Separation and Purification Technology 174 (2017) 338–344

Contents lists available at ScienceDirect

Separation and Purification Technology

journal homepage: www.elsevier.com/locate/seppur

Enhancing the recovery of cabbage through the monitoring of content and myrosinase activity during extraction by different methods ⇑ Patsaporn Pongmalai a, Sakamon Devahastin a, , Naphaporn Chiewchan a, Somchart Soponronnarit b a Advanced Food Processing Research Laboratory, Department of Food Engineering, Faculty of Engineering, King Mongkut’s University of Technology Thonburi, 126 Pracha u-tid Road, Tungkru, Bangkok 10140, Thailand b Division of Energy Technology, School of Energy, Environment and Materials, King Mongkut’s University of Technology Thonburi, 126 Pracha u-tid Road, Tungkru, Bangkok 10140, Thailand article info abstract

Article history: Since endogenous hydrolysis of glucoraphanin into sulforaphane is inefficient and difficult to control, a Received 16 June 2016 means to effectively extract glucoraphanin from its source, so that exogenous conversion can be later Received in revised form 26 October 2016 performed, is desired. In this study, selected extraction methods, i.e., combined ultrasound-assisted Accepted 4 November 2016 extraction and microwave-assisted extraction (UAE + MAE), vacuum microwave-assisted extraction Available online 5 November 2016 (VMAE), UAE + VMAE and Soxhlet extraction, with water as an extraction solvent, were employed to extract glucoraphanin from cabbage outer leaves, with the aim to enhance the extraction yield and at Keywords: the same time minimize the loss of the compound. Use of fresh versus steamed leaves for the extraction Enzymatic hydrolysis was also compared. Evolutions of the cabbage-water temperature as well as glucoraphanin and sul- Microwave foraphane contents of the extracts along with the myrosinase activity were monitored. UAE + VMAE Steaming led to a significantly higher content of glucoraphanin than UAE + MAE and VMAE; Soxhlet extraction Ultrasound resulted in a lower glucoraphanin content and required a much longer time. Nevertheless, UAE Temperature evolution + VMAE could not rapidly inactivate myrosinase in the fresh leaves, resulting in some conversion of glu- coraphanin into sulforaphane, which suffered degradation during the extraction. UAE + VMAE of steamed cabbages proved to yield the highest maximum content of glucoraphanin, with the value of 650.09 lmol/100 g dry weight, 87% higher than that obtained via the use of the fresh cabbages in combination with UAE + VMAE. Simple kinetic modeling of the extraction processes was also attempted. Ó 2016 Elsevier B.V. All rights reserved.

1. Introduction [6–8]. It is reported, however, that endogenous myrosinase is typ- ically inefficient in converting glucoraphanin into sulforaphane. Glucosinolates are a group of sulfur-containing plant secondary Jeffery and Stewart [9], for example, reported that endogenous metabolites that occur in Brassica vegetables [1]. These compounds myrosinase in homogenized could convert only 25% (or are naturally stable but biologically inactive. Glucosinolates can be lower) of glucoraphanin into sulforaphane. This implies that direct divided into three classes based on the structure of different amino intake of Brassica vegetables may not be sufficient to deliver the acid precursors, including aliphatic, aromatic and glucosino- intended health benefits as glucoraphanin is still in its native form. lates [2]. Glucoraphanin is well recognized as an aliphatic glucosi- Extraction of glucoraphanin from its source and addition of nolate found in Brassica vegetables. When plant tissues are exogenous myrosinase to hydrolyze glucoraphanin into sul- damaged, glucoraphanin is hydrolyzed by endogenous myrosinase, foraphane in a more effective and controllable manner should which is located in plant myrosin cells, to form sulforaphane [3]. therefore be conducted. Sulforaphane is an derivative formed from To achieve the aforementioned goal, evolutions of the gluco- glucoraphanin and is well known to possess high anticarcinogenic and sulforaphane contents as well as myrosinase activity activity [4,5]. Many studies have therefore been devoted to the during extraction need to be monitored. This information can be production and extraction of this compound from its source used to design an effective means to extract glucoraphanin without causing much degradation to the compound, either by thermal ⇑ Corresponding author. degradation or unintentionally hydrolyzing it into sulforaphane, E-mail address: [email protected] (S. Devahastin). which is very heat-sensitive and hence easily degradable and not http://dx.doi.org/10.1016/j.seppur.2016.11.003 1383-5866/Ó 2016 Elsevier B.V. All rights reserved. P. Pongmalai et al. / Separation and Purification Technology 174 (2017) 338–344 339 recoverable. Studies indeed exist on this kind of information. place. Such one-step procedure, if successfully validated, could Ghawi et al. [10], for example, investigated the effect of thermal help reduce the complexity, time and energy consumption of the treatment (at 30–100 °C) on the changes of glucoraphanin and whole process as no additional step to inactivate the native sulforaphane contents of broccoli; myrosinase activity was also enzyme would be required prior to the extraction. Cabbage outer determined. Thermal treatment at 80 °C for 4, 8 and 12 min led leaves, which are low-value residues of the vegetable processing to an increase in the glucoraphanin content in broccoli. However, industry but contain a significant amount of glucoraphanin, were glucoraphanin degraded at 100 °C due to thermal degradation. used as the test raw material. For comparison, steamed cabbage Myrosinase activity remained rather constant when mild-heat leaves, which no longer contained myrosinase of any significant treatment at 50 °C for 4 min was applied. A sudden decrease in activity, were allowed to undergo the extraction. The results the activity nevertheless occurred at higher temperatures obtained via Soxhlet extraction were also collected and compared. (50–100 °C), leading to a lower sulforaphane content in boiled broccoli. Pérez et al. [7] later reported that myrosinase activity 2. Materials and methods remained even after higher-temperature blanching (70–74 °C). Although the hydrolysis of glucoraphanin at a higher temperature 2.1. Materials (70–74 °C) took place, leading to sulforaphane formation, the con- tent of this latter compound was very low due to its high thermal Outer leaves of cabbage (Brassica oleracea L. var. capitata) were sensitivity. No information is so far available, however, on the evo- obtained from Pakklong Talad market in Bangkok; the leaves were lutions of glucoraphanin and sulforaphane contents as well as kept at 4 °C until the time of an experiment. Before starting of each myrosinase activity during extraction. experiment, the leaves were washed with tap water and drained In terms of extraction, a number of advanced techniques have on a screen to get rid of excess water. The leaves were then been applied to extract plant bioactive compounds from their chopped with an electric chopper (Moulinex, DPA141, Écully, sources. Ultrasound-assisted extraction (UAE) has recently France) for 2 min to obtain an average size of cabbages of received much attention due to its ability to modify plant structure 1.7–2.5 mm. The chopped cabbages were immediately introduced via acoustic cavitation; disruption of plant cell walls results in to an extraction process. enhanced release of internal cell compounds into an extraction solvent [11,12]. Microwave-assisted extraction (MAE) is also of interest as it helps induce rapid heating within plant cells, leading 2.2. Methods to disruption of cellular structure and hence increased extraction yield [8,13]. Pongmalai et al. [12] indeed combined UAE and MAE 2.2.1. UAE, MAE and VMAE to enhance the extraction of bioactive compounds from cabbages. Five g of chopped cabbages was dispersed in 50 mL of deionized UAE + MAE was noted to result in higher contents of extractable water (DI water), which was used as an extraction solvent, in a bioactive compounds than either UAE or MAE alone; the enhance- 250-mL beaker; the whole content was then placed in an ultra- ment is due to the combined effect of acoustic cavitation and rapid sonic bath (Elma, Elmasonic P, Singen, Germany) containing 1 L heating within the plant cells by microwave irradiation. On the of distilled water. UAE was performed by sonicating the mixture other hand, Hiranvarachat et al. [14], who monitored the evolu- of cabbages and DI water at a frequency of 37 kHz and a set power tions of extractable triterpene saponins and phenolics contents of 320 W (or absorbed ultrasonic power of 0.03 W/g of the mixture from Centella asiatica leaves during MAE and vacuum microwave- of cabbages and DI water) for 40 min; this condition was selected assisted extraction (VMAE), confirmed the benefit of VMAE over from the preliminary experiments as the value that led to the high- MAE, if the former is operated at an optimum condition. VMAE out- est extractable glucoraphanin content (data not shown). performed MAE due to the ability of the former to perform extrac- For combined extraction methods, i.e., UAE + MAE and UAE tion at a lower temperature as a result of the reduced boiling point + VMAE, an ultrasonically treated mixture was placed in a round- of an extraction solvent at a lower pressure. bottom flask and then subject to microwave irradiation at 180 W. In addition to their ability to enhance extraction, ultrasound MAE and VMAE were performed in the same domestic microwave and microwave have noted to help inactivate some enzymes dur- oven (Samsung, GE-872D, Port Klang, Malaysia), which was modi- ing processing. Benlloch-Tinoco et al. [15], for example, studied fied as detailed in Hiranvarachat et al. [19]. The microwave power the effect of microwave on the activity of enzyme in kiwifruit of 180 W was selected to prevent excessive boiling of the extrac- puree and found that microwave heating at a higher power with tion solvent at the tested vacuum pressure [9]. VMAE was con- a shorter treatment time was more effective in decreasing the ducted at an absolute pressure of 70 kPa as this pressure led to enzyme activity than heating in a water bath. Lopes et al. [16] con- the highest recoverable glucoraphanin content in comparison with firmed that microwave treatment led to a more extensive inactiva- those at the other tested pressures, i.e., 30 and 50 kPa (data not tion of horseradish peroxidase than heating in a thermostatic bath. shown). The duration of both MAE and VMAE was maintained at This is because the functional groups of the enzymes directly 10 min; this time was selected to deliberately extend the extrac- absorb microwave irradiation, leading to rapid heating and hence tion to the point where myrosinase was completely inactivated the destruction of the enzyme structure. In terms of the ability of as will be later discussed. VMAE without the prior use of UAE ultrasound to inactivate enzymes, Cheng et al. [17] observed the was also conducted; a sample was prepared in the same manner ability of ultrasound to help inactivate polyphenol oxidase (PPO) and subject directly to VMAE. in mushroom (Agaricus bisporus) during thermal processing. More The temperature evolution of the cabbage-water mixture dur- recently, Sulaiman et al. [18] also observed the effect of ultrasound ing UAE was measured using a type-T thermocouple, which was on PPO inactivation in fruit purees; ultrasonication is effective in inserted into a beaker containing the mixture and connected to a reducing the PPO activity due to the formation and collapse of data logger (Yokogawa, DX112, Tokyo, Japan). A fiber-optic ther- microbubbles that affect the enzyme structure. mometer (Luxtron, m600, Santa Clara, CA) was used to monitor Based on the above-mentioned arguments, it is interesting to the mixture temperature during MAE and VMAE; the thermometer determine if UAE + MAE, VMAE and UAE + VMAE can be used to was placed in the center of the flask containing the mixture. The extract glucoraphanin and at the same time inactivate myrosinase, specific absorbed microwave powers in the cases of UAE + MAE, so negligible hydrolysis of this compound into sulforaphane, which UAE + VMAE and VMAE were 1.3, 1.3 and 1.19 W/g, respectively. is very easily degradable and thus difficult to retain, would take Specific absorbed microwave powers, which are the powers 340 P. Pongmalai et al. / Separation and Purification Technology 174 (2017) 338–344 absorbed by a unit mass of the cabbages and DI water, were eval- 2.2.4. Determination of sulforaphane content uated according to the method of Hiranvarachat and Devahastin 2.2.4.1. Sample extraction from fresh cabbages. The extraction of sul- [20]. foraphane from fresh cabbages was conducted using the method Inactivation of endogenous myrosinase in fresh cabbages prior described by Tanongkankit et al. [22]. Five g of fresh cabbages to extraction was performed in selected cases. Whole leaves of was extracted two times with 50 mL of dichloromethane, which cabbages were steamed on a perforated tray over boiling water was combined with 2.5 g of sodium sulfate anhydrous. The dichlor- for 2 min; our preliminary study showed that this steaming omethane fraction was dehydrated using the rotary evaporator at duration was adequate to inactivate myrosinase, polyphenol 50 °C for 10 min. The residue was then dissolved in 2 mL of oxidase and peroxidase in the cabbages (data not shown). After acetonitrile. A diluted sample was kept at 18 °C in a vial until steaming the leaves were immediately cooled in cold water at further analysis. 4 °C. The steamed samples were then prepared and extracted following the procedures described earlier. 2.2.4.2. Quantification of sulforaphane. The determination of the After extraction an extract was filtered through a Whatman sulforaphane content was performed following the method of No. 1 filter paper, which is made from cellulose and has the pore Pongmalai et al. [12]. Two mL of an extract was introduced to Oasis l size of around 10 m. The filtrate was concentrated using a rotary HLB, 3 cc cartridge (Waters, Milford, MA). The eluate was purged evaporator (Buchi, R-215, Flawil, Switzerland) at 50 °C for 30 min with N2 for 10 min and dissolved with 0.5 mL of 1% (v/v) acetic before being diluted with another solvent; the type and amount acid. The whole content was filtered through a 0.2-lm nylon filter. of the added solvent depended on the required subsequent chem- Ten lL of the filtrate was injected into Xselect CSH C18 HPLC ical analysis. One mL of methanol and 2 mL of acetonitrile was column (4.6 250 mm) (Waters, Milford, MA) with 15% of acetoni- added, respectively, for glucoraphanin and sulforaphane analysis. trile and 85% of 1% (v/v) acetic acid as a mobile phase; the flow rate ° A diluted sample was kept at 18 C in a vial until further analysis. was set at 1.2 mL/min. A UV detector at a wavelength of 254 nm was used for detecting sulforaphane. The sulforaphane content was calculated from a standard curve of sulforaphane (Sigma- 2.2.2. Soxhlet extraction Aldrich, St. Louis, MO), which was prepared at the concentrations Five g of chopped cabbages was filled in a thimble made of cel- of 0–50 lg/mL. lulose with the pore size of around 10 lm. The thimble was then inserted into a Soxhlet extractor connected with an accurately weighed round bottom flask containing 250 mL of DI water. This 2.2.5. Determination of myrosinase activity volume was selected as the minimum volume required to allow 2.2.5.1. Preparation of enzyme extract from fresh cabbages. Myrosi- the reflux of DI water in the extractor. The water was refluxed at nase activity was determined through a decrease in the a temperature of 100 °C for 4 h using an electric heating mantle content via the enzymatic hydrolysis reaction as recommended (M TOPs, MS-E103, Kyunggi-do, Korea). The temperature of the by Charron et al. [23]. Fresh cabbages were cut into 1 1 cm using cabbage-water mixture was monitored using a type-T thermocou- a stainless steel knife. Twenty-five g of the sample was placed in a ple connected to the data logger. An extract in the bottom flask was blender (Waring, 32BL80, New Hartford, CT) with 50 mL of 30-mM filtered through a Whatman No. 1 filter paper; the filtrate was citrate/phosphate buffer (pH 7) and blended for 1 min. The mixture concentrated using the rotary evaporator at 50 °C for 2.5 h. The was filtered through a 425-lm stainless steel screen; the filtrate content was diluted with different extraction solvent, which was was then placed into a 16 75-mm test tube and centrifuged by again added depending on the type of the required chemical anal- a refrigerated centrifuge (Hitachi, Himax CR22 N, Ibaraki, Japan) ysis. A diluted sample was kept at 18 °C in a vial until further at 30,000g at 4 °C for 4 min. Supernatant from the tube was again analysis. filtered; the filtrate was kept at 4 °C until further analysis.

2.2.3. Determination of glucoraphanin content 2.2.5.2. Quantification of myrosinase activity. A reaction mixture l 2.2.3.1. Sample extraction from fresh cabbages. Glucoraphanin con- containing 50 L of the above-mentioned filtrate, 1.35 mL of tent was determined using the recommended method of Lee 32.22-mM citrate/phosphate buffer (pH 6.5) with 1.07-mM EDTA l et al. [21] with some modifications. Five g of fresh cabbages was and 100 L of 37.50-mM sinigrin was prepared in a 1 10-cm test ° l blended with 50 mL of methanol for 1 min; the mixture was then tube at 25 C. One-hundred L of DI water was used as a reference stirred in a stirring water bath (Major Science, SWB-10L-1, Sara- mixture instead of the sinigrin solution. The tube was mixed by a toga, CA) at 70 °C for 15 min. The mixture was cooled to room tem- vortex mixer (Scientific Industries, model G-560, Bohemia, NY) perature and filtered through a Whatman No. 1 filter paper and for 30 s. Both the sample mixture and reference solution were washed with 50 mL of methanol. The methanol fraction was dehy- placed into a pair of cuvettes and had their absorbance measured drated using the rotary evaporator at 50 °C for 30 min. The residue at 227 nm at every 5-s interval for a period of 5 min after the mix- was dissolved in 1 mL of methanol. A diluted sample was kept at ing. Myrosinase activity was determined from the linear slope of a 18 °C in a vial until further analysis. plot between the absorbance of sinigrin and the reaction time, which represents the disappearance of sinigrin from the reaction mixture. A unit (U) of activity is defined as the disappearance of 2.2.3.2. Quantification of glucoraphanin. One mL of an extract was 1 lmol sinigrin per min. first introduced to Sep-PakÒ Vac 6 cc cartridge (Waters, Milford, MA). The eluate was then filtered through a 0.2-lm nylon filter. l 2.2.6. Kinetic modeling of bioactive compounds contents and Ten L of the filtrate was injected into Xselect CSH C18 HPLC col- myrosinase activity umn (4.6 250 mm) (Waters, Milford, MA) with 30% of methanol The evolutions of the normalized bioactive compounds contents and 70% of 30-mM ammonium acetate buffer (pH 5) as a mobile and myrosinase activity were fitted to the following kinetic phase; the flow rate was set at 1 mL/min. A UV detector at a equation: wavelength of 233 nm was used for detecting glucoraphanin. The glucoraphanin content was calculated from a standard curve of Ct Ceq n glucoraphanin potassium salt (Chromadex, Irvine, CA), which was ¼ exp ðktÞ ð1Þ C0 Ceq prepared at the concentrations of 0–1000 lg/mL. P. Pongmalai et al. / Separation and Purification Technology 174 (2017) 338–344 341

where Ct is the normalized bioactive compound content and the process; the evolutions during the 40 min of UAE are not myrosinase activity at any instant, C0 is an initial normalized bioac- shown. The initial normalized glucoraphanin content was around tive compound content and myrosinase activity, Ceq is an equilib- 0.12 (or 12% of the content extractable from the fresh leaves as rium normalized bioactive compound content and myrosinase per the method described in Section 2.2.3.1), while the initial tem- activity, k is the rate constant and n is the order of the evolution perature of the mixture was 48 °C. The mixture temperature kinetics. dropped from 61 °C right after UAE to 48 °C although MAE started almost immediately after UAE. The rather low value of initial nor- 2.2.7. Statistical analysis malized glucoraphanin content is attributed to the fact that The experimental data were subject to the analysis of variance myrosinase was still active during UAE; the normalized myrosi- (ANOVA) and are presented as mean values with standard devia- nase activity after UAE was as high as 0.30. Although some studies tions. Differences between the mean values were established using have reported that ultrasonic treatment could help inactivate Duncan’s new multiple range tests; the differences were consid- enzymes, UAE could not completely inactivate endogenous ered at a confidence level of 95%. All statistical analyses were per- myrosinase; the enzyme could still act as a catalyst for the hydrol- formed using SPSSÒ software (version 17, SPSS Inc., Chicago, IL). All ysis of glucoraphanin into sulforaphane. The initial normalized sul- experiments were performed at least in duplicate. foraphane content in this case was around 0.29. During MAE the glucoraphanin content increased until reaching 3. Results and discussion its maximum value at 4 min. Subjecting the mixture to microwave irradiation led to the disruption of cabbage cell structure due to To reduce the variability of the experimental data as a result of rapid internal heating; cell rupture in turn resulted in enhanced the variation in the raw material, normalized contents of all bioac- glucoraphanin release from the cabbage matrix [12]. An increase tive compounds and myrosinase activity are reported in this study. in the glucoraphanin content could be observed despite the The normalized glucoraphanin and sulforaphane contents are remaining myrosinase activity. Although some amount of gluco- defined as the contents at any instant to those extractable from raphanin might be hydrolyzed by the remaining enzyme, gluco- the fresh leaves. Normalized myrosinase activity is defined as the raphanin release from the matrix must have taken place at a activity at any instant to that of the extract from the fresh leaves. much higher rate than the enzymatic hydrolysis. The loss of myrosinase activity was observed to be around 86% 3.1. UAE + MAE at 4 min of MAE. The decrease is due to the rapid increase in the mixture temperature (from 48 to 96 °C) within this period. Note The evolutions of the normalized bioactive compounds contents that white cabbage myrosinase is a glycoprotein and is stable up ° and myrosinase activity as well as the change of the mixture (cab- to only around 50 C [24,25]. Due to this rapid loss of the myrosi- bages and DI water) temperature are illustrated in Fig. 1a. Note nase activity, conversion of glucoraphanin into sulforaphane was that the results shown in this figure represent the MAE part of suppressed; normalized sulforaphane content was only 0.13 at

1.6 140 1.6 140 a) c) 1.4 120 1.4 120 1.2 1.2 100 100 1 1 80 80 0.8 0.8 60 60 0.6 0.6 Temperature (°C) 40 40 Temperature (°C) 0.4 0.4 20

contents and myrosinase activity contents and myrosinase 20 Normalized bioactive compounds bioactive Normalized 0.2 activity contents and myrosinase

Normalized bioactive compounds bioactive Normalized 0.2

0 0 0 0 0246810 0246810 Time (min) Time (min)

1.6 140 1.6 140 b) d) 1.4 120 1.4 120 1.2 1.2 100 100 1 1 80 80 0.8 0.8 60 60 0.6 0.6 Temperature (°C) 40 Temperature (°C) 40 0.4 0.4 20 20 activity contents and myrosinase Normalized bioactive compounds bioactive Normalized 0.2 contents and myrosinase activity contents and myrosinase

Normalized bioactive compounds bioactive Normalized 0.2

0 0 0 0 0246810 01234 Time (min) Time (h)

Fig. 1. Evolutions of normalized bioactive compounds contents, myrosinase activity and temperature during different extraction methods. (a) MAE (as part of UAE + MAE); (b) VMAE; (c) VMAE (as part of UAE + VMAE) and (d) Soxhlet extraction. (s) Glucoraphanin content; (4) Sulforaphane content; (d) Myrosinase activity and (h) Mixture (cabbages + DI water) temperature. 342 P. Pongmalai et al. / Separation and Purification Technology 174 (2017) 338–344 the end of this 4-min period. It is important to note that in addition help create suction across the porous structure of the sample [26], to the limited formation of sulforaphane, the compound also suf- leading to the more extensive flow of the extraction solvent fered degradation as the mixture temperature went far beyond through the structure and hence the more enhanced transport of its degradation temperature of around 52 °C [12]. glucoraphanin. The loss of myrosinase activity was around 87% After 4 min of MAE, glucoraphanin content decreased toward after VMAE at 90 °C for 4 min. Since some enzyme activity still the end of the extraction when the mixture temperature increased remained, hydrolysis of glucoraphanin into sulforaphane pro- to even higher than 96 °C, which is most probably the degradation ceeded. Normalized sulforaphane content was observed to be temperature of our cabbage glucoraphanin. Normalized sul- around 0.15 at 4 min of the extraction. foraphane content and myrosinase activity also decreased. No After 4 min the glucoraphanin content remained unchanged, myrosinase activity was observed after 6 min of MAE (mixture with the mixture temperature of around 90 °C. The normalized temperature was approximately 99 °C), rendering no further myrosinase activity after 4 min decreased toward the end of the hydrolysis of glucoraphanin into sulforaphane. extraction; this enzyme was completely inactivated at around 8 min of the extraction. As expected, the normalized sulforaphane 3.2. VMAE content also decreased toward the end of the extraction because of its degradation at a higher temperature and the lack of the myrosi- Evolutions of the normalized glucoraphanin and sulforaphane nase activity to further drive the hydrolysis reaction. contents and myrosinase activity as well as the mixture tempera- ture during VMAE are illustrated in Fig. 1b. Normalized gluco- 3.4. Soxhlet extraction raphanin content continuously increased until reaching the maximum value (of around 0.15) at 8 min of the extraction; the Fig. 1d presents the results on the normalized contents of bioac- content then remained almost unchanged until the end of the tive compounds and myrosinase activity during Soxhlet extraction. extraction. The maximum content of glucoraphanin extractable The normalized glucoraphanin content of around 0.12 was via VMAE was lower than those obtained via UAE + MAE at observed at 1 h of the extraction. After 1 h the content of this com- 4 min. This lower maximum value was observed despite the fact pound decreased toward the end of the extraction. The decrease in that vacuum could help lower the boiling temperature of the sol- the glucoraphanin content was due to the hydrolysis reaction by vent as well as the oxygen content in the system and should have endogenous myrosinase as well as thermal degradation of this therefore led to reduced thermal and oxidative degradation of the compound during its collection in the bottom flask. Slight myrosi- compound. Nevertheless, VMAE most probably led to less exten- nase activity was detected after 1 h when the mixture temperature sive modification of the cabbage structure than UAE + MAE, result- was around 100 °C, after which the activity was no longer detected. ing in less release of the compounds into the extraction solvent. Accordingly, small amount of sulforaphane was detected after 1 h The normalized myrosinase activity decreased when the mix- due to the early conversion of glucoraphanin into sulforaphane, ture was subject to VMAE; the activity of this enzyme after which subsequently degraded at a higher temperature. 2 min of the extraction was around 37% of the initial value. Although the mixture temperature increased to 60 °C within this 3.5. Extractable glucoraphanin contents from fresh versus steamed 2-min period, some myrosinase was still active. This remaining leaves activity in turn led to the hydrolysis of glucoraphanin into sul- foraphane. The normalized sulforaphane content reached around Based on the results presented in the previous sections, it is 0.31 at 2 min of the extraction. This rather high content of sul- seen that myrosinase could not be rapidly inactivated and could foraphane was noted probably because the degraded portion of still participate in the hydrolysis reaction. This led subsequently the compound was smaller than the initially formed and released to the losses of glucoraphanin. Sulforaphane that had been formed portion. also eventually suffered degradation. Inactivating myrosinase prior After 2 min the normalized myrosinase activity and sul- to extraction seemed therefore to be a reasonable option to take. In foraphane content decreased toward the end of the extraction; this our case, preliminary experiments showed that steaming for 2 min trend is similar to that of UAE + MAE. It took around 10 min of was adequate to inactivate myrosinase in the fresh cabbages. VMAE, with an increase in the mixture temperature to 90 °C, to Fig. 2 presents the evolutions of the normalized glucoraphanin completely inactivate endogenous myrosinase in cabbages. contents of the extracts from fresh and steamed cabbages undergo- ing different extraction methods. In the case of steamed cabbages 3.3. UAE + VMAE undergoing UAE + MAE, the initial normalized glucoraphanin content prior to MAE (after UAE for 40 min) was around 0.82, Fig. 1c shows the evolutions of the normalized glucoraphanin which was much higher than that observed in the case of the fresh and sulforaphane contents and myrosinase activity of the extract cabbages (around 0.12) because myrosinase was completely as well as the mixture temperature during VMAE part of the inactivated during steaming. Hydrolysis of glucoraphanin into UAE + VMAE; UAE was performed for 40 min prior to VMAE. The sulforaphane was inhibited and hence the reduced loss of gluco- initial normalized glucoraphanin content prior to VMAE (after raphanin. During MAE the normalized extractable glucoraphanin UAE for 40 min) was again around 0.12. The normalized myrosi- content from steamed cabbages increased to its maximum value nase activity and sulforaphane content prior to VMAE were 0.32 of around 0.95 in about 4 min; this maximum value was expect- and 0.28, respectively. edly higher than that in the case of the fresh cabbages. After that When the mixture was subject to microwave irradiation at an the content of glucoraphanin continuously decreased toward the absolute pressure of 70 kPa, glucoraphanin content increased to end of the extraction. In addition to the negligible enzyme activity, its maximum value of around 0.20 after 4 min. The maximum con- steaming might also help induce structural modification of the tent of glucoraphanin was higher than those in the cases of UAE cabbage matrix, hence facilitating the release of glucoraphanin into + MAE and VMAE alone. This is most probably because of the com- the extraction solvent [27]. bined effect of acoustic cavitation via ultrasound, which induced In the case of VMAE, the normalized extractable glucoraphanin modification of the cabbage structure, and of vacuum that helped content from steamed cabbages continuously increased to the reduce the boiling temperature of the water as well as limit the maximum value of around 0.83 in about 6 min. The content then oxygen content in the system. In addition, vacuum condition might remained almost unchanged toward the end of the extraction. P. Pongmalai et al. / Separation and Purification Technology 174 (2017) 338–344 343

1.4 140 Table 1 a) Normalized glucoraphanin contents of extracts from fresh and steamed cabbages via 1.2 120 different extraction methods.

1 100 Process Initial normalized Maximum normalized glucoraphanin content glucoraphanin content (time)* 0.8 80 Fresh cabbages UAE + MAE 0.12 ± 0.01b 0.18 ± 0.01b (4 min) 0.6 60 VMAE 0.00 ± 0.00a 0.15 ± 0.01a (8 min) UAE + VMAE 0.12 ± 0.01b 0.20 ± 0.00c (4 min) 0.4 40 Temperature (°C) Steamed cabbages 0.2 20 UAE + MAE 0.82 ± 0.04c 0.95 ± 0.04e (4 min) Notmalized glucoraphanin content glucoraphanin Notmalized VMAE 0.00 ± 0.00a 0.83 ± 0.05d (6 min) 0 0 UAE + VMAE 0.85 ± 0.02c 1.24 ± 0.02f (6 min) 0246810 Time (min) Same letters in the same column indicate that values are not significantly different (p > 0.05). * 1.4 140 Time at which the maximum content was observed. b) 1.2 120

1 100 was not performed on the steamed leaves as this process simply took too long. 0.8 80 Steaming in combination with the use of UAE + VMAE resulted 0.6 60 in the highest recovery of glucoraphanin. The maximum value in this case was around 650.09 lmol/100 g dry weight, which was

0.4 40 Temperature (°C) 87% higher than that when using the fresh leaves in combination with UAE + VMAE. It is not possible to compare this highest 0.2 20

Normalized glucoraphanin content Normalized content to any existing data as no study is so far available on the 0 0 glucoraphanin content in an extract. Nevertheless, in the case of 0246810 total glucosinolates, our preliminary results showed that the con- Time (min) tent of these compounds from steamed cabbages as extracted by UAE + VMAE was 34% higher than the value reported by Nugrahedi 1.4 120 c) et al. [28] who measured the glucosinolates content in blanched 1.2 100 white cabbages using heated methanol as a solvent. 1 80 0.8 3.6. Kinetic modeling of bioactive compounds contents and myrosinase 60 activity 0.6 40 The kinetic parameters, including the rate constant (k) and 0.4 Temperature (°C) order of the evolution kinetics (n), belonging to the normalized 0.2 20 bioactive compounds contents and myrosinase activity at different Normalized glucoraphanin content Normalized conditions are given in Table 2. In most cases, first-order kinetic 0 0 model (n = 1) could be used to explain the observed evolutions; 0246810 Time (min) Table 2 Fig. 2. Evolutions of normalized glucoraphanin content and temperature during Kinetic parameters (k and n) of normalized bioactive compounds and myrosinase different extraction methods. (A) MAE (as part of UAE + MAE); (B) VMAE and (C) activity during extraction. VMAE (as part of UAE + VMAE). (s) Glucoraphanin content of extracts from fresh leaves; (d) Glucoraphanin content of extracts from steamed leaves and (h) mixture Normalized bioactive Process Rate Order of (cabbages + DI water) temperature. compound content and constant evolution myrosinase activity (k) kinetics (n) Glucoraphanin Fresh cabbages UAE + MAE – – The maximum extractable glucoraphanin content from the VMAE 0.30 1 steamed cabbages was higher than that from the fresh cabbages. UAE + VMAE 0.40 1 This is again because steaming led to endogenous myrosinase inac- Soxhlet extraction – – tivation and also helped soften the cabbage structure. Steamed cabbages In the case of UAE + VMAE, when the cabbage-water mixture UAE + MAE – – VMAE 0.50 1 was subject to microwave irradiation at 70 kPa, the glucoraphanin UAE + VMAE 0.60 1 content increased to the maximum value of around 1.24 within Sulforaphane Fresh cabbages 6 min; the content then remained almost unchanged toward the UAE + MAE 0.30 1 end of the extraction. VMAE – – Table 1 summarizes the condition that resulted in the maxi- UAE + VMAE 0.25 1 mum extractable glucoraphanin content in each case. The maxi- Soxhlet extraction – – mum normalized glucoraphanin content from UAE + VMAE was Myrosinase activity Fresh cabbages significantly higher than those obtained from UAE + MAE and UAE + MAE 0.30 1 VMAE alone. The maximum normalized glucoraphanin contents VMAE 0.43 1 UAE + VMAE 0.30 1 extractable from steamed cabbages were significantly higher than Soxhlet extraction 0.04 1 those from fresh cabbages in all cases. Note that Soxhlet extraction 344 P. Pongmalai et al. / Separation and Purification Technology 174 (2017) 338–344 in some cases, Eq. (1) cannot be used to fit the observed evolutions [4] V. Briones-Labarca, M. Plaza-Morales, C. Giovagnoli-Vicuña, F. Jamett, High and the rate constant as well as the order of the evolution kinetics hydrostatic pressure and ultrasound extractions of antioxidant compounds, sulforaphane and fatty acids from Chilean papaya (Vasconcellea pubescens) in those cases are not reported. seeds: effects of extraction conditions and methods, LWT-Food Sci. Technol. 60 It is seen that UAE + VMAE exhibited higher rate constants of (2015) 525–534. glucoraphanin than VMAE. This implies that UAE + VMAE led to [5] Z.-X. Gu, Q.-H. Guo, Y.-J. Gu, Factors influencing glucoraphanin and sulforaphane formation in Brassica plants: a review, J. Integr. Agric. 11 a more extensive extraction of this compound. Glucoraphanin (2012) 1804–1816. extractable from fresh cabbages via UAE + MAE and Soxhlet [6] P. Lekcharoenkul, Y. Tanongkankit, N. Chiewchan, S. Devahastin, Enhancement extraction degraded after reaching the maximum values due to of sulforaphane content in cabbage outer leaves using hybrid drying technique and stepwise change of drying temperature, J. Food Eng. 122 (2014) 56–61. thermal degradation (as mentioned in Sections 3.1 and 3.4); k [7] C. Pérez, H. Barrientos, J. Román, A. Mahn, Optimization of a blanching step to and n values could not be determined since there is no degrada- maximize sulforaphane synthesis in broccoli florets, Food Chem. 145 (2014) tion term in the equation. The rate constants belonging to gluco- 264–271. [8] Y. Tanongkankit, S.S. Sablani, N. Chiewchan, S. Devahastin, Microwave-assisted raphanin contents extractable from steamed cabbages were extraction of sulforaphane from white cabbages: effects of extraction higher than those from fresh cabbages in almost all cases. This condition, solvent and sample pretreatment, J. Food Eng. 117 (2013) 151–157. is because of the inactivation of endogenous myrosinase by [9] E.H. Jeffery, K.E. Stewart, Upregulation of quinone reductase by steaming. hydrolysis products from dietary broccoli, in: H. Sies, L. Packer (Eds.), Quinones and Quinone Enzymes, Elsevier, London, 2004, pp. 457–472. UAE + MAE resulted in higher rate constants belonging to sul- [10] S.K. Ghawi, L. Methven, K. Niranjan, The potential to intensify sulforaphane foraphane than UAE + VMAE. This is probably because of the higher formation in cooked broccoli (Brassica oleracea var. italica) using mustard seeds extraction temperature during UAE + MAE. In terms of the myrosi- (Sinapis alba), Food Chem. 138 (2013) 1734–1741. [11] E. Luengo, S. Condón-Abanto, S. Condón, I. Álvarez, J. Raso, Improving the nase activity, Soxhlet extraction resulted in a much lower (inacti- extraction of carotenoids from tomato waste by application of ultrasound vation) rate constant than the other extraction methods. This is under pressure, Sep. Purif. Technol. 136 (2014) 130–136. because myrosinase in the cabbage matrix was not in direct con- [12] P. Pongmalai, S. Devahastin, N. Chiewchan, S. Soponronnarit, Enhancement of microwave-assisted extraction of bioactive compounds from cabbage outer tact with heat during Soxhlet extraction. The kinetic model results leaves via the application of ultrasonic pretreatment, Sep. Purif. Technol. 144 corresponded to the results reported in Sections 3.1–3.5. (2015) 37–45. [13] D.-T. Xie, Y.-Q. Wang, Y. Kang, Q.-F. Hu, N.-Y. Su, J.-M. Huang, C.-T. Che, J.-X. Guo, Microwave-assisted extraction of bioactive alkaloids from Stephania 4. Conclusions sinica, Sep. Purif. Technol. 130 (2014) 173–181. [14] B. Hiranvarachat, S. Devahastin, S. Soponronnarit, Comparative evaluation of The evolutions of the glucoraphanin and sulforaphane contents atmospheric and vacuum microwave-assisted extraction of bioactive compounds from fresh and dried Centella asiatica L. leaves, Int. J. Food Sci. of the extracts from cabbages along with the myrosinase activity Technol. 50 (2015) 750–757. and change of the extraction mixture temperature during UAE [15] M. Benlloch-Tinoco, M. Igual, D. Rodrigo, N. Martínez-Navarrete, Comparison + MAE, VMAE and UAE + VMAE were investigated with the aim to of microwaves and conventional thermal treatment on enzymes activity and antioxidant capacity of kiwifruit puree, Innovative Food Sci. Emer. Technol. 19 enhance the extraction of glucoraphanin. UAE + VMAE led to a sig- (2013) 166–172. nificantly higher maximum content of glucoraphanin than UAE [16] L.C. Lopes, M.T.M. Barreto, K.M. Gonçalves, H.M. Alvarez, M.F. Heredia, R.O.M.A. + MAE and VMAE due to the combined effect of acoustic cavitation de Souza, Y. Cordeiro, C. Dariva, A.T. Fricks, Stability and structural changes of horseradish peroxidase: microwave versus conventional heating treatment, and microwave irradiation at a lower temperature. The maximum Enzyme Microb. Technol. 69 (2015) 10–18. content of glucoraphanin extractable via Soxhlet extraction was [17] X.F. Cheng, M. Zhang, B. Adhikari, The inactivation kinetics of polyphenol significantly lower than those from the other extraction methods; oxidase in mushroom (Agaricus bisporus) during thermal and thermosonic treatments, Ultrason. Sonochem. 20 (2013) 674–679. the process also took much longer time. UAE + VMAE, when per- [18] A. Sulaiman, M.J. Soo, M. Farid, F.V.M. Silva, Thermosonication for formed on the fresh cabbages, could not adequately inactivate polyphenoloxidase inactivation in fruits: modeling the ultrasound and endogenous myrosinase, resulting in some conversion of gluco- thermal kinetics in pear, apple and strawberry purees at different raphanin into sulforaphane and hence the loss of the former com- temperatures, J. Food Eng. 165 (2015) 133–140. [19] B. Hiranvarachat, S. Devahastin, N. Chiewchan, G.S.V. Raghavan, Structural pound during the extraction. UAE + VMAE of steamed cabbages modification by different pretreatment methods to enhance microwave- proved to result in the highest recovery of glucoraphanin, with assisted extraction of b-carotene from carrots, J. Food Eng. 115 (2013) 190–197. the value of around 650.09 lmol/100 g dry weight, 87% higher [20] B. Hiranvarachat, S. Devahastin, Enhancement of microwave-assisted extraction via intermittent radiation: extraction of carotenoids from carrot than that obtained via the use of the fresh cabbages in combination peels, J. Food Eng. 126 (2014) 17–26. with UAE + VMAE. Simple kinetic model could be used to describe [21] K.-C. Lee, M.-W. Cheuk, W. Chan, A.W.-M. Lee, Z.-Z. Zhao, Z.-H. Jiang, Z. Cai, the extraction of glucoraphanin from both fresh and steamed cab- Determination of glucosinolates in traditional Chinese herbs by high- performance liquid chromatography and electrospray ionization mass bages via UAE + VMAE. spectrometry, Anal. Bioanal. Chem. 386 (2006) 2225–2232. [22] Y. Tanongkankit, N. Chiewchan, S. Devahastin, Evolution of anticarcinogenic Acknowledgements substance in dietary fibre powder from cabbage outer leaves during drying, Food Chem. 127 (2011) 67–73. [23] C.S. Charron, A.M. Saxton, C.E. Sams, Relationship of climate and genotype to The authors express their sincere appreciation to the Thailand seasonal variation in the glucosinolate-myrosinase system. I. Glucosinolate Research Fund (TRF) for financially supporting the study through content in ten cultivars of Brassica oleracea grown in fall and spring seasons, J. Sci. Food Agric. 85 (2005) 671–681. Grant no. RTA5880009 to Author Devahastin and Grant no. [24] R. Verkerk, M. Dekker, Glucosinolates and myrosinase activity in DPG5980004 to Author Soponronnarit. The authors are also grate- (Brassica oleracea L. Var. Capitata f. rubra DC.) after various microwave ful to Ms. Thitima Kuljarachanan for her help with the gluco- treatments, J. Agric. Food Chem. 52 (2004) 7318–7323. raphanin content determination procedures. [25] G.-C. Yen, Q.-K. Wei, Myrosinase activity and total glucosinolate content of , and some properties of cabbage myrosinase in Taiwan, J. Sci. Food Agric. 61 (1993) 471–475. References [26] L.G. Wilson, J.F. Artiola, Environmental monitoring and characterization, in: J. F. Artiola, I.L. Pepper, M.L. Brusseau (Eds.), Soil and Vadose Zone Sampling, Elsevier, Burlington, 2004, pp. 103–120. [1] S. Chaisamlitpol, B. Hiranvarachat, J. Srichumpoung, S. Devahastin, N. [27] J. Volden, G.I.A. Borge, M. Hansen, T. Wicklund, G.B. Bengtsson, Processing Chiewchan, Bioactive compositions of extracts from cabbage outer leaves as (blanching, boiling, steaming) effects on the content of glucosinolates and affected by drying pretreatment prior to microwave-assisted extraction, Sep. antioxidant-related parameters in cauliflower (Brassica oleracea L. ssp. Purif. Technol. 136 (2014) 177–183. botrytis), LWT-Food Sci. Technol. 42 (2009) 63–73. [2] I.R. Redovnikovic´, T. Glivetic´, K. Delonga, J. Vorkapic´-Furacˇ, Glucosinolates and [28] P.Y. Nugrahedi, M. Dekker, B. Widianarko, R. Verkerk, Quality of cabbage their potential role in plant, Period. Biol. 110 (2008) 297–309. during long term steaming; phytochemical, texture and colour evaluation, [3] J. Kapusta-Duch, B. Kusznierewicz, T. Leszczyn´ ska, B. Borczak, Effect of cooking LWT-Food Sci. Technol. 65 (2016) 421–427. on the contents of glucosinolates and their degradation products in selected Brassica vegetables, J. Funct. Foods. 23 (2016) 412–422.