- Email: [email protected]
or d -glucaro-1,4-lacton in different apple varieties through hydrophilic interaction chromatography
Food Chemistry 203 (2016) 1–7
Contents lists available at ScienceDirect
Food Chemistry journal homepage: www.elsevier.com/locate/foodchem
Analytical Methods
Determination of D-glucaric acid and/or D-glucaro-1,4-lacton in different apple varieties through hydrophilic interaction chromatography Baogang Xie a,b,⇑, Yalan Liu b, Huiqin Zou b, Yong Son c, Huiyun Wang b, Haipeng Wang b, Jianghua Shao a,⇑ a
Jiangxi Provincial Key Laboratory of Molecular Medicine, The Second Affiliated Hospital of Nanchang University, Nanchang, PR China School of Pharmaceutical Science, Nanchang University, Nanchang 330006, PR China c State Key Laboratory of Food Science and Technology, Institute for Advanced Study, University of Nanchang, Nanchang 330047, PR China b
a r t i c l e
i n f o
Article history: Received 2 May 2015 Received in revised form 20 January 2016 Accepted 1 February 2016 Available online 2 February 2016 Keywords: Apple D-Glucaric acid D-Glucaro-1,4-lacton Hydrophilic interaction liquid chromatography Quantification
a b s t r a c t D-Glucaric
acid (GA) derivatives exhibit anti-cancerogenic properties in vivo in apples, but quantitative information about these derivatives is limited. Hydrophilic interaction-based HPLC with ultraviolet detection or mass spectrometry was developed to quantify GA and/or D-glucaro-1,4-lacton (1,4-GL) in apples. Although the formation of 1,4-GL from GA could be the prerequisite to exert biological effects in vivo, only a small portion of GA (<5%) was identified and converted to 1,4-GL in the rat stomach. The 1,4-GL content in apples ranged from 0.3 mg/g to 0.9 mg/g, and this amount can substantiate health claims associated with apples. The amount of 1,4-GL was 1.5 times higher in Gala and the ratio of 1,4-GL to GA was lower in Green Delicious apples than those in the other varieties. Our findings suggested that the variety and maturity of apples at harvest are factors that determine 1,4-GL content. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Glucuronidation is a major detoxification process in the body that occurs through the balance between glucuronidation (UDPglucuronosyltransferase activity) and deglucuronidation (bglucuronidase activity). As b-Glucuronidase is a lysosomal enzyme that can hydrolyze glucuronide conjugates, the inhibition of this enzyme can enhance the excretion of toxins and ultimately decrease the risk of carcinogenesis (Heerdt, Young, & Borgen, 1995). D-Glucaro-1,4-lacton (1,4-GL) is a specific competitive inhibitor of b-glucuronidase. Many studies have established that oral supplementation of 1,4-GL and its precursor glucaric acid (GA) can control the progression of breast, prostate, and colon cancers at particular stages (Bespalov & Aleksandrov, 2012; Simone, Simone, Pallante, & Simone, 2001; Zoltaszek, Hanausek, Kilianska, & Walaszek, 2008). 1,4-GL exhibits antioxidative properties and reduces the activation of blood platelets (Saluk-Juszczak, Olas, Nowak, Staron, & Wachowicz, 2008) and total serum cholesterol in rats (Walaszek, Szemraj, Hanausek, Adams, & Sherman, 1996; Xie, Liu, Zhan, Ye, & Wei, 2014) and humans (Walaszek, Hanausek, Adams, & Sherman, 1991) in vivo.
⇑ Corresponding authors at: Jiangxi Provincial Key Laboratory of Molecular Medicine, The Second Affiliated Hospital of Nanchang University, PR China (B. Xie). E-mail addresses: [email protected] (B. Xie), [email protected] (J. Shao). http://dx.doi.org/10.1016/j.foodchem.2016.02.009 0308-8146/Ó 2016 Elsevier Ltd. All rights reserved.
Apple is one of the most ubiquitous and wide-spread fruits worldwide (Cornille et al., 2012). The GA content in apples is approximately 3.0 mg/g as determined through pyruvate methods (Walaszek et al., 1996). However, another study shows that the GA content in apples is approximately 0.02 mg/g by using the b-glucuronidase inhibition method (Dwivedi, Heck, Downie, Larroya, & Webb, 1990). This discrepancy can be attributed to the variety, geographical origin, sampling process, and analytical methods employed in each study. The majority of published methods for analyzing GA in fruits and vegetables are based on enzyme inhibition methods. Nevertheless, several compounds from natural products (e.g., baicalin, wogonoside, tea flavanols, and glycyrrhizin) (Narita et al., 1993; Yoshida & Okada, 1999) and compounds related to pyruvate metabolism have been reported to act by inhibiting b-glucuronidase activity and pyruvate content, respectively, and enzyme method cannot be used in this case. Poon, Villeneuve, Chu, and Kinach (1993) established a highperformance liquid chromatography (HPLC) method with an Aminex HPX-87H column to measure GA content in human urine; this method requires the sample to be treated with boronic acid gel and can be used to avoid interferences in the sample matrices. Another HPLC method with a Synergy hydro-RP column was developed to quantify GA content in grapefruits (Perez, Jayaprakasha, Yoo, & Patil, 2008). However, the retention time on the C18 column is insufficient to separate interfering substances for measuring GA content in apple extract because of the high polarity of GA.
2
B. Xie et al. / Food Chemistry 203 (2016) 1–7
Hydrophilic interaction chromatography (HILIC), which was developed in the 1990s, is similar to normal-phase chromatography and uses water as strong eluting solvent (Alpert, 1990). The HILIC separation mechanism is relatively complicated because it involves partitioning, hydrogen bonding, and electrostatic interaction (Chirita, West, Finaru, & Elfakir, 2010). HILIC is a suitable method for hydrophilic and polar compounds, such as sugars, oligosaccharides, and phosphorylated amino acids. This particular set of laboratory techniques has been widely applied to separate different highly polar compounds (Bertolini, Vicentini, Boschetti, Andreatta, & Gatti, 2014; Buck, Voehringer, & Ferger, 2009; Jian, Edom, Xu, & Weng, 2010; Zitka et al., 2014). In this study, a method was developed based on zwitterionic (ZIC) HILIC column coupled with an ultraviolet detection (UV) detector or electrospray ionization–tandem mass spectrometry (ESI-MS) and validated to determine GA and/or 1,4-GL contents in three apple varieties. To the best of our knowledge, this study is the first to provide quantitative information about GA and its bioactive form 1,4-GL in apples. 2. Experimental 2.1. Chemicals and apples 1,4 GL, D-glucarate potassium (GA), and acetonitrile (HPLC grade) were obtained from Sigma Chemical Co., Ltd. (USA). Phosphoric acid and ammonium acetate (HPLC grade) were acquired from Aladdin Reagent Co., Ltd. (China). Water was prepared using an Aquapro water purification system (Taiwan, China). All other chemicals were of analytical grade and used without further purification. Twenty-one apple samples belonging to three apple varieties that originated from five regions were purchased from Jingdong Mall in China. Information on apples used in this study is presented in Table 1. 2.2. Equipment
2.4. HPLC condition 2.4.1. HPLC-UV condition for simultaneous measurement of GA and 1,4-GL The samples were chromatographically separated with SeQuant ZIC-HILIC (250 mm 4.6 mm, 5 lm; Merck KGaA, Germany) protected by an HILIC pre-column. Solvent A consisted of 5 mM ammonium dihydrogen phosphate and 0.1% phosphoric acid (v/v), whereas solvent B contained acetonitrile. The mobile phase flow rate was 0.8 mL/min, and the gradient elution conditions were as follows: 0 min to 3 min, isocratic 83% B; 3 min to 15 min, linear gradient 83% to 65% B; 15 min to 18 min, isocratic 65% B; 18 min to 20 min, linear gradient 65% to 83% B; and 20 min to 30 min, isocratic 83% B. The column was maintained at 35 °C, and the injection volume of the sample was 20 lL. The detected and reference wavelengths were set at 205 (bandwidth 4 nm) and 360 nm (bandwidth 100 nm), respectively. 2.4.2. HPLC–MS/MS condition for measurement of 1,4-GL The HILIC column described in Section 2.4.1 was selected with a mobile phase that consisted of 20 mM ammonium formate solution (acetonitrile 30:70, v/v) for analysis at a flow rate of 0.8 mL/min in isocratic elution. The column was maintained at 35 °C, and the injection volume was 10 lL. The column was indirectly interfaced with the mass spectrometer with flow splitting at 7:3. The ESI source was operated in negative-ion mode, and full-scan mass spectral data were acquired from 50 m/z to 1000 m/z. The optimum values of the source parameter were as follows: capillary voltage, +3.5 kV; drying gas flow, 10.0 L/min; drying gas temperature, 350 °C; and nebulizing gas pressure, 40 psi. The collision energy was set at 15 eV, the fragmentor voltage was fixed at 120 V, and nitrogen was used as collision gas. The instrument was monitored in selected reaction monitoring (SRM) mode with the following monitoring transition for the analyte: m/z 191 ? 85. 2.5. Calibration graphs
The samples were chromatographically separated with an Agilent series 1260 HPLC system and Agilent series 1260 DAD detector. The HPLC-MS method was performed using a triple quadrupole mass spectrometer (6430 QqQ LC/MS system; Agilent Technologies, USA) equipped with an orthogonal ESI source. 2.3. Apple extraction Briefly, 2.0 g of apple flesh was accurately weighed and homogenized with 3.0 mL of distilled water in a glass tissue homogenizer. After homogenization, apple juice was transferred to an Eppendorf tube and added with water until 4.0 mL. The apple extract was centrifuged at 12,600g for 10 min. About 0.20 mL of the supernatant was diluted with 0.80 mL of methanol to prepare the sample for HPLC injection.
Standard stock solutions were prepared by dissolving 40.0 mg of GA or 1,4-GL in 10 mL of distilled water. The final solutions were then prepared from these stocks in 80% of methanol–water (v/v) to generate calibration curves from 20 lg/mL to 400 lg/mL. Each final solution was injected three times, and the peak area was plotted against the corresponding GA or 1,4-GL concentration to construct calibration graphs. 2.6. Method validation procedures 2.6.1. Specificity The specificity of the method was determined by analyzing three apple extracts. The extract was spiked with 50 lg/mL analyte to determine if any compounds were eluted at the same retention
Table 1 Varieties and geographical origins of apples used in this study and GA and 1,4-GL contents determined using HPLC–HILIC-UV (mean ± S.E., n = 3). Group
Variety
Origin
1,4-GL (lg/g)
GA (lg/g)
Ratio of 1,4-GL to GA
1 2 3 4 5 6 7
Fuji Gala Red Delicious Fuji Fuji Green Delicious Green Delicious
Shanxi, China USA Chile Xinjiang, China Shandong, China Chile USA
463.7 ± 34.8 833.7 ± 6.27 620.5 ± 17.3 393.7 ± 33.6 495.1 ± 55.1 553.6 ± 29.1 508.7 ± 33.9
301.7 ± 91.9 381.8 ± 67.3 382.4 ± 70.3 489.3 ± 30.0 236.7 ± 19.5 883.3 ± 38.6 873.4 ± 25.4
1.54 2.18 1.62 0.80 2.09 0.63 0.58
3
B. Xie et al. / Food Chemistry 203 (2016) 1–7
time to that of GA and 1,4-GL as these compounds can interfere with the resolution of the target analytes.
3. Results 3.1. HPLC-UV method for simultaneous measurement of GA and 1,4-GL in apple
2.6.2. Determination of precision, limit of detection (LOD), and limit of quantification (LOQ) Three levels of GA and 1,4-GL at 40, 120, and 200 lg/mL in the above solution were used as the quality control samples (QCs). Five replicates of these QCs were analyzed to determine intra-day precision. Inter-day precision was identified on two separate days for the QCs. LOD is the lowest concentration of the standard GA and 1,4-GL solution, which results in a signal-to-noise ratio of 3:1. LOQ is the sample that results in a signal-to-noise ratio of 10:1.
2.6.3. Method recovery Method recovery was established by separately spiking 20, 40, and 80 lg/mL 1,4-GL and/or GA to three aliquots of the predetermined amount of analytes in apple extract. Method recovery was calculated with the amount of the compound of interest, which was analyzed as a percentage to the theoretical amount presented in the medium.
2.7. Conversion of GA to 1,4-GL in water and in the rat stomach 2.7.1. Conversion of GA to 1,4-GL in water GA solution was prepared by dissolving 50 mg of GA in 10 mL of water (pH = 1.0, 0.05 M H2SO4, and pH = 7.0) in an Eppendorf tube. The tube was then immediately placed in a water bath with a constant-temperature vibrator at 37 °C or 98 °C. Approximately 0.20 mL of the solution was obtained at 0, 4, 8, 12, 24, 36, 48, 72, 84, and 120 h (or 0, 5, 10, 15, 30, 45, 60, 75, 90, and 105 min) by adding 0.80 mL of methanol for HPLC injection as described above. Conversion ratio was estimated using the ratio of the peak area of 1,4-GL at each time point to that of GA at 0 h (min).
2.7.2. Conversion of GA to 1,4-GL in the rat stomach Fourteen male Sprague–Dawley rats weighing 180 g to 200 g were randomly divided into two groups after acclimatizing to the environment for 3 d. All animals were housed in approved facilities by group at a controlled relative humidity (50–70%) and temperature (22 ± 2 °C). The rats were fasted for 12 h before the experiment. The control group (n = 2) received water through oral gavage, and the GA group (n = 12) was administered with 1.0 mL of GA (60 mg/mL). Two rats were sacrificed at 5, 15, 25, 35, 60, and 90 min after the treatment. The stomach was immediately removed, and the contents were washed using 5.0 mL of 10% methanol. Approximately 1.2 mL of the aliquot of the solution was centrifuged at 5600g for 10 min. The obtained supernatant was stored at 20 °C until analysis. Animal experiments were conducted in accordance with the Guidelines for Animal Experiments at Nanchang University (Nanchang, China). The protocol was approved by the Animal Ethics Committee at Nanchang University.
2.8. Quantification of GA and/or 1,4-GL in apples As mentioned above, GA and/or 1,4-GL was separated and quantified in apples through HILIC-based HPLC-UV. 1,4-GL content in samples from group 2 was further determined through HPLC-MS/ MS. The concentrations of GA and/or 1,4-GL in each sample were calculated with the regression equation using an appropriate dilution factor.
3.1.1. Optimized HPLC separation condition Venusil HILIC (5 lm, 100 A, 4.6 mm 250 mm) and SeQuant ZIC-HILIC (5 lm, 100 A, 250 mm 4.6 mm) columns were used to adequately separate the peaks of GA and 1,4-GL from those of the formulation ingredients. Effective separations and short balance time of the column were observed in analysis with the SeQuant ZIC-HILIC column. The mobile phases were optimized using a mixture of ammonium dihydrogen phosphate and 0.1% of phosphoric acid/acetonitrile as eluent. Ammonium dihydrogen phosphate and phosphoric acid were used because they do not absorb UV–vis.
3.1.2. Method validation The specificity of the method was presented on the chromatogram (Fig. 1) of GA and 1,4-GL (100 lg/mL) solution, apple extract, and apple extract spiked with 50 lg/mL analyte. The peak of the analyte in the apple extract was identified by comparing the retention time with that of the standard. GA and 1,4-GL were properly resolved with few interferences under assay conditions (Fig. 1). The precision of GA and 1,4-GL was evaluated by the QCs analysis expressed as relative standard deviation (RSD), which was calculated from five replicate injections of QCs within one day or in two separate days. Our data showed that the intra- and inter-day precisions were less than 3.5%. The analyte stability in the solution was detected by repeatedly examining the samples during the course of experimentation on the same day and after solution storage (GA or 1,4-GL, 100 lg/mL) for 12, 24, 48, and 72 h at 4 °C. GA and 1,4-GL solutions were stable for 48 h as indicated by the less than 5.0% mean RSD between the peak areas for the samples stored at 4 °C. Therefore, the experiment for quantifying GA and 1,4-GL in apples should be performed within 48 h after preparation and the samples should be stored at 4 °C. Linearity was calculated using linear regression with the least squares method on seven standard solutions from 20 lg/mL to 400 lg/mL. Each level was prepared once, and each solution was injected three times. Linear ranges, linear regression equations, LOD, and LOQ of GA and 1,4-GL are shown in Table 2. Accuracy was based on the recovery of known amounts of analytes, spiked GA, and 1,4-GL in the apple extract. Three samples were prepared in triplicate at three levels over a range of 80% to 120% of the target concentrations. The mean quantitative recoveries (mean ± SD, n = 3) for GA and 1,4-GL were 99.6% ± 2.1% and 101.5% ± 2.4%, respectively.
3.2. Conversion of GA to 1,4-GL in water and in the rat stomach 3.2.1. GA conversion to 1,4-GL in water The developed HPLC–HILIC-UV method enabled the monitoring of GA conversion. Three peaks (i.e., GA, 1,4-GL, and D-glucaro-6,3lactone) were observed in the HPLC-UV chromatogram (Fig. 2) when the conversion reached an equilibrium. At 37 °C, the equilibrium was reached at approximately 48 h and the conversion ratio was approximately 20% and 30% at pH = 7.0 and 1.0, respectively (SFig. 1a). Only 15 min was required to reach equilibrium at 98 °C (SFig. 1b). No differences were observed between the conversion ratio at 37 °C and 98 °C. Previous studies (Ho & Ho, 1990) demonstrated that the equilibrium among GA, 1,4-GL, and D-glucaro-6,3-lactone
is pH dependent; hence, low pH could result
4
B. Xie et al. / Food Chemistry 203 (2016) 1–7
Fig. 1. HPLC-UV (205 nm) chromatograms of apple; (a) apple extract spiked with 1,4-GL and GA; (b) apple extract; and (c) standard solution of 1,4-GL (100 lg/mL) and GA (100 lg/mL).
Table 2 Data for calibration graphs with the developed methods. Method
Linear ranges (lg/mL)
Linear regression equationa
Coefficient (r2)
LOD (lg/mL)
LOQ (lg/mL)
HPLC-UV (GA) (1,4-GL)
20 to 400 20 to 400 2 to 200
A = 0.0848C + 0.6958 A = 0.143C + 0.5917 A = 9.8008C 167.59
0.9993 0.9996 0.9932
12.5 20 1
20 40 5
HPLC-MS/MS (1,4-GL) a
According to A = a C + b, where C = compound concentration, A = peak area.
in generation of high amounts of 1,4-GL. The present results were consistent with these findings. 3.2.2. GA conversion to 1,4-GL in the rat stomach The stomach contents after the rat was administered with GA were analyzed using the HPLC–HILIC-UV method. SFig. 2 shows that high amounts of GA were observed in the stomach 15 min after GA administration, but a low GA content was detected in the sample obtained at 60 min. This finding suggested that the existing time of GA in the rat stomach was lower than 60 min. Only a small amount of GA could also be converted to 1,4-GL (less than 5%) in the stomach, which was consistent with GA conversion in water at pH = 1.0 and 37 °C. 3.3. HPLC-MS/MS method for determining 1,4-GL in apples The methods used in the literature to determine 1,4-GL are gas chromatography (GC) and HPLC. The GC method includes tedious sample preparation (Wang, Gan, Tang, Wang, & Tan, 2010; Yang et al., 2009), whereas the HPLC method with the Spherisorb NH2 column shows poor baseline separation of 1,4-GL (Laakso, Tokola, & Hirvisalo, 1983; Walaszek et al., 1997). In this regard, the present study developed an HILIC-based HPLC-MS/MS method for quantitative analysis of 1,4-GL in apple extracts to confirm 1,4-GL content
in apples. In particular, this study used SRM mode with m/z 191 ? 85 to quantify 1,4-GL. No interference from 1,4-GL was observed under the optimized conditions (Fig. 3). The developed analytical procedure was validated for the quantification of 1,4-GL. Excellent RSD values were obtained for repeatability (<3%) and reproducibility (<5%). The quantitative recoveries of 1,4-GL from the apple samples were measured as 99.4% ± 1.9% (n = 3), which reflects the sufficiently high reliability and accuracy of the method for quantification. The results of linear ranges, linear regression equations, LOD, and LOQ of 1,4-GL are shown in Table 2.
3.4. Contents of GA and 1,4-GL in apples Since UV detection is non-selective with the potential interferences, there is a possibility that co-eluting peaks behind the peak of 1,4-GL when dealing with real sample analysis. Therefore, 1,4GL content in samples from one group (group 2) was further determined through the HPLC-MS/MS method. The results showed that 1,4-GL contents determined using HPLC-MS/MS were 10% higher than those measured through HPLC-UV, suggesting that there is no overlapping components with1,4-GL. In this study, the valleyto-valley integration technique was employed for calculating the peak area of 1.4-GL as the response does not return to the baseline
B. Xie et al. / Food Chemistry 203 (2016) 1–7
5
Fig. 2. HPLC-UV (205 nm) chromatograms of GA conversion in water. (a) GA conversion reached equilibrium at pH = 1.0 and 37 °C; and (b) GA conversion reached equilibrium at pH = 1.0 and 98 °C.
Fig. 3. (a) MS/MS spectra of 1,4-GL; and (b) HPLC-MS/MS chromatogram of apples in SRM mode with m/z 191 ? 85 separated on the HILIC column.
between peaks (Fig. 1), which may be the reason for about 10% lower contents of 1.4-GL determined by HPLC-UV. Therefore, we supposed that both methods were reliable and robust to quantify 1,4-GL in apples. Although the HPLC-MS/MS method is more specific and sensitive to low LOD and LOQ than HPLC-UV, it requires the use of expensive mass spectrometers. The HPLC–HILIC-UV method was validated to be robust to allow its routine laboratory use for simultaneous measurement of GA and 1,4-GL in apples. Therefore, the method was applied to analyze three apple varieties that originated from five regions. The contents of GA and 1,4-GL in apples are presented in Table 1. 4. Discussion Apple is one of the most wide-spread fruits eaten daily by millions of people worldwide. 1,4-GL and GA are beneficial for general
health promotion (‘‘Cancer, Lifestyle Modification and Glucarate”, 2001). Two papers published in the last two decades described and quantified GA content in apples by using pyruvate (Walaszek et al., 1996) and b-glucuronidase inhibition methods (Dwivedi et al., 1990). As GA exhibits high polarity, its retention time on the C18 column is very short to separate interfering substances in apples. The HILIC column exhibited advantages in terms of retention and selectivity, which provided various ingredients of apples with a good separation of GA and 1,4-GL. As high proportion of organic solvent is required in sampling and in the mobile phase of the HILIC system, the supernatants from apple extraction can be directly injected into the HPLC system by adding a suitable volume of methanol. Thus, the procedure for sample preparation is simple. Collectively, these approaches can serve as a novel method for detecting GA derivatives in fruit and vegetables and are more
6
B. Xie et al. / Food Chemistry 203 (2016) 1–7
highly specific and sensitive than enzyme methods. Unlike HPLCMS/MS, the HPLC-UV method is stable, reproducible, and has practical advantages, such as low cost for simultaneous measurement of GA and 1,4-GL in apples. Heerdt et al. (1995) supposed that GA can be converted to 1,4-GL and obtained an equilibrium that consists of approximately 30% 1,4-GL and 30% D-glucaro-6,3-lactone in the acidic environment of the stomach. However, they did not describe the experimental process and analytical method for GA conversion in the stomach. The present study found that the conversion of GA to 1,4-GL to reach the equilibrium (conversion ratio of approximately 30%) was slow (approximately 48 h) at pH = 1.0 and 37 °C in water. In addition, the existing GA time in the rat stomach was lower than 60 min and only less than 5% GA can be converted to 1,4-GL in the stomach. Similar results were obtained in a previous study, in which potassium hydrogen D-[14C]glucarate was orally administered to rats and analyzed using HPLC and radioactive technology (Walaszek et al., 1997). 1,4-GL is a specific and potent inhibitor of b-glucuronidase compared with GA and D-glucaro-6,3-lactone (Kiyomoto, Harigaya, Ohshima, & Morita, 1963). Walaszek et al. (1997) demonstrated that the formation of 1,4-GL from GA can be the prerequisite to exert their benefits in general health promotion. Thus, the determination of 1,4-GL in apples plays a key role in quality evaluation of apples. Our data showed that 1,4-GL contents in apples ranged from 0.3 mg/g to 0.9 mg/g. To the best of our knowledge, this study is the first to present quantitative information about 1,4-GL in apples. Previous study (Kiyomoto et al., 1963) showed that hepatic b-glucuronidase activity caused 50% inhibition by single administration of 200 mg/kg 1,4-GL in rats. Therefore, we believe that the enterohepatic circulation of many drugs, toxins, steroids, and other compounds subject to glucuronidation can be modified if a person can consume 250 g of apple daily. Our findings provide key data to substantiate the positive effects of apples on general health promotion. The contents of 1,4-GL (x-axis) compared with those of GA (yaxis) in apples are plotted in Fig. 4 to determine the relationship between the samples. The plot shows that the samples can be classified into three classes. Samples from Fuji and Red Delicious formed a group, whereas Green Delicious and Gala were completely separated from each other in the plot. The classification of these samples suggested that the GA derivatives that constituted apple possibly differed among varieties. Approximately 1.5 times higher content of 1,4-GL was observed in Gala apples than that
1200
Variety Group 1 Fiji Group 2 Gala Group 3 Red Delicious Group 4 Fuji Group 5 Fuji Group 6 Green delicious Group 7 Green delicious
GA contents in apples (ug/g)
1000
800
600
400
200
300
350
400
450
500
550
600
650
700
750
1,4-GL contents in apples (ug/g)
Fig. 4. Plot that shows 1,4-GL and GA contents in apples.
800
850
in the other varieties. Furthermore, Table 1 shows that the ratio of 1,4-GL to GA in the Green Delicious group was significantly lower than that in the Red Delicious and other groups. Previous studies demonstrated that the concentration of esters could increase during climacteric ripening and reached maximum values at the climacteric peak (Guo, Yue, & Yuan, 2012; Ortiz, Graell, & Lara, 2011). Apples in the Green Delicious group were physiologically immature and produced lower concentrations of 1,4-GL. Therefore, our data indicated that the state of fruit maturity at harvest plays an important role in 1,4-GL contents in apples. 5. Conclusion An HILIC-based HPLC method coupled with UV or MS was developed and validated to quantify GA and/or 1,4-GL in three apple varieties that originated from five regions. The methods are highly recommended for the quality assessment of different GA derivative samples, including new, commercial juice, and pharmaceutical apple products. As the formation of 1,4-GL from GA can be the prerequisite to exert health benefits in vivo, GA conversion to 1,4-GL was investigated in water and rat stomach. Only a small portion of GA could be transformed to 1,4-GL in the rat stomach after GA oral administration. Furthermore, our data revealed that the variety and maturity of apples at harvest are factors that determine 1,4-GL contents and can be used to improve the nutritional quality of apples. Conflict of interest The authors have no conflict of interest. Acknowledgements The work was supported by a grant from the Natural Science Foundation of China (No. 81560631), China Postdoctoral Science Foundation (No. 143750), Postdoctoral Science Foundation of Jiangxi Province (No. 2015KY41), and the Innovation Fund Designated for Graduate Students of Jiangxi Province of China (No. YC2014-S076 and cx2015205). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.foodchem.2016. 02.009. References Alpert, A. J. (1990). Hydrophilic-interaction chromatography for the separation of peptides, nucleic acids and other polar compounds. Journal of Chromatography, 499, 177–196. Bertolini, T., Vicentini, L., Boschetti, S., Andreatta, P., & Gatti, R. (2014). A novel automated hydrophilic interaction liquid chromatography method using diodearray detector/electrospray ionization tandem mass spectrometry for analysis of sodium risedronate and related degradation products in pharmaceuticals. Journal of Chromatography A, 1365, 131–139. Bespalov, V. G., & Aleksandrov, V. A. (2012). Anticarcinogenic effect of potassium salts of glucaric and glucuronic acid in induced models of cervical and esophageal tumors. Voprosy Onkologii, 58(4), 537–540. Buck, K., Voehringer, P., & Ferger, B. (2009). Rapid analysis of GABA and glutamate in microdialysis samples using high performance liquid chromatography and tandem mass spectrometry. Journal of Neuroscience Methods, 182(1), 78–84. Cancer, Lifestyle Modification and Glucarate (2001). Chirita, R. I., West, C., Finaru, A. L., & Elfakir, C. (2010). Approach to hydrophilic interaction chromatography column selection: Application to neurotransmitters analysis. Journal of Chromatography A, 1217(18), 3091–3104. Cornille, A., Gladieux, P., Smulders, M. J., Roldan-Ruiz, I., Laurens, F., Le Cam, B., ... Giraud, T. (2012). New insight into the history of domesticated apple: Secondary contribution of the European wild apple to the genome of cultivated varieties. PLoS Genetics, 8(5), e1002703.
B. Xie et al. / Food Chemistry 203 (2016) 1–7 Dwivedi, C., Heck, W. J., Downie, A. A., Larroya, S., & Webb, T. E. (1990). Effect of calcium glucarate on beta-glucuronidase activity and glucarate content of certain vegetables and fruits. Biochemical Medicine and Metabolic Biology, 43(2), 83–92. Guo, J., Yue, T., & Yuan, Y. (2012). Feature selection and recognition from nonspecific volatile profiles for discrimination of apple juices according to variety and geographical origin. Journal of Food Science, 77(10), C1090–C1096. Heerdt, A. S., Young, C. W., & Borgen, P. I. (1995). Calcium glucarate as a chemopreventive agent in breast cancer. Israel Journal of Medical Sciences, 31 (2–3), 101–105. Ho, Y. D., & Ho, K. J. (1990). Biliary D-glucaric acid: Its quantitation and preventive role in gallstone formation. Scandinavian Journal of Gastroenterology, 25(6), 631–640. Jian, W. Y., Edom, R. W., Xu, Y. D., & Weng, N. D. (2010). Recent advances in application of hydrophilic interaction chromatography for quantitative bioanalysis. Journal of Separation Science, 33(6–7), 681–697. Kiyomoto, A., Harigaya, S., Ohshima, S., & Morita, T. (1963). Studies on glucosaccharolactones. I. Glucosaccharo-1:4-lactone. Biochemical Pharmacology, 12, 105–112. Laakso, E. I., Tokola, R. A., & Hirvisalo, E. L. (1983). Determination of D-glucaric acid by high-performance liquid chromatography. Journal of Chromatography, 278(2), 406–411. Narita, M., Nagai, E., Hagiwara, H., Aburada, M., Yokoi, T., & Kamataki, T. (1993). Inhibition of beta-glucuronidase by natural glucuronides of kampo medicines using glucuronide of SN-38 (7-ethyl-10-hydroxycamptothecin) as a substrate. Xenobiotica, 23(1), 5–10. Ortiz, A., Graell, J., & Lara, I. (2011). Volatile ester-synthesising capacity throughout on-tree maturation of ’Golden Reinders’ apples. Scientia Horticulturae, 131, 6–14. Perez, J. L., Jayaprakasha, G. K., Yoo, K. S., & Patil, B. S. (2008). Development of a method for the quantification of D-glucaric acid in different varieties of grapefruits by high-performance liquid chromatography and mass spectra. Journal of Chromatography A, 1190(1–2), 394–397. Poon, R., Villeneuve, D. C., Chu, I., & Kinach, R. (1993). HPLC determination of Dglucaric acid in human urine. Journal of Analytical Toxicology, 17(3), 146–150.
7
Saluk-Juszczak, J., Olas, B., Nowak, P., Staron, A., & Wachowicz, B. (2008). Protective effects of D-glucaro-1,4-lactone against oxidative modifications in blood platelets. Nutrition, Metabolism and Cardiovascular Diseases, 18(6), 422–428. Simone, Charles B., Simone, Nicole L., Pallante, Michael, & Simone, Charles B. (2001). Cancer, Lifestyle Modification and Glucarate. J. Orthomol. Med., 16(2), 83–90. Walaszek, Z., Hanausek, M., Adams, A. K., & Sherman, U. (1991). Cholesterollowering effects of dietary D-Glucarate. Faseb Journal, 5(5). A930-A930. Walaszek, Z., Szemraj, J., Hanausek, M., Adams, A. K., & Sherman, U. (1996). DGlucaric acid content of various fruits and vegetables and cholesterol-lowering effects of dietary D-glucarate in the rat. Nutrition Research, 16(4), 673–681. Walaszek, Z., Szemraj, J., Narog, M., Adams, A. K., Kilgore, J., Sherman, U., & Hanausek, M. (1997). Metabolism, uptake, and excretion of a D-glucaric acid salt and its potential use in cancer prevention. Cancer Detection and Prevention, 21 (2), 178–190. Wang, K., Gan, X., Tang, X., Wang, S., & Tan, H. (2010). Determination of D-saccharic acid-1,4-lactone from brewed kombucha broth by high-performance capillary electrophoresis. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci., 878(3–4), 371–374. Xie, B., Liu, A., Zhan, X., Ye, X., & Wei, J. (2014). Alteration of gut bacteria and metabolomes after glucaro-1,4-lactone treatment contributes to the prevention of hypercholesterolemia. Journal of Agriculture and Food Chemistry, 62(30), 7444–7451. Yang, Z.-W., Ji, B.-P., Zhou, F., Li, B., Luo, Y., Yang, L., & Li, T. (2009). Hypocholesterolaemic and antioxidant effects of kombucha tea in highcholesterol fed mice. Journal of the Science of Food and Agriculture, 89(1), 150–156. Yoshida, T., & Okada, T. (1999). Peptide separation in normal-phase liquid chromatography - Study of selectivity and mobile phase effects on various columns. Journal of Chromatography A, 840(1), 1–9. Zitka, O., Heger, Z., Kominkova, M., Skalickova, S., Krizkova, S., Adam, V., & Kizek, R. (2014). Preconcentration based on paramagnetic microparticles for the separation of sarcosine using hydrophilic interaction liquid chromatography coupled with coulometric detection. Journal of Separation Science, 37(5), 465–575. Zoltaszek, R., Hanausek, M., Kilianska, Z. M., & Walaszek, Z. (2008). The biological role of D-glucaric acid and its derivatives: Potential use in medicine. Postepy Hig Med Dosw (Online), 62, 451–462.
Contents lists available at ScienceDirect
Food Chemistry journal homepage: www.elsevier.com/locate/foodchem
Analytical Methods
Determination of D-glucaric acid and/or D-glucaro-1,4-lacton in different apple varieties through hydrophilic interaction chromatography Baogang Xie a,b,⇑, Yalan Liu b, Huiqin Zou b, Yong Son c, Huiyun Wang b, Haipeng Wang b, Jianghua Shao a,⇑ a
Jiangxi Provincial Key Laboratory of Molecular Medicine, The Second Affiliated Hospital of Nanchang University, Nanchang, PR China School of Pharmaceutical Science, Nanchang University, Nanchang 330006, PR China c State Key Laboratory of Food Science and Technology, Institute for Advanced Study, University of Nanchang, Nanchang 330047, PR China b
a r t i c l e
i n f o
Article history: Received 2 May 2015 Received in revised form 20 January 2016 Accepted 1 February 2016 Available online 2 February 2016 Keywords: Apple D-Glucaric acid D-Glucaro-1,4-lacton Hydrophilic interaction liquid chromatography Quantification
a b s t r a c t D-Glucaric
acid (GA) derivatives exhibit anti-cancerogenic properties in vivo in apples, but quantitative information about these derivatives is limited. Hydrophilic interaction-based HPLC with ultraviolet detection or mass spectrometry was developed to quantify GA and/or D-glucaro-1,4-lacton (1,4-GL) in apples. Although the formation of 1,4-GL from GA could be the prerequisite to exert biological effects in vivo, only a small portion of GA (<5%) was identified and converted to 1,4-GL in the rat stomach. The 1,4-GL content in apples ranged from 0.3 mg/g to 0.9 mg/g, and this amount can substantiate health claims associated with apples. The amount of 1,4-GL was 1.5 times higher in Gala and the ratio of 1,4-GL to GA was lower in Green Delicious apples than those in the other varieties. Our findings suggested that the variety and maturity of apples at harvest are factors that determine 1,4-GL content. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Glucuronidation is a major detoxification process in the body that occurs through the balance between glucuronidation (UDPglucuronosyltransferase activity) and deglucuronidation (bglucuronidase activity). As b-Glucuronidase is a lysosomal enzyme that can hydrolyze glucuronide conjugates, the inhibition of this enzyme can enhance the excretion of toxins and ultimately decrease the risk of carcinogenesis (Heerdt, Young, & Borgen, 1995). D-Glucaro-1,4-lacton (1,4-GL) is a specific competitive inhibitor of b-glucuronidase. Many studies have established that oral supplementation of 1,4-GL and its precursor glucaric acid (GA) can control the progression of breast, prostate, and colon cancers at particular stages (Bespalov & Aleksandrov, 2012; Simone, Simone, Pallante, & Simone, 2001; Zoltaszek, Hanausek, Kilianska, & Walaszek, 2008). 1,4-GL exhibits antioxidative properties and reduces the activation of blood platelets (Saluk-Juszczak, Olas, Nowak, Staron, & Wachowicz, 2008) and total serum cholesterol in rats (Walaszek, Szemraj, Hanausek, Adams, & Sherman, 1996; Xie, Liu, Zhan, Ye, & Wei, 2014) and humans (Walaszek, Hanausek, Adams, & Sherman, 1991) in vivo.
⇑ Corresponding authors at: Jiangxi Provincial Key Laboratory of Molecular Medicine, The Second Affiliated Hospital of Nanchang University, PR China (B. Xie). E-mail addresses: [email protected] (B. Xie), [email protected] (J. Shao). http://dx.doi.org/10.1016/j.foodchem.2016.02.009 0308-8146/Ó 2016 Elsevier Ltd. All rights reserved.
Apple is one of the most ubiquitous and wide-spread fruits worldwide (Cornille et al., 2012). The GA content in apples is approximately 3.0 mg/g as determined through pyruvate methods (Walaszek et al., 1996). However, another study shows that the GA content in apples is approximately 0.02 mg/g by using the b-glucuronidase inhibition method (Dwivedi, Heck, Downie, Larroya, & Webb, 1990). This discrepancy can be attributed to the variety, geographical origin, sampling process, and analytical methods employed in each study. The majority of published methods for analyzing GA in fruits and vegetables are based on enzyme inhibition methods. Nevertheless, several compounds from natural products (e.g., baicalin, wogonoside, tea flavanols, and glycyrrhizin) (Narita et al., 1993; Yoshida & Okada, 1999) and compounds related to pyruvate metabolism have been reported to act by inhibiting b-glucuronidase activity and pyruvate content, respectively, and enzyme method cannot be used in this case. Poon, Villeneuve, Chu, and Kinach (1993) established a highperformance liquid chromatography (HPLC) method with an Aminex HPX-87H column to measure GA content in human urine; this method requires the sample to be treated with boronic acid gel and can be used to avoid interferences in the sample matrices. Another HPLC method with a Synergy hydro-RP column was developed to quantify GA content in grapefruits (Perez, Jayaprakasha, Yoo, & Patil, 2008). However, the retention time on the C18 column is insufficient to separate interfering substances for measuring GA content in apple extract because of the high polarity of GA.
2
B. Xie et al. / Food Chemistry 203 (2016) 1–7
Hydrophilic interaction chromatography (HILIC), which was developed in the 1990s, is similar to normal-phase chromatography and uses water as strong eluting solvent (Alpert, 1990). The HILIC separation mechanism is relatively complicated because it involves partitioning, hydrogen bonding, and electrostatic interaction (Chirita, West, Finaru, & Elfakir, 2010). HILIC is a suitable method for hydrophilic and polar compounds, such as sugars, oligosaccharides, and phosphorylated amino acids. This particular set of laboratory techniques has been widely applied to separate different highly polar compounds (Bertolini, Vicentini, Boschetti, Andreatta, & Gatti, 2014; Buck, Voehringer, & Ferger, 2009; Jian, Edom, Xu, & Weng, 2010; Zitka et al., 2014). In this study, a method was developed based on zwitterionic (ZIC) HILIC column coupled with an ultraviolet detection (UV) detector or electrospray ionization–tandem mass spectrometry (ESI-MS) and validated to determine GA and/or 1,4-GL contents in three apple varieties. To the best of our knowledge, this study is the first to provide quantitative information about GA and its bioactive form 1,4-GL in apples. 2. Experimental 2.1. Chemicals and apples 1,4 GL, D-glucarate potassium (GA), and acetonitrile (HPLC grade) were obtained from Sigma Chemical Co., Ltd. (USA). Phosphoric acid and ammonium acetate (HPLC grade) were acquired from Aladdin Reagent Co., Ltd. (China). Water was prepared using an Aquapro water purification system (Taiwan, China). All other chemicals were of analytical grade and used without further purification. Twenty-one apple samples belonging to three apple varieties that originated from five regions were purchased from Jingdong Mall in China. Information on apples used in this study is presented in Table 1. 2.2. Equipment
2.4. HPLC condition 2.4.1. HPLC-UV condition for simultaneous measurement of GA and 1,4-GL The samples were chromatographically separated with SeQuant ZIC-HILIC (250 mm 4.6 mm, 5 lm; Merck KGaA, Germany) protected by an HILIC pre-column. Solvent A consisted of 5 mM ammonium dihydrogen phosphate and 0.1% phosphoric acid (v/v), whereas solvent B contained acetonitrile. The mobile phase flow rate was 0.8 mL/min, and the gradient elution conditions were as follows: 0 min to 3 min, isocratic 83% B; 3 min to 15 min, linear gradient 83% to 65% B; 15 min to 18 min, isocratic 65% B; 18 min to 20 min, linear gradient 65% to 83% B; and 20 min to 30 min, isocratic 83% B. The column was maintained at 35 °C, and the injection volume of the sample was 20 lL. The detected and reference wavelengths were set at 205 (bandwidth 4 nm) and 360 nm (bandwidth 100 nm), respectively. 2.4.2. HPLC–MS/MS condition for measurement of 1,4-GL The HILIC column described in Section 2.4.1 was selected with a mobile phase that consisted of 20 mM ammonium formate solution (acetonitrile 30:70, v/v) for analysis at a flow rate of 0.8 mL/min in isocratic elution. The column was maintained at 35 °C, and the injection volume was 10 lL. The column was indirectly interfaced with the mass spectrometer with flow splitting at 7:3. The ESI source was operated in negative-ion mode, and full-scan mass spectral data were acquired from 50 m/z to 1000 m/z. The optimum values of the source parameter were as follows: capillary voltage, +3.5 kV; drying gas flow, 10.0 L/min; drying gas temperature, 350 °C; and nebulizing gas pressure, 40 psi. The collision energy was set at 15 eV, the fragmentor voltage was fixed at 120 V, and nitrogen was used as collision gas. The instrument was monitored in selected reaction monitoring (SRM) mode with the following monitoring transition for the analyte: m/z 191 ? 85. 2.5. Calibration graphs
The samples were chromatographically separated with an Agilent series 1260 HPLC system and Agilent series 1260 DAD detector. The HPLC-MS method was performed using a triple quadrupole mass spectrometer (6430 QqQ LC/MS system; Agilent Technologies, USA) equipped with an orthogonal ESI source. 2.3. Apple extraction Briefly, 2.0 g of apple flesh was accurately weighed and homogenized with 3.0 mL of distilled water in a glass tissue homogenizer. After homogenization, apple juice was transferred to an Eppendorf tube and added with water until 4.0 mL. The apple extract was centrifuged at 12,600g for 10 min. About 0.20 mL of the supernatant was diluted with 0.80 mL of methanol to prepare the sample for HPLC injection.
Standard stock solutions were prepared by dissolving 40.0 mg of GA or 1,4-GL in 10 mL of distilled water. The final solutions were then prepared from these stocks in 80% of methanol–water (v/v) to generate calibration curves from 20 lg/mL to 400 lg/mL. Each final solution was injected three times, and the peak area was plotted against the corresponding GA or 1,4-GL concentration to construct calibration graphs. 2.6. Method validation procedures 2.6.1. Specificity The specificity of the method was determined by analyzing three apple extracts. The extract was spiked with 50 lg/mL analyte to determine if any compounds were eluted at the same retention
Table 1 Varieties and geographical origins of apples used in this study and GA and 1,4-GL contents determined using HPLC–HILIC-UV (mean ± S.E., n = 3). Group
Variety
Origin
1,4-GL (lg/g)
GA (lg/g)
Ratio of 1,4-GL to GA
1 2 3 4 5 6 7
Fuji Gala Red Delicious Fuji Fuji Green Delicious Green Delicious
Shanxi, China USA Chile Xinjiang, China Shandong, China Chile USA
463.7 ± 34.8 833.7 ± 6.27 620.5 ± 17.3 393.7 ± 33.6 495.1 ± 55.1 553.6 ± 29.1 508.7 ± 33.9
301.7 ± 91.9 381.8 ± 67.3 382.4 ± 70.3 489.3 ± 30.0 236.7 ± 19.5 883.3 ± 38.6 873.4 ± 25.4
1.54 2.18 1.62 0.80 2.09 0.63 0.58
3
B. Xie et al. / Food Chemistry 203 (2016) 1–7
time to that of GA and 1,4-GL as these compounds can interfere with the resolution of the target analytes.
3. Results 3.1. HPLC-UV method for simultaneous measurement of GA and 1,4-GL in apple
2.6.2. Determination of precision, limit of detection (LOD), and limit of quantification (LOQ) Three levels of GA and 1,4-GL at 40, 120, and 200 lg/mL in the above solution were used as the quality control samples (QCs). Five replicates of these QCs were analyzed to determine intra-day precision. Inter-day precision was identified on two separate days for the QCs. LOD is the lowest concentration of the standard GA and 1,4-GL solution, which results in a signal-to-noise ratio of 3:1. LOQ is the sample that results in a signal-to-noise ratio of 10:1.
2.6.3. Method recovery Method recovery was established by separately spiking 20, 40, and 80 lg/mL 1,4-GL and/or GA to three aliquots of the predetermined amount of analytes in apple extract. Method recovery was calculated with the amount of the compound of interest, which was analyzed as a percentage to the theoretical amount presented in the medium.
2.7. Conversion of GA to 1,4-GL in water and in the rat stomach 2.7.1. Conversion of GA to 1,4-GL in water GA solution was prepared by dissolving 50 mg of GA in 10 mL of water (pH = 1.0, 0.05 M H2SO4, and pH = 7.0) in an Eppendorf tube. The tube was then immediately placed in a water bath with a constant-temperature vibrator at 37 °C or 98 °C. Approximately 0.20 mL of the solution was obtained at 0, 4, 8, 12, 24, 36, 48, 72, 84, and 120 h (or 0, 5, 10, 15, 30, 45, 60, 75, 90, and 105 min) by adding 0.80 mL of methanol for HPLC injection as described above. Conversion ratio was estimated using the ratio of the peak area of 1,4-GL at each time point to that of GA at 0 h (min).
2.7.2. Conversion of GA to 1,4-GL in the rat stomach Fourteen male Sprague–Dawley rats weighing 180 g to 200 g were randomly divided into two groups after acclimatizing to the environment for 3 d. All animals were housed in approved facilities by group at a controlled relative humidity (50–70%) and temperature (22 ± 2 °C). The rats were fasted for 12 h before the experiment. The control group (n = 2) received water through oral gavage, and the GA group (n = 12) was administered with 1.0 mL of GA (60 mg/mL). Two rats were sacrificed at 5, 15, 25, 35, 60, and 90 min after the treatment. The stomach was immediately removed, and the contents were washed using 5.0 mL of 10% methanol. Approximately 1.2 mL of the aliquot of the solution was centrifuged at 5600g for 10 min. The obtained supernatant was stored at 20 °C until analysis. Animal experiments were conducted in accordance with the Guidelines for Animal Experiments at Nanchang University (Nanchang, China). The protocol was approved by the Animal Ethics Committee at Nanchang University.
2.8. Quantification of GA and/or 1,4-GL in apples As mentioned above, GA and/or 1,4-GL was separated and quantified in apples through HILIC-based HPLC-UV. 1,4-GL content in samples from group 2 was further determined through HPLC-MS/ MS. The concentrations of GA and/or 1,4-GL in each sample were calculated with the regression equation using an appropriate dilution factor.
3.1.1. Optimized HPLC separation condition Venusil HILIC (5 lm, 100 A, 4.6 mm 250 mm) and SeQuant ZIC-HILIC (5 lm, 100 A, 250 mm 4.6 mm) columns were used to adequately separate the peaks of GA and 1,4-GL from those of the formulation ingredients. Effective separations and short balance time of the column were observed in analysis with the SeQuant ZIC-HILIC column. The mobile phases were optimized using a mixture of ammonium dihydrogen phosphate and 0.1% of phosphoric acid/acetonitrile as eluent. Ammonium dihydrogen phosphate and phosphoric acid were used because they do not absorb UV–vis.
3.1.2. Method validation The specificity of the method was presented on the chromatogram (Fig. 1) of GA and 1,4-GL (100 lg/mL) solution, apple extract, and apple extract spiked with 50 lg/mL analyte. The peak of the analyte in the apple extract was identified by comparing the retention time with that of the standard. GA and 1,4-GL were properly resolved with few interferences under assay conditions (Fig. 1). The precision of GA and 1,4-GL was evaluated by the QCs analysis expressed as relative standard deviation (RSD), which was calculated from five replicate injections of QCs within one day or in two separate days. Our data showed that the intra- and inter-day precisions were less than 3.5%. The analyte stability in the solution was detected by repeatedly examining the samples during the course of experimentation on the same day and after solution storage (GA or 1,4-GL, 100 lg/mL) for 12, 24, 48, and 72 h at 4 °C. GA and 1,4-GL solutions were stable for 48 h as indicated by the less than 5.0% mean RSD between the peak areas for the samples stored at 4 °C. Therefore, the experiment for quantifying GA and 1,4-GL in apples should be performed within 48 h after preparation and the samples should be stored at 4 °C. Linearity was calculated using linear regression with the least squares method on seven standard solutions from 20 lg/mL to 400 lg/mL. Each level was prepared once, and each solution was injected three times. Linear ranges, linear regression equations, LOD, and LOQ of GA and 1,4-GL are shown in Table 2. Accuracy was based on the recovery of known amounts of analytes, spiked GA, and 1,4-GL in the apple extract. Three samples were prepared in triplicate at three levels over a range of 80% to 120% of the target concentrations. The mean quantitative recoveries (mean ± SD, n = 3) for GA and 1,4-GL were 99.6% ± 2.1% and 101.5% ± 2.4%, respectively.
3.2. Conversion of GA to 1,4-GL in water and in the rat stomach 3.2.1. GA conversion to 1,4-GL in water The developed HPLC–HILIC-UV method enabled the monitoring of GA conversion. Three peaks (i.e., GA, 1,4-GL, and D-glucaro-6,3lactone) were observed in the HPLC-UV chromatogram (Fig. 2) when the conversion reached an equilibrium. At 37 °C, the equilibrium was reached at approximately 48 h and the conversion ratio was approximately 20% and 30% at pH = 7.0 and 1.0, respectively (SFig. 1a). Only 15 min was required to reach equilibrium at 98 °C (SFig. 1b). No differences were observed between the conversion ratio at 37 °C and 98 °C. Previous studies (Ho & Ho, 1990) demonstrated that the equilibrium among GA, 1,4-GL, and D-glucaro-6,3-lactone
is pH dependent; hence, low pH could result
4
B. Xie et al. / Food Chemistry 203 (2016) 1–7
Fig. 1. HPLC-UV (205 nm) chromatograms of apple; (a) apple extract spiked with 1,4-GL and GA; (b) apple extract; and (c) standard solution of 1,4-GL (100 lg/mL) and GA (100 lg/mL).
Table 2 Data for calibration graphs with the developed methods. Method
Linear ranges (lg/mL)
Linear regression equationa
Coefficient (r2)
LOD (lg/mL)
LOQ (lg/mL)
HPLC-UV (GA) (1,4-GL)
20 to 400 20 to 400 2 to 200
A = 0.0848C + 0.6958 A = 0.143C + 0.5917 A = 9.8008C 167.59
0.9993 0.9996 0.9932
12.5 20 1
20 40 5
HPLC-MS/MS (1,4-GL) a
According to A = a C + b, where C = compound concentration, A = peak area.
in generation of high amounts of 1,4-GL. The present results were consistent with these findings. 3.2.2. GA conversion to 1,4-GL in the rat stomach The stomach contents after the rat was administered with GA were analyzed using the HPLC–HILIC-UV method. SFig. 2 shows that high amounts of GA were observed in the stomach 15 min after GA administration, but a low GA content was detected in the sample obtained at 60 min. This finding suggested that the existing time of GA in the rat stomach was lower than 60 min. Only a small amount of GA could also be converted to 1,4-GL (less than 5%) in the stomach, which was consistent with GA conversion in water at pH = 1.0 and 37 °C. 3.3. HPLC-MS/MS method for determining 1,4-GL in apples The methods used in the literature to determine 1,4-GL are gas chromatography (GC) and HPLC. The GC method includes tedious sample preparation (Wang, Gan, Tang, Wang, & Tan, 2010; Yang et al., 2009), whereas the HPLC method with the Spherisorb NH2 column shows poor baseline separation of 1,4-GL (Laakso, Tokola, & Hirvisalo, 1983; Walaszek et al., 1997). In this regard, the present study developed an HILIC-based HPLC-MS/MS method for quantitative analysis of 1,4-GL in apple extracts to confirm 1,4-GL content
in apples. In particular, this study used SRM mode with m/z 191 ? 85 to quantify 1,4-GL. No interference from 1,4-GL was observed under the optimized conditions (Fig. 3). The developed analytical procedure was validated for the quantification of 1,4-GL. Excellent RSD values were obtained for repeatability (<3%) and reproducibility (<5%). The quantitative recoveries of 1,4-GL from the apple samples were measured as 99.4% ± 1.9% (n = 3), which reflects the sufficiently high reliability and accuracy of the method for quantification. The results of linear ranges, linear regression equations, LOD, and LOQ of 1,4-GL are shown in Table 2.
3.4. Contents of GA and 1,4-GL in apples Since UV detection is non-selective with the potential interferences, there is a possibility that co-eluting peaks behind the peak of 1,4-GL when dealing with real sample analysis. Therefore, 1,4GL content in samples from one group (group 2) was further determined through the HPLC-MS/MS method. The results showed that 1,4-GL contents determined using HPLC-MS/MS were 10% higher than those measured through HPLC-UV, suggesting that there is no overlapping components with1,4-GL. In this study, the valleyto-valley integration technique was employed for calculating the peak area of 1.4-GL as the response does not return to the baseline
B. Xie et al. / Food Chemistry 203 (2016) 1–7
5
Fig. 2. HPLC-UV (205 nm) chromatograms of GA conversion in water. (a) GA conversion reached equilibrium at pH = 1.0 and 37 °C; and (b) GA conversion reached equilibrium at pH = 1.0 and 98 °C.
Fig. 3. (a) MS/MS spectra of 1,4-GL; and (b) HPLC-MS/MS chromatogram of apples in SRM mode with m/z 191 ? 85 separated on the HILIC column.
between peaks (Fig. 1), which may be the reason for about 10% lower contents of 1.4-GL determined by HPLC-UV. Therefore, we supposed that both methods were reliable and robust to quantify 1,4-GL in apples. Although the HPLC-MS/MS method is more specific and sensitive to low LOD and LOQ than HPLC-UV, it requires the use of expensive mass spectrometers. The HPLC–HILIC-UV method was validated to be robust to allow its routine laboratory use for simultaneous measurement of GA and 1,4-GL in apples. Therefore, the method was applied to analyze three apple varieties that originated from five regions. The contents of GA and 1,4-GL in apples are presented in Table 1. 4. Discussion Apple is one of the most wide-spread fruits eaten daily by millions of people worldwide. 1,4-GL and GA are beneficial for general
health promotion (‘‘Cancer, Lifestyle Modification and Glucarate”, 2001). Two papers published in the last two decades described and quantified GA content in apples by using pyruvate (Walaszek et al., 1996) and b-glucuronidase inhibition methods (Dwivedi et al., 1990). As GA exhibits high polarity, its retention time on the C18 column is very short to separate interfering substances in apples. The HILIC column exhibited advantages in terms of retention and selectivity, which provided various ingredients of apples with a good separation of GA and 1,4-GL. As high proportion of organic solvent is required in sampling and in the mobile phase of the HILIC system, the supernatants from apple extraction can be directly injected into the HPLC system by adding a suitable volume of methanol. Thus, the procedure for sample preparation is simple. Collectively, these approaches can serve as a novel method for detecting GA derivatives in fruit and vegetables and are more
6
B. Xie et al. / Food Chemistry 203 (2016) 1–7
highly specific and sensitive than enzyme methods. Unlike HPLCMS/MS, the HPLC-UV method is stable, reproducible, and has practical advantages, such as low cost for simultaneous measurement of GA and 1,4-GL in apples. Heerdt et al. (1995) supposed that GA can be converted to 1,4-GL and obtained an equilibrium that consists of approximately 30% 1,4-GL and 30% D-glucaro-6,3-lactone in the acidic environment of the stomach. However, they did not describe the experimental process and analytical method for GA conversion in the stomach. The present study found that the conversion of GA to 1,4-GL to reach the equilibrium (conversion ratio of approximately 30%) was slow (approximately 48 h) at pH = 1.0 and 37 °C in water. In addition, the existing GA time in the rat stomach was lower than 60 min and only less than 5% GA can be converted to 1,4-GL in the stomach. Similar results were obtained in a previous study, in which potassium hydrogen D-[14C]glucarate was orally administered to rats and analyzed using HPLC and radioactive technology (Walaszek et al., 1997). 1,4-GL is a specific and potent inhibitor of b-glucuronidase compared with GA and D-glucaro-6,3-lactone (Kiyomoto, Harigaya, Ohshima, & Morita, 1963). Walaszek et al. (1997) demonstrated that the formation of 1,4-GL from GA can be the prerequisite to exert their benefits in general health promotion. Thus, the determination of 1,4-GL in apples plays a key role in quality evaluation of apples. Our data showed that 1,4-GL contents in apples ranged from 0.3 mg/g to 0.9 mg/g. To the best of our knowledge, this study is the first to present quantitative information about 1,4-GL in apples. Previous study (Kiyomoto et al., 1963) showed that hepatic b-glucuronidase activity caused 50% inhibition by single administration of 200 mg/kg 1,4-GL in rats. Therefore, we believe that the enterohepatic circulation of many drugs, toxins, steroids, and other compounds subject to glucuronidation can be modified if a person can consume 250 g of apple daily. Our findings provide key data to substantiate the positive effects of apples on general health promotion. The contents of 1,4-GL (x-axis) compared with those of GA (yaxis) in apples are plotted in Fig. 4 to determine the relationship between the samples. The plot shows that the samples can be classified into three classes. Samples from Fuji and Red Delicious formed a group, whereas Green Delicious and Gala were completely separated from each other in the plot. The classification of these samples suggested that the GA derivatives that constituted apple possibly differed among varieties. Approximately 1.5 times higher content of 1,4-GL was observed in Gala apples than that
1200
Variety Group 1 Fiji Group 2 Gala Group 3 Red Delicious Group 4 Fuji Group 5 Fuji Group 6 Green delicious Group 7 Green delicious
GA contents in apples (ug/g)
1000
800
600
400
200
300
350
400
450
500
550
600
650
700
750
1,4-GL contents in apples (ug/g)
Fig. 4. Plot that shows 1,4-GL and GA contents in apples.
800
850
in the other varieties. Furthermore, Table 1 shows that the ratio of 1,4-GL to GA in the Green Delicious group was significantly lower than that in the Red Delicious and other groups. Previous studies demonstrated that the concentration of esters could increase during climacteric ripening and reached maximum values at the climacteric peak (Guo, Yue, & Yuan, 2012; Ortiz, Graell, & Lara, 2011). Apples in the Green Delicious group were physiologically immature and produced lower concentrations of 1,4-GL. Therefore, our data indicated that the state of fruit maturity at harvest plays an important role in 1,4-GL contents in apples. 5. Conclusion An HILIC-based HPLC method coupled with UV or MS was developed and validated to quantify GA and/or 1,4-GL in three apple varieties that originated from five regions. The methods are highly recommended for the quality assessment of different GA derivative samples, including new, commercial juice, and pharmaceutical apple products. As the formation of 1,4-GL from GA can be the prerequisite to exert health benefits in vivo, GA conversion to 1,4-GL was investigated in water and rat stomach. Only a small portion of GA could be transformed to 1,4-GL in the rat stomach after GA oral administration. Furthermore, our data revealed that the variety and maturity of apples at harvest are factors that determine 1,4-GL contents and can be used to improve the nutritional quality of apples. Conflict of interest The authors have no conflict of interest. Acknowledgements The work was supported by a grant from the Natural Science Foundation of China (No. 81560631), China Postdoctoral Science Foundation (No. 143750), Postdoctoral Science Foundation of Jiangxi Province (No. 2015KY41), and the Innovation Fund Designated for Graduate Students of Jiangxi Province of China (No. YC2014-S076 and cx2015205). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.foodchem.2016. 02.009. References Alpert, A. J. (1990). Hydrophilic-interaction chromatography for the separation of peptides, nucleic acids and other polar compounds. Journal of Chromatography, 499, 177–196. Bertolini, T., Vicentini, L., Boschetti, S., Andreatta, P., & Gatti, R. (2014). A novel automated hydrophilic interaction liquid chromatography method using diodearray detector/electrospray ionization tandem mass spectrometry for analysis of sodium risedronate and related degradation products in pharmaceuticals. Journal of Chromatography A, 1365, 131–139. Bespalov, V. G., & Aleksandrov, V. A. (2012). Anticarcinogenic effect of potassium salts of glucaric and glucuronic acid in induced models of cervical and esophageal tumors. Voprosy Onkologii, 58(4), 537–540. Buck, K., Voehringer, P., & Ferger, B. (2009). Rapid analysis of GABA and glutamate in microdialysis samples using high performance liquid chromatography and tandem mass spectrometry. Journal of Neuroscience Methods, 182(1), 78–84. Cancer, Lifestyle Modification and Glucarate (2001). Chirita, R. I., West, C., Finaru, A. L., & Elfakir, C. (2010). Approach to hydrophilic interaction chromatography column selection: Application to neurotransmitters analysis. Journal of Chromatography A, 1217(18), 3091–3104. Cornille, A., Gladieux, P., Smulders, M. J., Roldan-Ruiz, I., Laurens, F., Le Cam, B., ... Giraud, T. (2012). New insight into the history of domesticated apple: Secondary contribution of the European wild apple to the genome of cultivated varieties. PLoS Genetics, 8(5), e1002703.
B. Xie et al. / Food Chemistry 203 (2016) 1–7 Dwivedi, C., Heck, W. J., Downie, A. A., Larroya, S., & Webb, T. E. (1990). Effect of calcium glucarate on beta-glucuronidase activity and glucarate content of certain vegetables and fruits. Biochemical Medicine and Metabolic Biology, 43(2), 83–92. Guo, J., Yue, T., & Yuan, Y. (2012). Feature selection and recognition from nonspecific volatile profiles for discrimination of apple juices according to variety and geographical origin. Journal of Food Science, 77(10), C1090–C1096. Heerdt, A. S., Young, C. W., & Borgen, P. I. (1995). Calcium glucarate as a chemopreventive agent in breast cancer. Israel Journal of Medical Sciences, 31 (2–3), 101–105. Ho, Y. D., & Ho, K. J. (1990). Biliary D-glucaric acid: Its quantitation and preventive role in gallstone formation. Scandinavian Journal of Gastroenterology, 25(6), 631–640. Jian, W. Y., Edom, R. W., Xu, Y. D., & Weng, N. D. (2010). Recent advances in application of hydrophilic interaction chromatography for quantitative bioanalysis. Journal of Separation Science, 33(6–7), 681–697. Kiyomoto, A., Harigaya, S., Ohshima, S., & Morita, T. (1963). Studies on glucosaccharolactones. I. Glucosaccharo-1:4-lactone. Biochemical Pharmacology, 12, 105–112. Laakso, E. I., Tokola, R. A., & Hirvisalo, E. L. (1983). Determination of D-glucaric acid by high-performance liquid chromatography. Journal of Chromatography, 278(2), 406–411. Narita, M., Nagai, E., Hagiwara, H., Aburada, M., Yokoi, T., & Kamataki, T. (1993). Inhibition of beta-glucuronidase by natural glucuronides of kampo medicines using glucuronide of SN-38 (7-ethyl-10-hydroxycamptothecin) as a substrate. Xenobiotica, 23(1), 5–10. Ortiz, A., Graell, J., & Lara, I. (2011). Volatile ester-synthesising capacity throughout on-tree maturation of ’Golden Reinders’ apples. Scientia Horticulturae, 131, 6–14. Perez, J. L., Jayaprakasha, G. K., Yoo, K. S., & Patil, B. S. (2008). Development of a method for the quantification of D-glucaric acid in different varieties of grapefruits by high-performance liquid chromatography and mass spectra. Journal of Chromatography A, 1190(1–2), 394–397. Poon, R., Villeneuve, D. C., Chu, I., & Kinach, R. (1993). HPLC determination of Dglucaric acid in human urine. Journal of Analytical Toxicology, 17(3), 146–150.
7
Saluk-Juszczak, J., Olas, B., Nowak, P., Staron, A., & Wachowicz, B. (2008). Protective effects of D-glucaro-1,4-lactone against oxidative modifications in blood platelets. Nutrition, Metabolism and Cardiovascular Diseases, 18(6), 422–428. Simone, Charles B., Simone, Nicole L., Pallante, Michael, & Simone, Charles B. (2001). Cancer, Lifestyle Modification and Glucarate. J. Orthomol. Med., 16(2), 83–90. Walaszek, Z., Hanausek, M., Adams, A. K., & Sherman, U. (1991). Cholesterollowering effects of dietary D-Glucarate. Faseb Journal, 5(5). A930-A930. Walaszek, Z., Szemraj, J., Hanausek, M., Adams, A. K., & Sherman, U. (1996). DGlucaric acid content of various fruits and vegetables and cholesterol-lowering effects of dietary D-glucarate in the rat. Nutrition Research, 16(4), 673–681. Walaszek, Z., Szemraj, J., Narog, M., Adams, A. K., Kilgore, J., Sherman, U., & Hanausek, M. (1997). Metabolism, uptake, and excretion of a D-glucaric acid salt and its potential use in cancer prevention. Cancer Detection and Prevention, 21 (2), 178–190. Wang, K., Gan, X., Tang, X., Wang, S., & Tan, H. (2010). Determination of D-saccharic acid-1,4-lactone from brewed kombucha broth by high-performance capillary electrophoresis. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci., 878(3–4), 371–374. Xie, B., Liu, A., Zhan, X., Ye, X., & Wei, J. (2014). Alteration of gut bacteria and metabolomes after glucaro-1,4-lactone treatment contributes to the prevention of hypercholesterolemia. Journal of Agriculture and Food Chemistry, 62(30), 7444–7451. Yang, Z.-W., Ji, B.-P., Zhou, F., Li, B., Luo, Y., Yang, L., & Li, T. (2009). Hypocholesterolaemic and antioxidant effects of kombucha tea in highcholesterol fed mice. Journal of the Science of Food and Agriculture, 89(1), 150–156. Yoshida, T., & Okada, T. (1999). Peptide separation in normal-phase liquid chromatography - Study of selectivity and mobile phase effects on various columns. Journal of Chromatography A, 840(1), 1–9. Zitka, O., Heger, Z., Kominkova, M., Skalickova, S., Krizkova, S., Adam, V., & Kizek, R. (2014). Preconcentration based on paramagnetic microparticles for the separation of sarcosine using hydrophilic interaction liquid chromatography coupled with coulometric detection. Journal of Separation Science, 37(5), 465–575. Zoltaszek, R., Hanausek, M., Kilianska, Z. M., & Walaszek, Z. (2008). The biological role of D-glucaric acid and its derivatives: Potential use in medicine. Postepy Hig Med Dosw (Online), 62, 451–462.