Article pubs.acs.org/JAFC
High Concentrations of Furan Fatty Acids in Organic Butter Samples from the German Market Christine Wendlinger and Walter Vetter* Institute of Food Chemistry (170b), University of Hohenheim, Garbenstraße 28, D-70599 Stuttgart, Germany S Supporting Information *
ABSTRACT: Furan fatty acids (F-acids) are valuable antioxidants containing a furan moiety in the central part of the molecule. They occur in the lipids of different foodstuffs and plants, with grass being the main source for their presence in milk fat and butter. Because cows from organic farming receive higher portions of grass-based feed, it was tested whether organic butter samples (n = 26) contain more F-acids than conventional ones (n = 25) in Germany. For this purpose, samples were melted, and the lipid phase was separated and transesterified into methyl esters, which were enriched using silver ion chromatography and analyzed by GC-EI/MS in the selected ion monitoring (SIM) mode. Levels of F-acids in butter were higher in summer than in winter, and in both seasons, organic samples contained significantly higher levels of F-acids than conventional ones (one-way ANOVA: p < 0.001). Furthermore, the daily intake of F-acids via milk fat and other foodstuffs was calculated. KEYWORDS: furan fatty acids, organic butter, conventional butter, silver ion chromatography, daily intake
■
■
INTRODUCTION Recent lipid research has shown that not only major fatty acids but also different classes of bioactive minor fatty acids, such as furan fatty acids (F-acids), contribute to the nutritional benefits of fats or oils.1−3 F-acids are fatty acids containing a furan moiety in the central part of the acyl chain. The α-position of the furan ring carries a carboxyalkyl chain typically with 7, 9, 11, or 13 carbons, and the α′-position is substituted with a propyl or pentyl chain.1 In addition, the β- and β′-positions may be substituted with one (methyl-substituted F-acids) or two methyl groups (dimethyl-substituted F-acids). F-acids are known to serve as hydroxyl and peroxyl radical scavengers and thus are able to protect polyunsaturated fatty acids from lipid peroxidation.4,5 Furthermore, the anti-inflammatory activity of F-acids had been confirmed in vivo.6 These natural antioxidants occur widely in foods including fish,7−10 olive oil,11 soybean oil,12−14 rapeseed oil,12 and butter.15 Plants are able to synthesize F-acids from acetate, methionine, and oxygen from air, whereas F-acids in fish originate from the intake of algae and those in milk fat from the intake of grass and other plants by cows.1 Concentrations of F-acids in butter ranged from 4.5 to 47.6 mg/100 g.15,16 High amounts of F-acids also occurred in grass,17 and because F-acids in milk fat originate from the cow’s feed, we aimed to investigate whether there is a difference in the concentrations of F-acids in butter samples from organic and conventional farming due to higher proportions of pasture and hay fed in organic farming. For this purpose, F-acid methyl esters of 51 butter samples were enriched using silver ion microcolumn chromatography and analyzed by gas chromatography with mass spectrometry (GC-MS) operated in the selected ion monitoring (SIM) mode according to the method of Vetter et al.16 On the basis of our results for butter and published concentrations of F-acids in other foodstuffs, we estimated the dietary daily intake of F-acids, as well as the corresponding share of milk fat in Germany. © XXXX American Chemical Society
MATERIALS AND METHODS
Samples. Butter samples (26 organic and 25 conventional ones) were purchased from different supermarkets located in Stuttgart, Germany, between January 2012 and January 2014. Most of the samples originated from southwestern, southeastern, and northeastern Germany. Samples from summer and winter were included in the study. Information about the type of butter (sweet or sour cream), the farming system (organic or conventional), and the minimum durability dates (MDD) were provided on the packaging. Samples were assigned to winter/summer season by backdating MDD for 2 months to take into account the period from milking to the date printed as MDD on the label. The “summer” period was set from May to October and “winter” from November to April. This division resulted in four groups: winter conventional (n = 14), winter organic (n = 13), summer conventional (n = 11), and summer organic (n = 13). Chemicals and Standards. Methanol and n-hexane (both of HPLC grade, ≥95% purity) were from Th. Geyer (Renningen, Germany). Sulfuric acid was from BASF (Ludwigshafen, Germany). Diethyl ether (for synthesis, ≥99%) and NaCl (≥99.5%) were from Carl Roth (Karlsruhe, Germany). Silica gel 60, AgNO3, and myristic acid were from Sigma-Aldrich (Steinheim, Germany). The abbreviated nomenclature and preparation of the internal standard (IS) 10,13epoxy-11-methyloctadeca-10,12-dienoic acid ethyl ester (9M5-EE) were according to Vetter et al.16 Myristic acid ethyl ester (14:0-EE) was prepared by heating myristic acid with 1% sulfuric acid in ethanol. Sample Preparation and Transesterification. Butter samples (∼4 g) were melted at ∼70 °C, and the supernatant clear lipid phase was separated (∼2 g). For the formation of fatty acid methyl esters (FAME) about ∼20 mg per sample was treated with 1.5 mL of 1% sulfuric acid in methanol in a small sealed tube at 80 °C for 1 h, which was shaken from time to time. Then, 1 mL of saturated aqueous NaCl solution and 1 mL of demineralized water were added, and FAME were extracted once using 2 mL of n-hexane. Received: June 23, 2014 Revised: August 6, 2014 Accepted: August 7, 2014
A
dx.doi.org/10.1021/jf502975b | J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Journal of Agricultural and Food Chemistry
Article
Enrichment of F-Acids by Silver Ion Micro Column Chromatography. Silver ion microcolumn chromatography was performed according to the method of Vetter et al. with slight amendments.16 In brief, an aliquot of the organic layer (∼180−250 μL) was supplemented with 0.87 μg of the internal standard 9M5-EE, and the solution was placed on a Pasteur pipet (the tip was plugged with glass wool) containing ∼0.6 g of 1% deactivated silica gel with 20% AgNO3. Fraction 1 containing saturated and the majority of the monounsaturated fatty acids was eluted with 15 mL of n-hexane/ diethyl ether (99.5:0.5, v/v). Fraction 2 containing F-acids was eluted with 15 mL of n-hexane/diethyl ether (97:3, v/v). Both fractions were individually evaporated at 38 °C by means of a gentle stream of nitrogen, the residue was transferred with n-hexane into a GC vial, and the volume of the solution was adjusted to 1 mL. Afterward, the second internal standard, 14:0-EE, was added, and the solution was analyzed by GC-MS. During the sample preparation, oxidation of Facids was minimized by using amber glass and performing each step under the lowest possible lighting level. Gas Chromatography Coupled to Electron Ionization Mass Spectrometry (GC-EI/MS). All analyses were carried out with a 5890 series II/5971A GC/MS system in combination with a 7673A autosampler (Hewlett-Packard/Agilent, Waldbronn, Germany). The carrier gas, helium (purity 5.0), was transported at a flow rate of 1 mL/ min through two serially coupled 30 m columns (Rtx-2330, internal diameter = 0.25 mm, coated with 0.1 μm of 10% cyanopropylphenyl, 90% biscyanopropyl polysiloxane, Restek, Bellefonte, PA, USA). The GC oven program started at 60 °C for 1 min. It followed ramps of (i) 13 °C/min to 180 °C, (ii) 3 °C/min to 225 °C, and (iii) 20 °C/min to 250 °C. The final temperature was held for 5 min, and the solvent delay was set to 8 min. GC-MS measurements in the full scan mode covered m/z 50−500. In the SIM mode four ions (m/z 137, 151, 165, and 179) were measured throughout the run. Additionally, three further ions, divided into four time windows, were recorded: (I) 8−13 min, m/z 74, 88, 101; (II) 13−19.5 min, m/z 74, 294, 308; (III) 19.5− 23 min, m/z 322, 336, 350; (IV) 23−31 min, m/z 364, 378, 392. Peak identification was carried out in the full scan mode, whereas quantification was performed in the SIM mode using 14:0-EE and 9M5-EE as internal standards.16 Quantification Mode, Data Reporting, and Statistical Analysis. The IS 9M5-EE was calibrated against 12,15-epoxy-13methyleicosa-12,14-dienoic acid ethyl ester (11M5-EE) and 12,15epoxy-13,14-dimethyleicosa-10,12-dienoic acid ethyl ester (11D5-EE). Because the peak areas at the same concentration did not deviate by more than 5%, 9M5-EE was used for the quantification of all F-acids as methyl esters. The quantitative results were further verified by standard addition experiments using 11D5-ME. All preparation steps were performed in duplicate, and for each sample mean values were calculated (individual values of replicates did not vary by more than 10%); the results were expressed as milligrams per 100 g of fat. The limit of quantification (LOQ) and the limit of detection (LOD) were 0.4 and 0.12 mg/100 g fat, respectively. For the calculation of total concentrations of F-acids in the samples, individual values below LOD were taken into account by using half the LOD (i.e., 0.06 mg/100 g fat). The F-acid concentrations were evaluated by one-way ANOVA as integrated in EXCEL. Significance was based on a probability of 0.05.
Figure 1. Frequency of the number of F-acids in butter samples from the German market.
dimethyleicosa-14,16-dienoic acid (13D3). Depending on the abundance of the individual F-acids, most commonly four (7D5, 9M5, 9D5, and 11D5; 18 samples) or five (additionally 11M5; 17 samples) F-acids were present in the samples (Figure 1; Table S1). In five butter samples only two F-acids and in nine samples only three F-acids were detectable, whereas six Facids were present in two cases (Figure 1; Table S1). A similar F-acid pattern was also reported in grass (blade).16,17 Thus, it seems obvious that the occurrence of (especially dimethylsubstituted) F-acids in grass lipids and pasture was linked with the presence of F-acids in milk and butter fat. Concentrations of F-Acids in Organically or Conventionally Produced Butter Samples. The total F-acid content in the samples ranged from 1.6 to 16.5 mg/100 g fat. Sweet cream and sour cream butter samples showed no difference in the F-acid pattern, yet the concentrations of F-acids were 3-fold lower than those reported by Guth and Grosch.15 Concentrations of F-acids in organic butter samples (5.4−16.5 mg/100 g fat, n = 26) were significantly higher than in conventional samples (1.6−7.1 mg/100 g fat, n = 25) (one-way ANOVA: p < 0.001). The same result could be obtained by the sole evaluation of the summed concentration of 9D5 and 11D5 (one-way ANOVA: p < 0.001). However, 13 organic samples (all winter samples) with the lowest content of F-acids showed lower concentrations than the conventional sample with the highest F-acid content. This overlap did not allow unequivocally distinguishing organic from conventional butter samples. Subsequently, the samples were further divided depending on the season into four groups (see Materials and Methods), that is, winter conventional (n = 14), winter organic (n = 13), summer conventional (n = 11), and summer organic (n = 13) (Table 1). Organic samples collected in summer (May− October) generally contained higher amounts of F-acids than samples from the winter months (November−April) (Table 1; one-way ANOVA: p < 0.001). Also, no overlap of the F-acid concentration range was observed for conventional samples collected in summer versus winter (Table 1; one-way ANOVA: p < 0.001). The widest ranges for F-acid concentrations were observed for conventional samples in winter and organic samples in summer, whereas ranges in organic butter samples collected in winter and conventional ones of the summer season were surprisingly small (Table 1). These wider ranges in the F-acid content could be due to a higher flexibility in the composition of the feed. Analysis of samples from the same producer in summer and winter showed the typical seasonal dependency of the F-acid concentrations. Accordingly, the
■
RESULTS AND DISCUSSION F-Acid Variety and Pattern in Butter Samples. Between two and six F-acids were detected by GC-MS in the 51 butter samples (Figure 1; Supporting Information Table S1). Highest concentrations were generally determined for 10,13-epoxy11,12-dimethyloctadeca-10,12-dienoic acid (9D5) and 11D5, which represented 60−100% of the total amount of F-acids in the samples. A clear dominance of dimethyl-substituted F-acids (11D5 and 9D5) over methyl-substituted F-acids was also reported in previous studies.15,16 These two major F-acids were partly accompanied by 11M5, 8,11-epoxy-9,10-dimethylhexadeca-8,10-dienoic acid (7D5), 9M5, and 14,17-epoxy-15,16B
dx.doi.org/10.1021/jf502975b | J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Journal of Agricultural and Food Chemistry
Article
Table 1. F-Acid Concentrations in Butter Samples, Dependent on Season (Summer from May to October, Winter from November to April, According to the Backdated Minimum Durability Dates) winter samples (mg/100 g fat) F-acid 7D5 9M5 9D5 11M5 11D5 13D3
Σ mean (Σ) median (Σ)
conventional (n=14)
organic (n=13)
summer samples (mg/100 g fat) conventional (n=11)
organic (n=13)
0.1a−0.9 0.1a−0.9 0.1a−1.9 0.1a−0.5 (n = 1) 0.6−1.8 0.1a
0.6−1.2 0.1a−1.2 1.7−2.5 0.1a−0.8
0.1a−1.1 0.1a−1.1 1.7−2.6 0.1a−0.6
0.7−2.2 0.5−1.5 1.7−5.6 0.1a−1.1
1.8−2.5 0.1a
1.8−2.8 0.1a−0.4 (n = 1)
2.8−6.4 0.1a−0.4 (n = 1)
1.6−5.1 3.4 3.4
5.4−6.6 5.9 6.0
5.2−7.1 6.0 6.0
7.5−16.5 10.6 9.5
a
Values below LOQ were included to sum with 1/2 LOD (i.e., 0.06 mg/100 g fat).
processing technology of butter did not have a primary effect on the F-acid concentrations in butter. Moreover, the potential impact of the geographical origin of the milk samples was evaluated as follows. The four previously mentioned groups were further divided according to their geographical origins in southeastern, southwestern, and northeastern Germany. Irrespective of the sample group, the F-acid concentration ranges of the geographical groups were overlapping, and data analysis by one-way ANOVA did not show significant deviations (p > 0.2). Accordingly, organic or conventional farming systems could not be differentiated by their geographical position in Germany. However, by taking the season into account, it was possible to distinguish organic from conventional samples without exception (Figure 2). Thus, the concentrations of F-acids may serve as a useful marker for the authentication of organic butter, especially in combination with other analytical markers such as the concentration of n-3 fatty acids,18−21 the concentration and diastereomer ratio of phytanic acid,22−25 the δ13C values of milk fat,26,27 or the fatty acid and triacylglycerol profiles.28,29 Nutritional Intake of F-Acids in Germany. Our data suggested that the intake of F-acids through milk and dairy products may vary up to the factor of 3, depending on the season and farming system (Table 1). From these findings we estimated the daily intake of F-acids by the consumption of milk and butter, as well as their percentage contribution of the nutritional intake. The average daily milk fat intake in Germany was reported to be about 20−45 g per day.30 On the basis of the mean values (Table 1), the daily intake of F-acids via milk fat amounted to 0.7−1.5 mg (full range, 0.3−2.3 mg; 1.1 mg mean of mean, which corresponds to 32.5 g milk fat) for conventional samples in winter, 1.2−2.7 mg (full range, 1.1−3.0 mg, mean of mean 1.9 mg) for organic samples in winter, 1.2−2.7 mg (full range, 1.0−3.2 mg, mean of mean 2.0 mg) for conventional samples in summer, and up to 2.1−4.8 mg (full range, 1.5−7.4 mg, mean of mean 3.4 mg) for organic ones in summer (Figure 3). Thus, the daily intake of F-acids via milk fat was highest for organic samples in summer.
Figure 2. Concentration ranges of F-acids (mg/100 g fat) in organic and conventional butter samples collected in summer and winter in Germany.
Figure 3. F-acid intake per capita and day (mg) from different food items in Germany with mean value (line in the middle) and minimum to maximum ranges.
According to national consumption data in Germany, the average per capita consumption of fish was about 40 g/day in 2012.31 The intake of F-acids via fish was calculated on the basis of the six most popular fish species consumed in Germany: Theragra chalcogramma, Alaska pollock; Clupea harengus, herring; Salmo salar, salmon; Thunnus, tuna; Pangasius hypophthalmus, iridescent shark; and Salmo trutta, trout.31 These six species contributed ∼80% to total fish consumption in Germany.31 On the basis of these species, fish consumed in Germany contained an average of ∼8% lipids (weighted average). With an average F-acid content of 200 mg/100 g lipids (own determinations, data not shown) or 500 mg/100 g lipids (according to Pacetti et al.10), the daily intake of F-acids C
dx.doi.org/10.1021/jf502975b | J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Journal of Agricultural and Food Chemistry
Article
by the consumption of fish accounted for ∼6.6−16.5 mg (Figure 3). However, it should be noted that these values are calculated for raw fish and that cooking may result in lower Facid contents due to their partial decomposition. About 1.7 g of olive oil was consumed per capita and day in Germany in 2009.32 According to Boselli et al. the average Facid content in olive oil was about 2 mg/kg.11 On the basis of these data, the daily F-acid intake through olive oil amounted to 0.003 mg. In Germany, the daily intake for rapeseed oil was 14 g and for soybean oil, 15 g, in 2009.32 The average F-acid content was determined to be 1.3−3.4 mg/100 g in rapeseed oil, 26.3−42.4 mg/100 g in crude soybean oil, and 9.3−16.8 mg/100 g in refined soybean oil.12 Thus, the F-acid intake via refined soybean oil, which is the primary source for human consumption, is much lower than the calculated values for crude soybean oil (Figure 3). Accordingly, the F-acid intake per capita and day would be 0.2−0.5 mg via rapeseed oil and 1.4− 2.5 mg via refined soybean oil (Figure 3). Accordingly, the intake of F-acids by the consumption of vegetables and fruits is rather low.1 However, plants are the origin of F-acids in both milk and fish.1 On the basis of these individual contributions of different foodstuffs to the overall nutritional daily intake of F-acids, it was apparent that intake based on raw fish represented the highest single source (Figure 3). Compared to fish, the F-acid intake via milk fat was relatively small for winter conventional samples (8 vs 77% from fish to the total F-acid intake), whereas organic samples milked during the summer had a larger share (20 vs 67% from fish to the total F-acid intake). By contrast, olive oil did not play any noticeable role in the dietary intake of F-acids in Germany. Even for Mediterranean diets with a per capita consumption for olive oil of about 38 g/day,32 the F-acid intake would be only 0.08 mg/day, that is,
High Concentrations of Furan Fatty Acids in Organic Butter Samples from the German Market Christine Wendlinger and Walter Vetter* Institute of Food Chemistry (170b), University of Hohenheim, Garbenstraße 28, D-70599 Stuttgart, Germany S Supporting Information *
ABSTRACT: Furan fatty acids (F-acids) are valuable antioxidants containing a furan moiety in the central part of the molecule. They occur in the lipids of different foodstuffs and plants, with grass being the main source for their presence in milk fat and butter. Because cows from organic farming receive higher portions of grass-based feed, it was tested whether organic butter samples (n = 26) contain more F-acids than conventional ones (n = 25) in Germany. For this purpose, samples were melted, and the lipid phase was separated and transesterified into methyl esters, which were enriched using silver ion chromatography and analyzed by GC-EI/MS in the selected ion monitoring (SIM) mode. Levels of F-acids in butter were higher in summer than in winter, and in both seasons, organic samples contained significantly higher levels of F-acids than conventional ones (one-way ANOVA: p < 0.001). Furthermore, the daily intake of F-acids via milk fat and other foodstuffs was calculated. KEYWORDS: furan fatty acids, organic butter, conventional butter, silver ion chromatography, daily intake
■
■
INTRODUCTION Recent lipid research has shown that not only major fatty acids but also different classes of bioactive minor fatty acids, such as furan fatty acids (F-acids), contribute to the nutritional benefits of fats or oils.1−3 F-acids are fatty acids containing a furan moiety in the central part of the acyl chain. The α-position of the furan ring carries a carboxyalkyl chain typically with 7, 9, 11, or 13 carbons, and the α′-position is substituted with a propyl or pentyl chain.1 In addition, the β- and β′-positions may be substituted with one (methyl-substituted F-acids) or two methyl groups (dimethyl-substituted F-acids). F-acids are known to serve as hydroxyl and peroxyl radical scavengers and thus are able to protect polyunsaturated fatty acids from lipid peroxidation.4,5 Furthermore, the anti-inflammatory activity of F-acids had been confirmed in vivo.6 These natural antioxidants occur widely in foods including fish,7−10 olive oil,11 soybean oil,12−14 rapeseed oil,12 and butter.15 Plants are able to synthesize F-acids from acetate, methionine, and oxygen from air, whereas F-acids in fish originate from the intake of algae and those in milk fat from the intake of grass and other plants by cows.1 Concentrations of F-acids in butter ranged from 4.5 to 47.6 mg/100 g.15,16 High amounts of F-acids also occurred in grass,17 and because F-acids in milk fat originate from the cow’s feed, we aimed to investigate whether there is a difference in the concentrations of F-acids in butter samples from organic and conventional farming due to higher proportions of pasture and hay fed in organic farming. For this purpose, F-acid methyl esters of 51 butter samples were enriched using silver ion microcolumn chromatography and analyzed by gas chromatography with mass spectrometry (GC-MS) operated in the selected ion monitoring (SIM) mode according to the method of Vetter et al.16 On the basis of our results for butter and published concentrations of F-acids in other foodstuffs, we estimated the dietary daily intake of F-acids, as well as the corresponding share of milk fat in Germany. © XXXX American Chemical Society
MATERIALS AND METHODS
Samples. Butter samples (26 organic and 25 conventional ones) were purchased from different supermarkets located in Stuttgart, Germany, between January 2012 and January 2014. Most of the samples originated from southwestern, southeastern, and northeastern Germany. Samples from summer and winter were included in the study. Information about the type of butter (sweet or sour cream), the farming system (organic or conventional), and the minimum durability dates (MDD) were provided on the packaging. Samples were assigned to winter/summer season by backdating MDD for 2 months to take into account the period from milking to the date printed as MDD on the label. The “summer” period was set from May to October and “winter” from November to April. This division resulted in four groups: winter conventional (n = 14), winter organic (n = 13), summer conventional (n = 11), and summer organic (n = 13). Chemicals and Standards. Methanol and n-hexane (both of HPLC grade, ≥95% purity) were from Th. Geyer (Renningen, Germany). Sulfuric acid was from BASF (Ludwigshafen, Germany). Diethyl ether (for synthesis, ≥99%) and NaCl (≥99.5%) were from Carl Roth (Karlsruhe, Germany). Silica gel 60, AgNO3, and myristic acid were from Sigma-Aldrich (Steinheim, Germany). The abbreviated nomenclature and preparation of the internal standard (IS) 10,13epoxy-11-methyloctadeca-10,12-dienoic acid ethyl ester (9M5-EE) were according to Vetter et al.16 Myristic acid ethyl ester (14:0-EE) was prepared by heating myristic acid with 1% sulfuric acid in ethanol. Sample Preparation and Transesterification. Butter samples (∼4 g) were melted at ∼70 °C, and the supernatant clear lipid phase was separated (∼2 g). For the formation of fatty acid methyl esters (FAME) about ∼20 mg per sample was treated with 1.5 mL of 1% sulfuric acid in methanol in a small sealed tube at 80 °C for 1 h, which was shaken from time to time. Then, 1 mL of saturated aqueous NaCl solution and 1 mL of demineralized water were added, and FAME were extracted once using 2 mL of n-hexane. Received: June 23, 2014 Revised: August 6, 2014 Accepted: August 7, 2014
A
dx.doi.org/10.1021/jf502975b | J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Journal of Agricultural and Food Chemistry
Article
Enrichment of F-Acids by Silver Ion Micro Column Chromatography. Silver ion microcolumn chromatography was performed according to the method of Vetter et al. with slight amendments.16 In brief, an aliquot of the organic layer (∼180−250 μL) was supplemented with 0.87 μg of the internal standard 9M5-EE, and the solution was placed on a Pasteur pipet (the tip was plugged with glass wool) containing ∼0.6 g of 1% deactivated silica gel with 20% AgNO3. Fraction 1 containing saturated and the majority of the monounsaturated fatty acids was eluted with 15 mL of n-hexane/ diethyl ether (99.5:0.5, v/v). Fraction 2 containing F-acids was eluted with 15 mL of n-hexane/diethyl ether (97:3, v/v). Both fractions were individually evaporated at 38 °C by means of a gentle stream of nitrogen, the residue was transferred with n-hexane into a GC vial, and the volume of the solution was adjusted to 1 mL. Afterward, the second internal standard, 14:0-EE, was added, and the solution was analyzed by GC-MS. During the sample preparation, oxidation of Facids was minimized by using amber glass and performing each step under the lowest possible lighting level. Gas Chromatography Coupled to Electron Ionization Mass Spectrometry (GC-EI/MS). All analyses were carried out with a 5890 series II/5971A GC/MS system in combination with a 7673A autosampler (Hewlett-Packard/Agilent, Waldbronn, Germany). The carrier gas, helium (purity 5.0), was transported at a flow rate of 1 mL/ min through two serially coupled 30 m columns (Rtx-2330, internal diameter = 0.25 mm, coated with 0.1 μm of 10% cyanopropylphenyl, 90% biscyanopropyl polysiloxane, Restek, Bellefonte, PA, USA). The GC oven program started at 60 °C for 1 min. It followed ramps of (i) 13 °C/min to 180 °C, (ii) 3 °C/min to 225 °C, and (iii) 20 °C/min to 250 °C. The final temperature was held for 5 min, and the solvent delay was set to 8 min. GC-MS measurements in the full scan mode covered m/z 50−500. In the SIM mode four ions (m/z 137, 151, 165, and 179) were measured throughout the run. Additionally, three further ions, divided into four time windows, were recorded: (I) 8−13 min, m/z 74, 88, 101; (II) 13−19.5 min, m/z 74, 294, 308; (III) 19.5− 23 min, m/z 322, 336, 350; (IV) 23−31 min, m/z 364, 378, 392. Peak identification was carried out in the full scan mode, whereas quantification was performed in the SIM mode using 14:0-EE and 9M5-EE as internal standards.16 Quantification Mode, Data Reporting, and Statistical Analysis. The IS 9M5-EE was calibrated against 12,15-epoxy-13methyleicosa-12,14-dienoic acid ethyl ester (11M5-EE) and 12,15epoxy-13,14-dimethyleicosa-10,12-dienoic acid ethyl ester (11D5-EE). Because the peak areas at the same concentration did not deviate by more than 5%, 9M5-EE was used for the quantification of all F-acids as methyl esters. The quantitative results were further verified by standard addition experiments using 11D5-ME. All preparation steps were performed in duplicate, and for each sample mean values were calculated (individual values of replicates did not vary by more than 10%); the results were expressed as milligrams per 100 g of fat. The limit of quantification (LOQ) and the limit of detection (LOD) were 0.4 and 0.12 mg/100 g fat, respectively. For the calculation of total concentrations of F-acids in the samples, individual values below LOD were taken into account by using half the LOD (i.e., 0.06 mg/100 g fat). The F-acid concentrations were evaluated by one-way ANOVA as integrated in EXCEL. Significance was based on a probability of 0.05.
Figure 1. Frequency of the number of F-acids in butter samples from the German market.
dimethyleicosa-14,16-dienoic acid (13D3). Depending on the abundance of the individual F-acids, most commonly four (7D5, 9M5, 9D5, and 11D5; 18 samples) or five (additionally 11M5; 17 samples) F-acids were present in the samples (Figure 1; Table S1). In five butter samples only two F-acids and in nine samples only three F-acids were detectable, whereas six Facids were present in two cases (Figure 1; Table S1). A similar F-acid pattern was also reported in grass (blade).16,17 Thus, it seems obvious that the occurrence of (especially dimethylsubstituted) F-acids in grass lipids and pasture was linked with the presence of F-acids in milk and butter fat. Concentrations of F-Acids in Organically or Conventionally Produced Butter Samples. The total F-acid content in the samples ranged from 1.6 to 16.5 mg/100 g fat. Sweet cream and sour cream butter samples showed no difference in the F-acid pattern, yet the concentrations of F-acids were 3-fold lower than those reported by Guth and Grosch.15 Concentrations of F-acids in organic butter samples (5.4−16.5 mg/100 g fat, n = 26) were significantly higher than in conventional samples (1.6−7.1 mg/100 g fat, n = 25) (one-way ANOVA: p < 0.001). The same result could be obtained by the sole evaluation of the summed concentration of 9D5 and 11D5 (one-way ANOVA: p < 0.001). However, 13 organic samples (all winter samples) with the lowest content of F-acids showed lower concentrations than the conventional sample with the highest F-acid content. This overlap did not allow unequivocally distinguishing organic from conventional butter samples. Subsequently, the samples were further divided depending on the season into four groups (see Materials and Methods), that is, winter conventional (n = 14), winter organic (n = 13), summer conventional (n = 11), and summer organic (n = 13) (Table 1). Organic samples collected in summer (May− October) generally contained higher amounts of F-acids than samples from the winter months (November−April) (Table 1; one-way ANOVA: p < 0.001). Also, no overlap of the F-acid concentration range was observed for conventional samples collected in summer versus winter (Table 1; one-way ANOVA: p < 0.001). The widest ranges for F-acid concentrations were observed for conventional samples in winter and organic samples in summer, whereas ranges in organic butter samples collected in winter and conventional ones of the summer season were surprisingly small (Table 1). These wider ranges in the F-acid content could be due to a higher flexibility in the composition of the feed. Analysis of samples from the same producer in summer and winter showed the typical seasonal dependency of the F-acid concentrations. Accordingly, the
■
RESULTS AND DISCUSSION F-Acid Variety and Pattern in Butter Samples. Between two and six F-acids were detected by GC-MS in the 51 butter samples (Figure 1; Supporting Information Table S1). Highest concentrations were generally determined for 10,13-epoxy11,12-dimethyloctadeca-10,12-dienoic acid (9D5) and 11D5, which represented 60−100% of the total amount of F-acids in the samples. A clear dominance of dimethyl-substituted F-acids (11D5 and 9D5) over methyl-substituted F-acids was also reported in previous studies.15,16 These two major F-acids were partly accompanied by 11M5, 8,11-epoxy-9,10-dimethylhexadeca-8,10-dienoic acid (7D5), 9M5, and 14,17-epoxy-15,16B
dx.doi.org/10.1021/jf502975b | J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Journal of Agricultural and Food Chemistry
Article
Table 1. F-Acid Concentrations in Butter Samples, Dependent on Season (Summer from May to October, Winter from November to April, According to the Backdated Minimum Durability Dates) winter samples (mg/100 g fat) F-acid 7D5 9M5 9D5 11M5 11D5 13D3
Σ mean (Σ) median (Σ)
conventional (n=14)
organic (n=13)
summer samples (mg/100 g fat) conventional (n=11)
organic (n=13)
0.1a−0.9 0.1a−0.9 0.1a−1.9 0.1a−0.5 (n = 1) 0.6−1.8 0.1a
0.6−1.2 0.1a−1.2 1.7−2.5 0.1a−0.8
0.1a−1.1 0.1a−1.1 1.7−2.6 0.1a−0.6
0.7−2.2 0.5−1.5 1.7−5.6 0.1a−1.1
1.8−2.5 0.1a
1.8−2.8 0.1a−0.4 (n = 1)
2.8−6.4 0.1a−0.4 (n = 1)
1.6−5.1 3.4 3.4
5.4−6.6 5.9 6.0
5.2−7.1 6.0 6.0
7.5−16.5 10.6 9.5
a
Values below LOQ were included to sum with 1/2 LOD (i.e., 0.06 mg/100 g fat).
processing technology of butter did not have a primary effect on the F-acid concentrations in butter. Moreover, the potential impact of the geographical origin of the milk samples was evaluated as follows. The four previously mentioned groups were further divided according to their geographical origins in southeastern, southwestern, and northeastern Germany. Irrespective of the sample group, the F-acid concentration ranges of the geographical groups were overlapping, and data analysis by one-way ANOVA did not show significant deviations (p > 0.2). Accordingly, organic or conventional farming systems could not be differentiated by their geographical position in Germany. However, by taking the season into account, it was possible to distinguish organic from conventional samples without exception (Figure 2). Thus, the concentrations of F-acids may serve as a useful marker for the authentication of organic butter, especially in combination with other analytical markers such as the concentration of n-3 fatty acids,18−21 the concentration and diastereomer ratio of phytanic acid,22−25 the δ13C values of milk fat,26,27 or the fatty acid and triacylglycerol profiles.28,29 Nutritional Intake of F-Acids in Germany. Our data suggested that the intake of F-acids through milk and dairy products may vary up to the factor of 3, depending on the season and farming system (Table 1). From these findings we estimated the daily intake of F-acids by the consumption of milk and butter, as well as their percentage contribution of the nutritional intake. The average daily milk fat intake in Germany was reported to be about 20−45 g per day.30 On the basis of the mean values (Table 1), the daily intake of F-acids via milk fat amounted to 0.7−1.5 mg (full range, 0.3−2.3 mg; 1.1 mg mean of mean, which corresponds to 32.5 g milk fat) for conventional samples in winter, 1.2−2.7 mg (full range, 1.1−3.0 mg, mean of mean 1.9 mg) for organic samples in winter, 1.2−2.7 mg (full range, 1.0−3.2 mg, mean of mean 2.0 mg) for conventional samples in summer, and up to 2.1−4.8 mg (full range, 1.5−7.4 mg, mean of mean 3.4 mg) for organic ones in summer (Figure 3). Thus, the daily intake of F-acids via milk fat was highest for organic samples in summer.
Figure 2. Concentration ranges of F-acids (mg/100 g fat) in organic and conventional butter samples collected in summer and winter in Germany.
Figure 3. F-acid intake per capita and day (mg) from different food items in Germany with mean value (line in the middle) and minimum to maximum ranges.
According to national consumption data in Germany, the average per capita consumption of fish was about 40 g/day in 2012.31 The intake of F-acids via fish was calculated on the basis of the six most popular fish species consumed in Germany: Theragra chalcogramma, Alaska pollock; Clupea harengus, herring; Salmo salar, salmon; Thunnus, tuna; Pangasius hypophthalmus, iridescent shark; and Salmo trutta, trout.31 These six species contributed ∼80% to total fish consumption in Germany.31 On the basis of these species, fish consumed in Germany contained an average of ∼8% lipids (weighted average). With an average F-acid content of 200 mg/100 g lipids (own determinations, data not shown) or 500 mg/100 g lipids (according to Pacetti et al.10), the daily intake of F-acids C
dx.doi.org/10.1021/jf502975b | J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Journal of Agricultural and Food Chemistry
Article
by the consumption of fish accounted for ∼6.6−16.5 mg (Figure 3). However, it should be noted that these values are calculated for raw fish and that cooking may result in lower Facid contents due to their partial decomposition. About 1.7 g of olive oil was consumed per capita and day in Germany in 2009.32 According to Boselli et al. the average Facid content in olive oil was about 2 mg/kg.11 On the basis of these data, the daily F-acid intake through olive oil amounted to 0.003 mg. In Germany, the daily intake for rapeseed oil was 14 g and for soybean oil, 15 g, in 2009.32 The average F-acid content was determined to be 1.3−3.4 mg/100 g in rapeseed oil, 26.3−42.4 mg/100 g in crude soybean oil, and 9.3−16.8 mg/100 g in refined soybean oil.12 Thus, the F-acid intake via refined soybean oil, which is the primary source for human consumption, is much lower than the calculated values for crude soybean oil (Figure 3). Accordingly, the F-acid intake per capita and day would be 0.2−0.5 mg via rapeseed oil and 1.4− 2.5 mg via refined soybean oil (Figure 3). Accordingly, the intake of F-acids by the consumption of vegetables and fruits is rather low.1 However, plants are the origin of F-acids in both milk and fish.1 On the basis of these individual contributions of different foodstuffs to the overall nutritional daily intake of F-acids, it was apparent that intake based on raw fish represented the highest single source (Figure 3). Compared to fish, the F-acid intake via milk fat was relatively small for winter conventional samples (8 vs 77% from fish to the total F-acid intake), whereas organic samples milked during the summer had a larger share (20 vs 67% from fish to the total F-acid intake). By contrast, olive oil did not play any noticeable role in the dietary intake of F-acids in Germany. Even for Mediterranean diets with a per capita consumption for olive oil of about 38 g/day,32 the F-acid intake would be only 0.08 mg/day, that is,
