Research Article pubs.acs.org/journal/ascecg
Optimization of Ecofriendly Extraction of Bioactive Monomeric Phenolics and Useful Flavor Precursors from Grape Waste Rebecca E. Jelley,† Mandy Herbst-Johnstone,† Steffen Klaere,‡ Lisa I. Pilkington,† Claire Grose,§ Damian Martin,§ David Barker,*,†,∥ and Bruno Fedrizzi*,†,∥ †
School of Chemical Sciences, University of Auckland, Private Bag 92019, Auckland, New Zealand Department of Statistics, University of Auckland, Private Bag 92019, Auckland, New Zealand § The New Zealand Institute for Plant & Food Research Limited, Marlborough, P.O. Box 845, Blenheim, New Zealand ∥ Centre for Green Chemical Science, School of Chemical Sciences, University of Auckland, Private Bag 92019, Auckland, New Zealand ‡
S Supporting Information *
ABSTRACT: Grape marc, also known as grape pomace, is an underutilized biowaste composed predominantly of grape skin and seeds that is produced as a byproduct of winemaking on the milliontonne scale annually. The most important high-value current use of grape marc is in the production of oenological tannins, widely used additives in the food and beverage industry. More commonly, grape marc is simply either disposed of or used as feed or fertilizer. With recent evidence showing that extracts enriched in grape tannins contain significant amounts of the thiol precursors Cys-3SH and GSH-3SH and the possibility that these could influence food and beverage aroma, it was decided to investigate grape pomace extraction procedures in order to try to define extraction protocols that could maximize the recovery of these compounds from grape marc. Should such an extraction protocol be shown to be commercially viable, this could lead to further utilization of material that would otherwise be disposed of. Two thiol precursors and eight monomeric phenolics (quercetin 3-O-glucoside, quercetin, resveratrol, grape reaction product, trans-caftaric acid, catechin, epicatechin, and gallic acid) were identified and simultaneously extracted from Sauvignon Blanc grape marc using solid−liquid extractions. The optimal solvent ratio of acetone:water:EtOH was explored for each compound. Ten ternary diagrams were constructed, showing the effectiveness of extraction across 66 different solvent combinations using the aforementioned solvents. Effective extraction of thiol precursors was dependent on a high water content which is an advantage from an economic and environmental perspective, while for the most abundant phenolic, quercetin 3-O-glucoside, optimal extraction levels (1017 mg per kg of grape marc) were achieved using a 40:50:10 solvent mixture. In addition, this manuscript details the first extraction of thiol precursors from grape pomace which adds a significant potential commercial value to this underutilized byproduct. KEYWORDS: Grape marc, Thiol precursors, Phenolics, Sauvignon blanc, Tannins
■
INTRODUCTION There has been considerable discussion in the literature concerning the extraction of valuable materials from horticultural waste.1 Of particular interest in this respect is the wine industry, which produces large amounts of such waste. It has been estimated that up to 18%−20% (w/w) of all harvested grapes used for wine production end up as grape marc,2 which gives a potential worldwide annual production of approximately 13 million tonnes.3 The Marlborough (New Zealand) wine industry alone generated approximately 50,000 tonnes of grape marc in 2014.4 Grape waste is normally considered to have limited economic value and is generally used for animal feed or as soil fertilizer.5 Recent studies have shown that grape marc can be successfully used as a source of carbohydrates for bioethanol3 and of a variety of bioactive compounds.6 With regard to the latter case, most of these studies have focused simply on phenolic © XXXX American Chemical Society
compounds. These compounds are well known for their numerous health benefits as a result of both their antioxidant and antimicrobial properties.7 The extraction of phenolics from grape waste has been achieved previously using a variety of methods, ranging from the use of supercritical CO2 to microwave-assisted extraction.8,9 However, solid−liquid extraction (SLE) is the most commonly used method due to its simplicity and commercial feasibility.10 Additionally, SLE has a great capacity for extensive adaptation to follow green chemistry ideals.11 Of particular importance in the SLE method is the solvent system, and many studies have attempted to identify the optimal solvent or solvent system to extract phenolic compounds from both red and white grape marc.6 Received: July 5, 2016 Revised: July 28, 2016
A
DOI: 10.1021/acssuschemeng.6b01551 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX
Research Article
ACS Sustainable Chemistry & Engineering
fermented beverages to other foodstuffs and soft drinks, as these nonvolatile molecules can be released by enzymes present in the saliva.27 The precursors can also potentially be used in the commercial production of fragrances after the application of a suitable fermentation.28 However, to the best of our knowledge, neither the identification of these thiol precursors in grape pomace extracts nor the subsequent SLE optimization of precursors from grape marc have been explored. In this study, we report the optimization of a mixed solvent system comprising acetone, water, and ethanol for the extraction of the thiol precursors, Cys-3SH and GSH-3SH. These solvents were chosen as they are considered to be relatively “green” solvents in environmental, health, and safety (EHS) assessments29 and also as they have been commonly reported in the literature as being effective for extraction of phenolics. Previously, both methanol and ethanol have been reported for the extraction of phenolic compounds.6 These solvents are chemically very similar resulting in comparable extraction capabilities, yet as ethanol is readily used in the food industry, and can be obtained in food grade, it was chosen in preference to methanol. It was decided that extractions would be undertaken at ambient temperature as it is the most energy and cost efficient, convenient, and commercially viable condition to perform on a large scale. We also provide, for the first time, an in-depth profile of the extractable phenolic compounds from Sauvignon Blanc grape marc using this combination of solvents.
Four main solvents, and combinations thereof, appear to dominate the literature in this field: acetone, ethanol, methanol, and water.6 The most common approach taken to quantify the ability of such solvent systems to extract phenolics has been to measure the total phenolic content of the extract, with little or no attention paid to the individual profile of the phenolics being extracted with each solvent system. With many potential applications for specific phenolics, the ability to selectively enhance the extraction of a particular phenolic compound, or family, by means of optimizing the solvent system becomes of paramount importance, as it can allow the efficient extraction of valuable materials from grape marc. Brief studies aimed at quantifying individual phenolics extracted from red grape marc, using a small number of solvent systems, have been previously reported.12−14 Recently, a study by Bosso et al. reports the investigation into a selection of solvent systems on the extraction of tannins in the seeds from red wine grape pomace.15 This report focuses on total phenolic content but also investigates the molar % of individual phenolics isolated using binary solvent systems. However, there has been limited, if any, interest in the in-depth optimization of solvent systems for a straightforward SLE approach to monitor the extraction of specific phenolic compounds from white grape waste, which can potentially be used as either industrially important materials or useful chemical building blocks. As the majority of phenolic compounds are located in the grape skin, Sauvignon Blanc wine, which represents approximately 70% of grape production in New Zealand,16 is reported to be low in antioxidants compared to red wines due to undergoing a light pressing involving little grape skin contact during the winemaking process.17,18 For these reasons, the lightly pressed Sauvignon Blanc grape marc should retain higher amounts of organic compounds and is a good candidate for the study of the extraction of phenolic compounds. With phenolic compounds typically being the primary target in grape marc extractions, there has been little interest, until now, into the potential extraction of important aroma precursors from grape marc in light of their potential use as food additives in commercial tannins. The varietal thiol precursors 3-S-cysteinylhexan-1-ol (Cys-3SH) and 3-S-glutathionylhexan-1-ol (GSH-3SH) have been detected in grape skins and grape must, as well as in wine.19,20 Furthermore, the presence of leaves and stems has been found to significantly increase the presence of these thiols in grape pomace.21 To the best of our knowledge, in-depth exploration of grape marc as a viable source of these compounds has not been extensively studied. The significance of these thiol-containing precursors on the ultimate sensory profile of fruits and beverages has long been discussed.19 The aroma profile of Sauvignon Blanc and other wines is heavily influenced by the concentrations of varietal thiols which are known to provide the tropical, fruity aroma typically associated with NZ Sauvignon Blanc.22 3Mercaptohexan-1-ol (3SH) can be released from Cys-3SH and GSH-3SH by yeast activity during alcoholic fermentation23 or via the reaction between (E)-2-hexenal and H2S.24 Grapederived oenological tannins, previously reported to contain important levels of Cys-3SH and GSH-3SH,25 were added to grape juice prior to fermentation as a potential external source of thiol precursors, and this successfully increased the levels of 3SH that were detected,26 suggesting a potential role for the external addition of aroma precursors by means of technological strategies in the aroma modulation of wine. The application of these products could be extended beyond
■
EXPERIMENTAL SECTION
Grape Pomace. Sauvignon Blanc grape pomace, which included skins, seeds, and stems, was supplied by Indevin New Zealand (Cloudy Bay Business Park, Riverlands, Marlborough) from grapes harvested in the Marlborough region from the 2015 vintage. A representative 30 kg marc sample was collected on April 11, 2015, post-pressing during the unloading phase of the press. Grapes were pressed for a duration of 4.5 h yielding approximately 780−800 L of juice per tonne of crushed grape must. The marc sample was packaged, frozen, and stored at −20 °C for 10 weeks until analysis. Reagents. Commercial standards of rutin hydrate, caffeic acid, gallic acid, epicatechin, and catechin hydrate were purchased from Sigma-Aldrich (NSW, Australia). Water was of Milli-Q grade (resistivity = 18.1 MΩ at 22 °C), and analytical grade ethanol, acetone, acetonitrile, acetic acid, and formic acid were purchased from ECP Laboratory and Research Chemicals (Auckland, NZ), SigmaAldrich (NSW, Australia), and Merck (Darmstadt, Germany). d3-Cys3SH and d3-GSH-3SH were purchased from Buchem (Apeldoorn, The Netherlands). Solvent Systems. Various combinations of the solvents acetone, water, and ethanol were investigated and are expressed in volume ratios (v:v:v). The proportion of each solvent in a particular mix ranged from 0% to 100%. Increments of 10% were used to uniformly cover the entire spectrum of solvent combinations. This gave a total of 66 solvent profiles (see Supporting Information for corresponding figure). The extraction procedure described below was repeated in triplicate for each of the 66 solvent profiles. The average of the three analyses is reported, along with the associated standard deviation (SD). Extraction. Defrosted grape waste (2.5 kg) was homogenized using a 600 W kitchen blender to give a uniform material and was stored at −20 °C until use. Defrosted grape pomace (10 g) and internal standard (30 μL, d3-Cys-3SH and d3-GSH-3SH, final concentrations of 44 and 78 μg/L, respectively) were stirred in a solvent mix of acetone/ water/ethanol (50 mL) for 1 h at room temperature (22 °C). The solvent mix was decanted and centrifuged for 10 min at 6000 rpm. A portion (5 mL) of the resulting supernatant was isolated, and the solvent was removed in vacuo (rotary evaporator). The resulting B
DOI: 10.1021/acssuschemeng.6b01551 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX
Research Article
ACS Sustainable Chemistry & Engineering residue was dissolved in 2 mL of model white wine (12% ethanol, 5 g/ L tartaric acid, pH adjusted to 3.2 with NaOH), passed through a Phenomenex regenerated cellulose (RC) membrane, 0.45 μm, 15 mm syringe filter, and stored at −20 °C until HPLC analysis. UHPLC-MS/MS Conditions. Thiol precursors (Cys-3SH and GSH-3SH) were identified and quantified using a modification of a previously reported LCMS method,30,31 using an Agilent 6460 Triple Quadrupole LC/MS system with an Agilent 1290 Infinity LC. The ion source was AJS ESI, gas temperature of 200 °C, and gas flow of 9 L/ min. A C18 Phenomenex Kinetex column (CA, U.S.A.) was used (100 mm × 3 mm, 100 Å, 2.6 μm) operating at 25 °C. The solvent system was 0.1% formic acid in Milli-Q water (solvent A) and 100% acetonitrile (solvent B) with a flow rate of 500 μL/min. The gradient for solvent B over the 20 min run was as follows: 0 min 0%, 6 min 5%, 10 min 15%, 12 min 80%, 15 min 100%, and 17 min 5%. A 10 μL injection volume was used for each sample. Detection was carried out in Multiple Reaction Monitoring (MRM) mode with the mass spectrometer parameters described in Table 1.
epicatechin, gallic acid, and rutin hydrate) in absolute ethanol before serial dilution in model white wine to give a concentration range between 0.34 and 630 mg/L appropriate to the particular compound. Calibration curves were constructed with excellent linearity to give a R2 value of greater than 0.9996 in each case. Flavonols were quantified as mg of rutin per kg of grape pomace and hydroxycinnamates as mg of caffeic acid per kg of grape pomace. Statistical Analysis. To identify regions of high and low effectiveness in compound extraction, locally smoothed models were fitted to the extraction data for each compound using R (version 3.2.1).33 To visualize these models, smoothed ternary surface plots were generated using the R-package “ggtern”,34 and Pearson correlation coefficients were calculated using R.33 Significance was tested using a correlation test with false discovery rate correction for multiple testing. The correlation matrix was visually represented using the R package “corrplot”.35 The smoothing approach uses the observed extractions over the discrete grid of analyzed solvent combinations to predict the outcome of other solvent combinations within the ternary diagram. The nature of this approach can lead to peaks in the smoothed surface that do not correspond to the measured optimum, e.g., if combinations close to the measured optimum are suboptimal too, the smoother might assume a higher peak in the center of the group of optima and suboptima.
Table 1. Mass Spectral Parameters for Each Thiol Precursor Analyseda
a
compound
precursor ion (m/z)
product ion (m/z)
fragmentation voltage (V)
collison energy (V)
Cys-3SH
222
205 101 83 208 104 86 333 279 262 162 336 282 265
120 120 120 120 120 120 110 110 110 110 110 110 110
5 16 40 5 16 40 15 8 15 22 15 8 15
d3-Cys-3SH
225
GSH-3SH
408
d3-GSH-3SH
411
■
RESULTS AND DISCUSSION Thiol Precursors. The thiol precursors, Cys-3SH and GSH3SH, were detected in all grape marc extracts across the entire solvent profile investigated. Cys-3SH was detected in concentrations ranging from 0.87 ± 0.12 to 1.19 ± 0.03 mg/ kg grape marc. GSH-3SH was detected in concentrations approximately 3-fold greater than Cys-3SH, with values ranging from 2.33 ± 0.04 to 3.76 ± 0.03 mg/kg grape marc. Our findings give a significantly larger average concentration of Cys-3SH and GSH-3SH extracted from grape marc than those previously reported. Average concentrations of these thiol precursors in the grape skins of three French Sauvignon Blanc were about 30 μg Cys-3SH/kg grape skin and about 32 μg GSH-3SH/kg grape skin.19 The average concentration of Cys3SH extracted from the marc in this study was 35-fold larger than reported in the Sauvignon Blanc grape skins and 100-fold greater for GSH-3SH. In addition to this difference, the concentrations of GSH-3SH and Cys-3SH extracted from the grape marc showed greater relative variation than those found in the grape skins. The concentration of GSH-3SH detected in the grape marc extract was on average 3 times higher than that
Dwell time = 50 ms for each transition.
HPLC-DAD Procedure. Monomeric white wine phenolics were identified and quantified on the basis of a 90 min reversed-phase HPLC method previously described by Patel et al. 32 The spectrophotometric detector was set at 280 nm (hydroxybenzoic acids and flavan-3-ols), 320 nm (hydroxycinnamates), and 365 nm (flavonols). External standards were prepared for each compound investigated by dissolving standards of phenolics (caffeic acid, catechin hydrate,
Figure 1. Extraction of the thiol precursors Cys-3SH (a) and GSH-3SH (b) from Sauvignon Blanc grape marc with various proportions of acetone, water, and ethanol. Values are reported in mg of precursor extracted per kg of grape marc. C
DOI: 10.1021/acssuschemeng.6b01551 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX
Research Article
ACS Sustainable Chemistry & Engineering of Cys-3SH, whereas the Sauvignon Blanc grape skins gave approximately equal concentrations of the two thiol precursors. Grape marc derived from machine-harvested grapes contains additional organic matter (stems and leaves) compared to grape marc that originates from hand-harvested grapes. As stems and leaves have both been reported to be sources of the thiol precursors Cys-3SH and GSH-3SH,36 grape marc from machine-harvested grapes could reasonably provide a larger quantity of these thiol precursors on extraction. Additionally, the levels of thiol precursors obtained using the extraction conditions reported here suggest that our solvent systems could provide a better medium for the release of thiol precursors from the grape marc than those previously reported or facilitate a de novo formation of these precursors via the reaction between GSH and (E)-2-hexenal,24,37 both of which could be present in grape marc. Figure 1 illustrates the effect of solvent variation on the efficiency of extraction of Cys-3SH and GSH-3SH. As shown in Figure 1a, the most efficient extraction conditions are those that contain significant amounts of water, while solvent mixtures composed primarily of ethanol and acetone display the poorest efficiencies. The proportion of ethanol in the solvent mix can be seen to have little to no effect on the amounts of these thiol precursors that were extracted. However, it is important to note that the quantity of Cys-3SH extracted from the grape marc showed little variation between the different solvent systems. Therefore, the composition of the solvent system is not overly important for the extraction of this thiol precursor. Figure 1b shows a similar trend, with a high water content, again, leading to the most efficient extractions. However, a second region of interest can also be seen close to the second highest extraction of GSH-3SH at a ratio of 40:60:0 (acetone:water:EtOH). After pure water, this is the next-best solvent system region to target for the extraction of this thiol precursor. This second region may become important if simultaneous extraction of GSH-3SH together with other poorly water-extractable compounds is required. The expected solubility of Cys-3SH and GSH-3SH in pure water is greater than that in pure acetone or pure ethanol. Therefore, it might be anticipated that a high water content in the solvent system is necessary for optimal extraction and in fact this study shows that a mixed solvent system provides no further benefit for the extraction of these compounds over pure water. Water is inexpensive, environmentally friendly, and very suitable for use in the food industry. Phenolics. Eight phenolic compounds were detected in the grape marc extracts in sufficient amounts to quantify. These phenolic compounds were quercetin 3-O-glucoside, quercetin, resveratrol, gallic acid, catechin, epicatechin, trans-caftaric acid, and grape reaction product (GRP). Caffeic acid, vanillic acid, cis-coutaric acid, and trans-coutaric acid were also detected; however, these were only present in trace amounts and were therefore not quantified. Table 2 shows a summary of the optimal solvent system for the targeted extraction of each of these eight phenolic compounds. In all cases, the most efficient extractions were achieved using combinations of the three solvents investigated. Quercetin 3-O-glucoside, a member of the flavonol family, was by far the most abundant phenolic compound extracted. Grape skins are known to be very rich in flavonols38 and because of the light pressing process described earlier in the Introduction, a large quantity of this family of phenolics will remain in the grape marc. Figure 2a shows that the optimal
Table 2. Optimal Solvent Systems for Extraction of Thiol Precursors and Phenolic Compounds from Sauvignon Blanc Grape Marc compound
optimal solvent system (acetone:water:EtOH)
Cys-3SH GSH-3SH quercetin 3-O-glucosidea catechin epicatechin quercetina gallic acid GRPb trans-caftaric acidb resveratrolb
0:100:0 0:100:0 40:50:10 50:50:0 30:50:20 40:30:30 20:70:10 40:50:10 30:30:40 50:20:30
concentration (mg/kg) 1.19 3.76 1017.61 466.29 151.14 31.89 21.34 14.47 6.66 4.06
± ± ± ± ± ± ± ± ± ±
0.03 0.03 168.71 83.01 7.91 2.77 1.99 2.65 1.42 0.69
a
Expressed as rutin equivalent. bExpressed as caffeic acid equivalent. Values are reported as an average of three analyses ± SD.
solvent system for the extraction of quercetin 3-O-glucoside lies around the ratio of 40:50:10 (acetone:water:EtOH). The pure solvents alone, in particular water, are the poorest media for extraction of quercetin 3-O-glucoside. It has been proposed that alcohol-dominated solvent systems should be used to extract flavonols and resveratrol from grape waste, owing to the low solubility of these compounds in water.10 We have indeed found that pure water is the poorest medium for the extraction of quercetin 3-O-glucoside. However, in contrast to the above contention, we have found that a solvent system dominated by acetone and water with only a small proportion of ethanol (ca. 0%−30%) provides the best conditions for the extraction of quercetin 3-O-glucoside. Figure 2b shows similar trends to Figure 2a; this is notable, as even though quercetin and quercetin 3-O-glucoside are very similar, one would expect that the polarity of the carbohydrate portion of the glycoside versus the parent compound would cause their extraction profiles to differ. Again, a mixed solvent system, around the ratio of 40:30:30 (acetone:water:EtOH), is the best for extracting this flavonol. There has been some debate concerning the presence of ́ flavonols in Sauvignon Blanc grape marc. Rodriguez-Montealegre et al. originally found a total flavonol content of 25 ± 5.6 mg/kg in Sauvignon Blanc grape skins.38 However, a more recent study investigating the SLE extraction of phenolic compounds from Sauvignon Blanc grape marc reported that no flavonols were detected in the extract.13 On the basis of these findings, the authors suggested that this could be due to the absence of flavonols in white grapes. Our data support the earlier study with respect to the presence of flavonols in ́ Sauvignon Blanc grapes. Rodriguez-Montealegre et al. also showed that no flavonol aglycones were detected in Sauvignon Blanc grape skins and seeds.38 Our data seem to suggest that these compounds are indeed present in Sauvignon Blanc grape marc, possibly contained within components other than grape skins and seeds. Our results also show that a solvent mix dominated by ethanol is not the optimal medium for the extraction of resveratrol, again in contrast to the earlier report.10 Figure 2c illustrates that a mixed solvent system around the ratio 50:20:30 (acetone:water:EtOH) provided the best conditions for the efficient extraction of resveratrol from grape marc. As shown in Figure 2c, pure water is the poorest solvent system for the extraction of this compound. D
DOI: 10.1021/acssuschemeng.6b01551 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX
Research Article
ACS Sustainable Chemistry & Engineering
Figure 2. Extraction of quercetin 3-O-glucoside (a), quercetin (b), resveratrol (c), and GRP (d) from Sauvignon Blanc grape waste with various proportions of acetone, water, and ethanol. All values are reported in mg per kg of grape pomace. Quercetin 3-O-glucoside and quercetin are reported in rutin equivalents. Resveratrol and GRP are reported in caffeic acid equivalents.
10:50:40 (acetone:water:EtOH) in the plot for catechin. Catechin was extracted in similar quantities (466.29 ± 83.01 mg/kg) to those previously reported (477.2 ± 38.6 mg/kg).13 However, epicatechin was extracted in lower amounts (151.14 ± 79.1 mg/kg) with the previously reported value being over 3fold greater than the result in this study (506.1 ± 67.0 mg/ kg).13 While the amounts of catechin and epicatechin extracted in the previous study were similar, we found 3 times more catechin than epicatechin present in our grape marc extracts. A high proportion of water, as shown in Figure 3d, was found to be the best for extraction of gallic acid, with an optimal extraction region positioned around a ratio of 20:70:10 (acetone:water:EtOH). The eight monomeric phenolic compounds investigated were extracted to different extents depending on the acetone:water:EtOH ratio of the solvent system. In comparison to the commonly reported solvent system optimization studies, where total phenolic content is measured and the individual levels of the component phenolic compounds are not investigated, these findings provide a straightforward means to selectively enhance the extraction of targeted phenolic compounds by altering the solvent system, an advance that will prove to be extremely useful for those wishing to obtain extracts rich in a specific compound. The experimental methods used in the previous study by de la Cerda-Carrasco et al.13 differ slightly from those reported here. In particular, the grape marc extract previously reported
Ethanol is shown to be unimportant in the extraction of GRP and changing the ratio of acetone to water provides the largest effect (Figure 2d). The poorest media are systems containing large amounts of either ethanol or acetone, and it is apparent that a significant proportion of water is required to obtain efficient extraction of this compound. Although pure water is shown to be a good medium for extraction, the optimal region centers around a ratio of 40:50:10 (acetone:water:EtOH). This mixed solvent system, somewhat surprisingly, gives the most efficient extraction of GRP. As shown in Figure 3a, the optimal solvent system for extracting trans-caftaric acid is positioned around a ratio of 30:30:40 (acetone:water:EtOH). Increasing the relative amount of any of the three solvents results in a decrease in the extraction efficiency. In addition to the pure solvents, another region of poor extraction efficiency can be seen around a ratio of 60:10:30 (acetone:water:EtOH). In contrast to our findings, no caftaric acid was detected in the Sauvignon Blanc grape marc extracts reported by de la Cerda-Carrasco et al.13 After quercetin 3-O-glucoside, catechin and epicatechin were the most abundant phenolic compounds to be extracted. The ternary plots, Figure 3b and c, respectively show many similarities which are to be expected given that the two compounds are diastereoisomers. Similar regions of optimal extraction are seen around ratios of 50:50:0 and 30:50:20 (acetone:water:EtOH) for catechin and epicatechin, respectively. A second region of interest can be seen at around E
DOI: 10.1021/acssuschemeng.6b01551 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX
Research Article
ACS Sustainable Chemistry & Engineering
Figure 3. Extraction of trans-caftaric acid (a), catechin (b), epicatechin (c), and gallic acid (d) from Sauvignon Blanc grape marc with various proportions of acetone, water, and ethanol. Values are reported in mg of caffeic acid extracted per kg of grape marc. All values are reported in mg per kg grape pomace. trans-Caftaric acid is measured in caffeic acid equivalents.
This study has also shown for the first time that extraction of individual monomeric phenolic compounds (catechin, epicatechin, trans-caftaric acid, gallic acid, GRP, resveratrol, quercetin, and quercetin 3-O-glucoside) from Sauvignon Blanc grape marc can be optimized or manipulated through varying the acetone:water:EtOH ratio of the solvent system in a straightforward SLE. Quercetin 3-O-glucoside was extracted in the greatest quantities (up to 1017.61 ± 168.71 mg/kg grape marc using a 40:50:10, acetone:water:EtOH solvent system) across the entire solvent system profile investigated. This result was in contrast to a previous Sauvignon Blanc grape marc study where no flavonols were detected in the extract. These results highlight for the first time the ability of SLE procedures to obtain significant levels of valuable secondary metabolites that could be employed for industrial (thiol precursors) and nutraceutical (phenolics) purposes. The former class of compounds could be extracted from grape marc to obtain tannins with the potential to condition the aroma of food and beverages as well as for the production of fragrances which are currently derived from horticultural materials. The latter class of molecules (phenolics) could be selectively extracted for their known beneficial activities. This research particularly addresses many of the cornerstone principles of green chemistry, namely, the use of solvents and techniques that have little to no environmental impact, as well as the utilization of material that would otherwise be discarded as waste. Additionally, this research attempts to improve existing techniques to increase the effectiveness of extraction, thereby allowing for an overall decrease in the amount of solvent used
was stored in water and was re-extracted with organic solvents. It is therefore possible that some compounds may have been left behind in the aqueous layer. This may explain why some of the phenolic compounds were not detected previously but were detected in this study. The statistical correlation between the extracted compounds, the thiol precursors, and the eight phenolic compounds is illustrated in Figure 4. It can be seen that the strongest positive correlation is between the extractions of quercetin 3-Oglucoside and quercetin; the extraction profiles for these two compounds are most closely aligned, i.e., conditions that are the most suitable for the extraction of quercetin are too for the extraction of its corresponding glycoside. The strongest negative correlation is between the extractions of gallic acid and resveratrol; the better the extraction conditions are for one, the worse they for the other. Combining the insights from the correlation plot and the ternary plots, we obtain valuable information on groupings of compounds suitable for simultaneous extraction.
■
CONCLUSIONS
This study has demonstrated for the first time that Sauvignon Blanc grape marc is a valuable source of the thiol precursors Cys-3SH and GSH-3SH. In addition to this, the extraction of GSH-3SH was up to 100-fold greater than the previously reported extraction from Sauvignon Blanc grape skins. The extraction efficiency of these thiol precursors was shown to increase with an increasing proportion of water in the solvent system. F
DOI: 10.1021/acssuschemeng.6b01551 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX
Research Article
ACS Sustainable Chemistry & Engineering
Figure 4. Statistical correlation between the 10 extracted compounds. The pie charts and color gradients illustrate the strength of each of the correlations. Pairs of compounds without pie charts showed no statistically significant correlation (e.g., gallic acid and epicatechin).
Innovation and Employment to generate innovative products from primary sector byproducts. M.H.J. is supported by grants to B.F. from the New Zealand Ministry of Business, Innovation and Employment, New Zealand Winegrowers and Plant and Food Research. We also acknowledge the University of Auckland for additional funding.
to obtain comparable amounts of extracted material. The extraction conditions reported here will be important in the scale-up of this protocol, which could ultimately lead to a significant added value of an underutilized horticultural byproduct.
■
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acssuschemeng.6b01551. Figure showing the various combinations of solvents that were investigated. (PDF)
■
REFERENCES
(1) Makris, D. P.; Boskou, G.; Chiou, A.; Andrikopoulos, N. K. An investigation on factors affecting recovery of antioxidant phenolics and anthocyanins from red grape (Vitis vinifera L.) pomace employing water/ethanol-based solutions. Am. J. Food Technol. 2008, 3 (3), 164− 173. (2) Spanghero, M.; Salem, A. Z. M.; Robinson, P. H. Chemical composition, including secondary metabolites, and rumen fermentability of seeds and pulp of Californian (USA) and Italian grape pomaces. Anim. Feed Sci. Technol. 2009, 152 (4), 243−255. (3) Corbin, K. R.; Hsieh, Y. S. Y.; Betts, N. S.; Byrt, C. S.; Henderson, M.; Stork, J.; DeBolt, S.; Fincher, G. B.; Burton, R. A. Grape marc as a source of carbohydrates for bioethanol: Chemical composition, pretreatment and saccharification. Bioresour. Technol. 2015, 193, 76−83. (4) Robertson, J. Winery Wastewater & Grape Marc Monitoring Report; Malbourough District Council, 2014. http://www. marlborough.govt.nz/Services/~/~/media/Files/MDC/Home/ Services/Solid%20and%20Liquid%20Waste/2014_Winery_Waste_ Environment_Committee_Report.pdf (accessed August 2016). (5) Iora, S. R. F.; Maciel, G. M.; Zielinski, A. A. F.; da Silva, M. V.; Pontes, P. V. d. A.; Haminiuk, C. W. I.; Granato, D. Evaluation of the
AUTHOR INFORMATION
Corresponding Authors
*Tel: + 64 9 9238473. Fax: + 64 9 3737422. E-mail: d.barker@ auckland.ac.nz (D.B.). *Tel: + 64 9 9238473. Fax: + 64 9 3737422. E-mail: b.fedrizzi@ auckland.ac.nz (B.F.). Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work is supported by the Bioresource Processing Alliance, established by the New Zealand Ministry of Business, G
DOI: 10.1021/acssuschemeng.6b01551 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX
Research Article
ACS Sustainable Chemistry & Engineering
Stages of Must Fermentation. J. Agric. Food Chem. 2013, 61 (15), 3703−3713. (25) Larcher, R.; Tonidandel, L.; Nicolini, G.; Fedrizzi, B. First evidence of the presence of S-cysteinylated and S-glutathionylated precursors in tannins. Food Chem. 2013, 141 (2), 1196−1202. (26) Larcher, R.; Tonidandel, L.; Román Villegas, T.; Nardin, T.; Fedrizzi, B.; Nicolini, G. Pre-fermentation addition of grape tannin increases the varietal thiols content in wine. Food Chem. 2015, 166, 56−61. (27) Starkenmann, C.; Le Calvé, B.; Niclass, Y.; Cayeux, I.; Beccucci, S.; Troccaz, M. Olfactory Perception of Cysteine−S-Conjugates from Fruits and Vegetables. J. Agric. Food Chem. 2008, 56 (20), 9575−9580. (28) Mantzouridou, F. T.; Paraskevopoulou, A.; Lalou, S. Yeast flavour production by solid state fermentation of orange peel waste. Biochem. Eng. J. 2015, 101, 1−8. (29) Capello, C.; Fischer, U.; Hungerbuehler, K. What is a green solvent? A comprehensive framework for the environmental assessment of solvents. Green Chem. 2007, 9, 927−934. (30) Larcher, R.; Nicolini, G.; Tonidandel, L.; Román Villegas, T.; Malacarne, M.; Fedrizzi, B. Influence of oxygen availability during skincontact maceration on the formation of precursors of 3-mercaptohexan-1-ol in Müller-Thurgau and Sauvignon Blanc grapes. Aust. J. Grape Wine Res. 2013, 19 (3), 342−348. (31) Mattivi, F.; Fedrizzi, B.; Zenato, A.; Tiefenthaler, P.; Tempesta, S.; Perenzoni, D.; Cantarella, P.; Simeoni, F.; Vrhovsek, U. Anal. Chim. Acta 2012, 732, 194−202. (32) Patel, P.; Herbst-Johnstone, M.; Lee, S. A.; Gardner, R. C.; Weaver, R.; Nicolau, L.; Kilmartin, P. A. Influence of juice pressing conditions on polyphenols, antioxidants, and varietal aroma of Sauvignon blanc microferments. J. Agric. Food Chem. 2010, 58 (12), 7280−7288. (33) R Core Team (R Foundation for Statistical Computing). R: A Language and Environment for Statistical Computing, 2015. https:// www.R-project.org (accessed September 10, 2015). (34) Hamilton, N. ggtern: An Extension to ‘ggplot2’ for the creation of Ternary Diagrams. R package version 3.2.1, 2015. http://CRAN.Rproject.org/package=ggtern (accessed September 10, 2015). (35) Wei, T. corrplot: Visualization of a correlation matrix. R package version 3.2.1, 2013. http://CRAN.R-project.org/package=corrplot (accessed September 10, 2015). (36) Kobayashi, H.; Takase, H.; Kaneko, K.; Tanzawa, F.; Takata, R.; Suzuki, S.; Konno, T. Analysis of S-3-(hexan-1-ol)-glutathione and S-3(hexan-1-ol)-L-cysteine in Vitis vinifera L. cv. Koshu for aromatic wines. Am. J. Enol. Viticult. 2010, 61 (2), 176−185. (37) Schneider, R.; Charrier, F.; Razungles, A.; Baumes, R. Evidence for an alternative biogenetic pathway leading to 3-mercaptohexanol and 4-mercapto-4-methylpentan-2-one in wines. Anal. Chim. Acta 2006, 563 (1), 58−64. (38) Rodríguez-Montealegre, R.; Romero-Peces, R.; ChacónVozmediano, J. L.; Martínez-Gascueña, J.; García-Romero, E. Phenolic compounds in skins and seeds of ten grape Vitis vinifera varieties grown in a warm climate. J. Food Compos. Anal. 2006, 19 (6−7), 687− 693.
bioactive compounds and the antioxidant capacity of grape pomace. Int. J. Food Sci. Technol. 2015, 50 (1), 62−69. (6) Fontana, A. R.; Antoniolli, A.; Bottini, R. Grape Pomace as a Sustainable Source of Bioactive Compounds: Extraction, Characterization, and Biotechnological Applications of Phenolics. J. Agric. Food Chem. 2013, 61 (38), 8987−9003. (7) Xia, E.; He, X.; Li, H.; Wu, S.; Li, S.; Deng, G. Biological activities of polyphenols from grapes. In Polyphenols in Human Health and Disease; Watson, R. R.; Preedy, V. R.; Zibadi, S., Eds.; Academic Press: San Diego, CA, 2014. (8) Krishnaswamy, K.; Orsat, V.; Gariépy, Y.; Thangavel, K. Optimization of microwave-assisted extraction of phenolic antioxidants from grape seeds (Vitis vinifera). Food Bioprocess Technol. 2013, 6 (2), 441−455. (9) Louli, V.; Ragoussis, N.; Magoulas, K. Recovery of phenolic antioxidants from wine industry by-products. Bioresour. Technol. 2004, 92 (2), 201−208. (10) Davidov-Pardo, G.; Navarro, M.; Arozarena, I.; Marin-Arroyo, M. R. Obtaining polyphenolic extracts from wine by-products. In Grapes: Production, Phenolic Composition and Potential Biomedical Effects; Câmara, J. S., Eds.; Nova Science Publishers, Inc., 2014. (11) Anastas, P. T.; Warner, J. C. Green Chemistry: Theory and Practice; Oxford University Press: New York, 1998. (12) Agustin-Salazar, S.; Medina-Juárez, L. A.; Soto-Valdez, H.; Manzanares-López, F.; Gámez-Meza, N. Influence of the solvent system on the composition of phenolic substances and antioxidant capacity of extracts of grape (Vitis vinifera L.) marc. Aust. J. Grape Wine Res. 2014, 20 (1), 208−213. (13) de la Cerda-Carrasco, A.; López-Solís, R.; Nuñez-Kalasic, H.; Peñ a-Neira, A.; Obreque-Slier, E. Phenolic composition and antioxidant capacity of pomaces from four grape varieties (Vitis vinifera L.). J. Sci. Food Agric. 2015, 95 (7), 1521−1527. (14) Makris, D. P.; Boskou, G.; Andrikopoulos, N. K. Polyphenolic content and in vitro antioxidant characteristics of wine industry and other agri-food solid waste extracts. J. Food Compos. Anal. 2007, 20 (2), 125−132. (15) Bosso, A.; Guaita, M.; Petrozziello, M. Influence of solvents on the composition of condensed tannins in grape pomace seed extracts. Food Chem. 2016, 207, 162−169. (16) New Zealand Winegrowers’ Annual Report, 2015. New Zealand Wine. http://www.nzwine.com/assets/sm/upload/rj/wk/2d/01/ NZW%20Annual%20Report%202015.pdf (accessed August 2016). (17) Olejar, K. J.; Fedrizzi, B.; Kilmartin, P. A. Influence of harvesting technique and maceration process on aroma and phenolic attributes of Sauvignon blanc wine. Food Chem. 2015, 183, 181−189. (18) Waterhouse, A. L. Wine phenolics. Ann. N. Y. Acad. Sci. 2002, 957, 21−36. (19) Roland, A.; Schneider, R.; Charrier, F.; Cavelier, F.; Rossignol, M.; Razungles, A. Distribution of varietal thiol precursors in the skin and the pulp of Melon B. and Sauvignon Blanc grapes. Food Chem. 2011, 125 (1), 139−144. (20) Roland, A.; Schneider, R.; Razungles, A.; Cavelier, F. Varietal thiols in wine: discovery, analysis and applications. Chem. Rev. 2011, 111 (11), 7355−7376. (21) Villegas, T. R.; Tonidandel, L.; Fedrizzi, B.; Larcher, R.; Nicolini, R. Novel technological strategies to enhance tropical thiol precursors in winemaking by-products. Food Chem. 2016, 207, 16−19. (22) Jouanneau, S.; Weaver, R. J.; Nicolau, L.; Herbst-Johnstone, M.; Benkwitz, F.; Kilmartin, P. A. Subregional survey of aroma compounds in Marlborough Sauvignon Blanc wines. Aust. J. Grape Wine Res. 2012, 18 (3), 329−343. (23) Winter, G.; Van Der Westhuizen, T.; Higgins, V. J.; Curtin, C.; Ugliano, M. Contribution of cysteine and glutathione conjugates to the formation of the volatile thiols 3-mercaptohexan-1-ol (3MH) and 3mercaptohexyl acetate (3MHA) during fermentation by Saccharomyces cerevisiae. Aust. J. Grape Wine Res. 2011, 17 (2), 285−290. (24) Harsch, M. J.; Benkwitz, F.; Frost, A.; Colonna-Ceccaldi, B.; Gardner, R. C.; Salmon, J.-M. New Precursor of 3-Mercaptohexan-1-ol in Grape Juice: Thiol-Forming Potential and Kinetics during Early H
DOI: 10.1021/acssuschemeng.6b01551 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX
