu et al., 1998), this saponin is the rst reported in scarlet runner bean. According to the above results, cultivars and environmental stresses (temperature, light, location, year) may have been the main factors for the variations of saponin proles in common bean and scarlet runner bean (Lopes-da-Silva et al., 2007). For our ongoing research on saponins in legumes, an investigation of the saponin prole in lentil seeds was carried out. The 80% methanol extract of this species was observed more than six peaks in UPLC chromatogram (Fig. 1F). Even though two peaks (tR = 4.5 and 4.8 min) remain unidentied, we characterised four peaks with retention times of 9.4, 10.1, 11.2, and 12.9 min. Their MS spectrums presented the molecular ions [M]+ at m/z 942, m/z 912, m/z 956, and m/z 1068 (Fig. 2CE, and G). Based on the mentioned molecular weights, the four peaks have already been elucidated as soyasaponin Bb (3), soyasaponin Bc (4), soyasaponin Bd (5), and soyasaponin bg (7). Also, their peaks were conrmed by comparison with the saponin prole of soybean and previously published data (Chang et al., 2010; Decroos et al., 2005; Hu et al., 2002). As far as we know, the present research was the rst to characterise the saponins in lentil seeds. To access the saponin prole of chick pea, we also investigated the 80% methanol extract using UPLC PDAESI/MS analysis (Fig. 1G). The identication of each peak was achieved according to the retention times and molecular weights in published data (Decroos et al., 2005; Hu et al., 2002). Even though ve peaks (tR = 1.9, 2.3, 3.2, 7.1 and 8.0 min) were not elucidated on the basis of MS spectra, the remaining peaks 3 and 7 (tR = 9.4 and 12.9 min) were evaluated as soyasaponin Bb (3) and soyasaponin bg (7) by fragment ions at m/z 942 and m/z 1068 (Decroos et al., 2005; Hu et al., 2002). In our continuing survey of legumes, we determined the saponin prole in hyacinth bean. As shown in Fig. 1H, many peaks were detected in UPLC chromatogram. Among them, the chemical structures of six peaks were established based on the characteristics of molecular ions [M]+ and previous works (Komatsu et al., 1998; Nie et al., 1984; Yoshiki et al., 1995). Peaks 16, 17, and 18 had retention times at 1.5, 1.9, and 3.6 min and their molecular ions [M]+ were detected at m/z 1410, m/z 1394, and m/z 1262 (Fig. 2PR). Based on the above considerations, their structures were conrmed as lablaboside E (16), lablaboside F (17), and lablaboside D (18) (Komatsu et al., 1998). Peak 19 (tR = 7.1 min) possessed an identical molecular ion [M]+ at m/z 794 (Fig. 2S) and was tentatively assigned as chikusetsusaponin IVa (Nie et al., 1984). Although this compound was identied from Panax japonicus (Nie et al., 1984), we reported for the rst time that chikusetsusaponin IVa was present in the seeds of hyacinth bean. Major peak 20 (tR = 7.6 min) exhibited the molecular ion [M]+ at m/z 1082 (Fig. 2T). This structure was tentatively elucidated as lablab saponin I (20) by comparison of the data reported in Yoshiki et al. (1995). The MS analysis of the remaining peak 7 (tR = 12.9 min) presented the molecular ion [M]+ at m/z 1068, and its structure was determined as soyasaponin bg (7) according to the results of other legumes. Lastly, we have examined the saponin compositions in broad bean seeds. As indicated in Fig. 1I, minor peaks were observed in the 80% methanol extract of this species. It is commonly documented that soyasaponin Bb (3) is distributed in the methanol extract of broad bean (Decroos et al., 2005). However, this peak and

its molecular weight were not detected in the present study. This phenomenon may be affected by environmental stresses (light, area, year, temperature, and moisture), cultivars, and genetics (Lopes-da-Silva et al., 2007). Although, three peaks (tR = 7.5, 7.8, and 10.3 min) remained unidentied, we demonstrated two saponins. Their structures were elucidated as soyasaponin bg (7) and azukisaponin IV (12) in comparison with retention times (tR = 6.3 and 12.9 min) and molecular weights (m/z 972 and m/z 1068) as shown in other legumes. To our knowledge, little information has been reported for the saponin prole in broad bean seeds. Thus, our result may be important factor for the determination of the quality of broad bean. The results of the current work were undertaken to provide information on a rapid and reliable procedure for the determination of saponins using UPLCPDAESI/MS from various legume species. 4. Conclusion In the current work, we characterised the saponin proles of legume seeds within 15 min using rapid and sensitive UPLCPDA ESI/MS analysis. Moreover, this research was the rst to investigate comparisons of saponin compositions from various legume species. Twenty saponin structures were demonstrated as follows: soyasaponin Ab (1), soyasaponin Ba (2), soyasaponin Bb (3), soyasaponin Bc (4), soyasaponin Bd (5), soyasaponin ag (6), soyasaponin bg (7), soyasaponin ba (8), soyasaponin cg (9), soyasaponin ca (10), azukisaponin VI (11), azukisaponin IV (12), azukisaponin II (13), AzII (14), AzIV (15), lablaboside E (16), lablaboside F (17), lablaboside D (18), chikusetusaponin IVa (19), and lablab saponin I (20). Saponin compositions exhibited important differences in the 80% methanol extract of each legume. Especially, soybean, adzuki bean, common bean, and scarlet runner bean were present many saponins. Interestingly, soyasaponin bg (7) was detected in all legume species at different concentrations. For the development of nutraceuticals and functional foods, future studies are needed to evaluate the potential benecial health properties of saponins and other phytochemicals in legume species. References
Aguilera, Y., Martn-Cabrejas, M. A., Bentez, V., Moll, E., Lpez-Andru, F. J., & Esteban, R. M. (2009). Changes in carbohydrate fraction during dehydration process of common legumes. Journal of Food Composition and Analysis, 22, 678683. Almeida Costa, G. E., Queiroz-Monici, K. S., Reis, S. M. P. M., & Oliveira, A. C. (2006). Chemical composition, dietray bre and resistant starch contents of raw and cooked pea, common bean, chickpea and lentil legumes. Food Chemistry, 94, 327330. Amarowicz, R., & Pegg, R. B. (2008). Legumes as a source of natural antioxidants. European Journal of Lipid Science and Technology, 110, 865878. Applebaum, S. W., Marco, S., & Birk, Y. (1969). Saponins as possible factors of resistance of legume seeds to the attack of insects. Journal of Agricultural and Food Chemistry, 17, 618622. Boschin, G., & Arnoldi, A. (2011). Legumes are valuable sources of tocopherols. Food Chemistry, 127, 11991203. Chang, S. Y., Han, M. J., Joh, E. H., & Kim, D. H. (2010). Liquid chromatography/mass spectrometry-based structural analysis of soyasaponin Ab metabolites by human fecal microora. Journal of Pharmaceutical and Biomedical Analysis, 52, 752756. Dartora, N., de Souza, L. M., Santana-Filho, A. P., Iacomini, M., Valduga, A. T., Gorin, P. A. J., et al. (2011). UPLCPDAMS evaluation of bioactive compounds from leaves of Ilex paraguariensis with different growth conditions, treatments, and ageing. Food Chemistry, 129, 14531461. Decroos, K., Vincken, J. P., Heng, L., Bakker, R., Gruppen, H., & Verstraete, W. (2005). Simultaneous quantication of differently glycosylated, acetylated, and 2,3dihydro-2,5-dihydroxy-6-methyl-4H-pyran-4-one-conjugated soyasaponins using reversed-phase high-performance liquid chromatography with evaporative light scattering detection. Journal of Chromatography A, 1072, 185193. Guajardo-Flores, D., Garca-Patio, M., Serna-Guerrero, D., Gutirrez-Uribe, J. A., & Serna-Saldvar, S. O. (2012). Characterization and quantication of saponins and avonoids in sprouts, seed coats and cotyledons of germinated black beans. Food Chemistry, 134, 13121319.

T.J. Ha et al. / Food Chemistry 146 (2014) 270277 Guan, J., Lai, C. M., & Li, S. P. (2007). A rapid method for the simultaneous determination of 11 saponins in Panax notoginseng using ultra performance liquid chromatography. Journal of Pharmaceutical and Biomedical Analysis, 44, 9961000. Gurnkel, D. M., & Rao, A. V. (2003). Soyasaponins: The relationship between chemical structure and colon anticarcinogenic activity. Nutrition and Cancer, 47, 2433. Han, L., Pan, G., Wang, Y., Song, X., Gao, X., Ma, B., et al. (2011). Rapid proling and identication of triterpenoid saponins in crude extracts from Albizia julibrissin Durazz by ultra high-performance liquid chromatography coupled with electrospray ionization quadrupole time-of-ight tandem mass spectrometry. Journal of Pharmaceutical and Biomedical Analysis, 55, 9961009. Harbourne, N., Marete, E., Christophe, J., Jacquier, C., & ORiordan, D. (2013). Stability of phytochemicals as sources of anti-inammatory nutraceuticals in beverages. Food Research International, 50, 480486. Hu, J., Lee, S. O., Hendrich, S., & Murphy, P. A. (2002). Quantication of the group B soyasaponins by high-performance liquid chromatography. Journal of Agricultural and Food Chemistry, 50, 25872594. Hubert, J., Berger, M., & Dayd, J. (2005). Use of a simplied HPLCUV analysis for soyasaponin B determination: Study of saponin and isoavone variability in soybean cultivars and soy-based health food products. Journal of Agricultural and Food Chemistry, 53, 39233930. Iida, T., Yoshiki, Y., Okubo, K., Ohrui, H., Kinjo, J., & Nohara, T. (1999). Triterpenoid saponins from Vigna angularis. Phytochemistry, 51, 10551058. Iqbal, A., Khalil, I. A., Ateeq, N., & Khan, M. S. (2006). Nutritional quality of important food legumes. Food Chemistry, 97, 331335. Kalogeropoulos, N., Chiou, A., Ioannou, M., Karathanos, V. T., Hassapidou, M., & Andrikopoulos, N. K. (2010). Nutritional evaluation and bioactive microconstituents (phytosterols, tocopherols, polyphenols, triterpenic acids) in cooked dry legumes usually consumed in the Mediterranean countries. Food Chemistry, 121, 682690. Kinjo, J. K., Hatakeyama, M., Udayama, M., Tsutanaga, Y., Yamashita, M., Nohara, T., et al. (1998). HPLC prole analysis of oleanane-glucuronides in several edible beans. Bioscience Biotechnology and Biochemistry, 62, 429433. Kitagawa, I., Wang, H. K., Saito, M., & Yoshikawa, M. (1983). Saponin and sapogenol. XXXII. Chemical constituents of the seeds of Vigna angularis (WILLD) OHWI et OHASHI. (2). Azukisaponin I, II, III, and IV. Chemical and Pharmaceutical Bulletin, 31, 674682. Komatsu, H., Murakami, T., Matsuda, H., & Yoshikawa, M. (1998). Medicinal foodstuffs. XIII.1 Saponin constituents with adjuvant activity from hyacinth bean, the seeds of Dolichos lablab L. (2): Structures of lablabosides D, E, and F. Heterocycles, 48, 703710. Konoshima, T. (1996). Antitumor-promoting activities of triterpenoid glycosides; cancer chemoprevention by saponins. Advances in Experimental Medicine and Biology, 404, 87100.

277

Lai, C. M., Li, S. P., Yu, H., Wan, J. B., Kan, K. W., & Wang, Y. T. (2006). A rapid HPLC ESIMS/MS for qualitative and quantitative analysis of saponins in XUESETONG injection. Journal of Pharmaceutical and Biomedical Analysis, 40, 669678. Lee, J. H., & Cho, K. M. (2012). Changes occurring in compositional components of black soybeans maintained at room temperature for different storage periods. Food Chemistry, 131, 161169. Li, F., Liu, X., Tang, M., Chen, B., Ding, L., Chen, L., et al. (2012). Structure revision of hupehensis saponin F and G and characterization of new trace triterpenoid saponins from Anemone hupehensis by tandem electrospray ionization mass spectrometry. Carbohydrate Research, 353, 4956. Liu, F. F., Ang, C. Y. W., & Springer, D. (2000). Optimization of extraction conditions for active components in Hypericum perforatum using surface methodology. Journal of Agricultural and Food Chemistry, 48, 33643371. Lopes-da-Silva, F., Escribano-Bailon, M. T., Perez Alonso, J. J., Rivas-Gonzalo, J. C., & Santos-Buelga, C. (2007). Anthocyanin pigments in strawberry. LWT-Food Science and Technology, 40, 374382. Luthria, D. L., & Pastor-Corrales, M. A. (2006). Phenolic acids content of fteen dry edible bean (Phaseolus vulgaris L.) varieties. Journal of Food Composition and Analysis, 19, 205211. Mbagwu, F. N., Okafor, V. U., & Ekeanyanwu, J. (2011). Phytochemical screening on four edible legumes (Vigna subterranean, Glycine max, Arachis hypogea, and Vigna uniguiculata) found in eastern Nigeria. African Journal of Plant Science, 5, 370372. Nagata, T., Tsushida, T., Hamaya, E., Enoki, N., Manabe, S., & Nishino, C. (1985). Camellidins, antifungal saponins isolated from Camellia japonica. Agricultural and Biological Chemistry, 49, 11811186. Nie, R. L., Morita, T., Kasai, R., Zhou, J., Wu, C. Y., & Tanaka, O. (1984). Saponins from Chinese medicinal plants. (I). Isolation and structures of hemslosides. Planta Medica, 322327. Price, K. R., Curl, C. L., & Fenwick, G. R. (1986). The saponin content and sapogenol composition of the seed of 13 varieties of legume. Journal of the Science of Food and Agriculture, 37, 11851191. Shahidi, F., & Naczk, M. (2004). Phenolics in food and nutraceuticals. Washington, USA: CRC Press. Sparg, S. G., Light, M. E., & van Staden, J. (2004). Biological activities and distribution of plant saponins. Journal of Ethnopharmacology, 94, 219243. Yoshiki, Y., Kim, J. H., Okubo, K., Nagoya, I., Sakabe, T., & Tamura, N. (1995). A saponin conjugated with 2,3-dihydro-2,5-dihydroxy-6-methyl-4H-pyran-4-one from Dolichos labrab. Phytochemistry, 38, 229231.