Gas chromatographic investigation of acrylamide formation in browning model systems

Acrylamide formed in browning model systems was analyzed using a gas chromatograph with a nitrogen-phosphorus detector. Asparagine alone produced acrylamide via thermal degradation at the level of 0.99 microgram/g of asparagine. When asparagine was heated with triolein-which produced acrolein at the level of 1.82 +/- 0.31 (n = 5) mg/L of headspace by heat treatment-acrylamide was formed at the level of 88.6 microgram/g of asparagine. When acrolein gas was sprayed onto asparagine heated at 180 degrees C, a significant amount of acrylamide was formed (114 microgram/g of asparagine). On the other hand, when acrolein gas was sprayed onto glutamine under the same conditions, only a trace amount of acrylamide was formed (0.18 microgram/g of glutamine). Relatively high levels of acrylamide (753 microgram/g of ammonia) were formed from ammonia and acrolein heated at 180 degrees C in the vapor phase. The reaction of acrylic acid, which is an oxidation product of acrolein and ammonia, produced a high level of acrylamide (190 000 microgram/g of ammonia), suggesting that ammonia and acrolein play an important role in acrylamide formation in lipid-rich foods. Acrylamide can be formed from asparagine alone via thermal degradation, but carbonyl compounds, such as acrolein, promote its formation via a browning reaction.     

Development of two stable isotope dilution assays for the quantitation of acrolein in heat-processed fats

Two stable isotope dilution assays were developed for the quantitation of acrolein in fats and oils using [(13)C(3)]-acrolein as the internal standard. First, a direct GC-MS headspace method, followed by an indirect GC-MS method using derivatization with pentafluorophenyl hydrazine, was established. Analysis of six different types of oils varying in their pattern of fatty acids showed significant differences in the amounts of acrolein formed after heating at various temperatures and for various times. For example, after 24 h at 140 °C, coconut oil contained 6.7 mg/kg, whereas linseed oil was highest with 242.3 mg/kg. A comparison of the results showed that the extent of acrolein formation seemed to be correlated with the amount of linolenic acid in the oils. Although the acrolein concentrations were lowered in all six oils after frying of potato crisps, linseed and rapeseed oil still contained the highest amounts of acrolein after frying. By applying both methods on different thermally treated fats and oils, nearly identical quantitative data were obtained.     

Development of a stable isotope dilution assay for the quantification of 5-methyl-(E)-2-hepten-4-one: application to hazelnut oils and hazelnuts.

A stable isotope dilution assay was developed for the quantitation of the hazelnut odorant 5-methyl-(E)-2-hepten-4-one by mass chromatography using synthesized [(2)H](2)-5-methyl-(E)-2-hepten-4-one as the internal standard. Application of the method on two batches of commercial hazelnut oils, processed from either roasted or unroasted nuts, revealed 6.4 microg 5-methyl-(E)-2-hepten-4-one per kg of unroasted oil whereas 315.8 microg per kg was determined in the roasted nut oil. The about 50-fold higher amount of 5-methyl-(E)-2-hepten-4-one in roasted hazelnut oil suggested the necessity of a thermal treatment to generate the flavor compound. Pan frying of raw hazelnuts (9 to 15 min) or boiling of the crushed nut material for 1 h in water led to an increase of 5-methyl-(E)-2-hepten-4-one by factors of 600 and 800, respectively, thereby corroborating that the major part of the nut flavorant is formed during heat treatment from a yet unknown precursor in hazelnuts.     

Antioxidant activity of flavonoids isolated from young green barley leaves toward biological lipid samples

Natural plant flavonoids, saponarin/lutonarin=4.5/1, isolated from young green barley leaves were examined for their antioxidant activity using cod liver oil, omega-3 fatty acids, phospholipids, and blood plasma. The saponarin/lutonarin (S/L) mixture inhibited malonaldehyde (MA) formation from cod liver oil by 76.47+/-0.11% at a level of 1 micromol and 85.88+/-0.12% at a level of 8 micromol. The S/L mixture inhibited MA formation from the omega-3 fatty acids eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) by 45.60+/-1.08 and 69.24+/-0.24%, respectively, at a level of 8 micromol. The S/L mixture inhibited MA formation from the phospholipids lecithin I and II by 43.08+/-0.72 and 69.16+/-2.92%, respectively, at a level of 8 micromol. It also inhibited MA formation from blood plasma by 62.20+/-0.11% at a level of 8 micromol. The antioxidant activities obtained from the S/L mixture were comparable to those obtained from alpha-tocopherol and butylated hydroxy toluene (BHT) in all lipids tested.     

Liquid chromatographic determination of total formaldehyde in drinking water

Formaldehyde is readily quantitated at micrograms/L levels in drinking water. The analyte present in 1 L water samples is derivatized with 2,4-dinitrophenylhydrazine in a 2M acid medium and then extracted with chloroform. After the solvent is exchanged for methanol, the product is separated and quantitated using reverse-phase liquid chromatography with UV detection (365 nm). Reporting limits as low as 20 micrograms/L (corrected for laboratory blank) are routinely achieved. Formaldehyde recovery typically exceeds 90% at 20-200 micrograms/L. The method was applied to hot and cold water samples from thirty-four 1- and 2-story houses equipped with poly(acetal) plumbing elbows and tees. The drinking water samples were accompanied by sets of blanks and spikes specified by a quality assurance/quality control plan. Formaldehyde was observed above the reporting limit in 80% of hot and cold water samples from the 1-story dwellings, but in less than 50% of those from the 2-story dwellings. The results may depend on both the construction of the houses (1- vs 2-story) and the time of year the water is sampled (mid-summer).     

Formaldehyde quantitation in air samples by thiazolidine derivatization: factors affecting analysis

A new method for the determination of trace levels of formaldehyde in air was developed and validated. The method is based on the reaction of formaldehyde with cysteamine to form thiazolidine. Air samples containing trace levels of formaldehyde were prepared from paraformaldehyde. The percent yield of formaldehyde from paraformaldehyde was 85.1 +/- 1.14%. Air samples were bubbled into an aqueous cysteamine trap. Thiazolidine formed from formaldehyde and cysteamine in the trap was determined by gas chromatography with a fused silica capillary column and a nitrogen-phosphorus detector (NPD). The lowest detection level for thiazolidine was 17.2 pg, equivalent to 5.80 pg formaldehyde. The recovery efficiency of trace gas phase formaldehyde in air was greater than 90%. Formaldehyde levels in ambient laboratory air were 48.9-56.2 ppb (v/v).     

Determination of cadmium, copper, and lead in sodium chloride food salts by flame atomic absorption spectroscopy

A method for determination of Cd, Cu, and Pb in sodium chloride food salt samples has been developed. It consists of extraction in 4-methyl-2-pentanone of the complexes formed with ammonium pyrrolidine dithiocarbamate and further analysis of the extracts by flame atomic absorption spectroscopy. Detection limits in ng/g salt were 0.2 for Cd, 0.7 for Cu, and 10.0 for Pb. The coefficients of variation of 12 independent analyses were 13% for Cd (at a level of 0.4 ppb), 18% for Cu (1.6 ppb), and 5% for Pb (40 ppb). The recoveries were 100 +/- 0% for Cd, 115 +/- 14% for Cu, and 100 +/- 13% for Pb.     

Multiresidue screen for organophosphorus insecticides using gel permeation chromatography--silica gel cleanup.

A multiresidue screen for quantitative determination of 43 organophosphorus insecticides in 5 g of plant and animal tissues is described. The organophosphorus insecticides are extracted with methanol-dichloromethane (10 + 90, v/v) and cleaned up using automated gel permeation chromatography with hexane-ethyl acetate (60 + 40) eluant and in-line silica gel minicolumns. Concentrated extracts are analyzed by gas chromatography with flame photometric detection. The method recovers 43 organophosphorus insecticides in the range of 72 to 115%. Analysis of fortified bovine liver (n = 5) shows an average 95.9 +/- 4.8% recovery at the 0.05 micrograms/g level and 93 +/- 3.8% at the 0.5 micrograms/g level. Analysis of fortified bovine rumen content (n = 5) shows an average 98 +/- 4.2% recovery at the 0.1 micrograms/g level and 98.7 +/- 2.8% at the 1 micrograms/g level. Method detection limits ranged from 0.01 to 0.05 micrograms/g for the compounds studied using a nominal 5 gram sample.     

Capillary column gas chromatographic determination of ethyl carbamate in alcoholic beverages with confirmation by gas chromatography/mass spectrometry

A method is described for determining ethyl carbamate at low microgram/kg levels in several types of alcoholic beverages by capillary column gas chromatography with Hall electrolytic conductivity detection and confirmation by mass spectrometry. Samples are diluted to obtain a uniform concentration of ethanol (ca 10%) then saturated with NaCl and extracted with methylene chloride. Extracts are evaporated to a small volume and injected in ethyl acetate solution for chromatographic analysis. The method was evaluated by 5 laboratories, 4 employing the Hall detector and one using mass spectrometric detection. Overall between-laboratory mean percent recoveries were: wine, 85.3 +/- 21.0% coefficient of variation (CV) (spiking level 20-45 micrograms/kg); sherry, 83.8 +/- 16.1% CV (spiking level, 81-142 micrograms/kg); whiskey, 79.5 +/- 13.9% CV (spiking level 127-190 micrograms/kg); and brandy, 85.0 +/- 12.5% CV (spiking level 297-446 micrograms/kg). Mass spectrometric results agreed well with the Hall results for all commodities. Detection limits were about 5 micrograms/kg for the Hall detector and about 0.5 microgram/kg for mass spectrometric detection.     

Two enzyme immunoassays to screen for 2,4-dichlorophenoxyacetic acid in water

Two solid-phase enzyme immunoassays were developed to measure 2,4-dichlorophenoxyacetic acid (2,4-D), using 2 sets of structurally distinct immunogens and enzyme ligands. The 2,4-D analog, 2-methyl-4-chlorophenoxyacetic acid (MCPA), gave a similar response with both methods, whereas other phenoxy herbicides cross-reacted differently. In method A, the aromatic moiety of 2,4-D was distal from the carrier protein and labeled enzyme, whereas in method B, the acetic acid portion of the herbicide was distal. The use of both methods to screen for this herbicide in ground water and municipal and river water reduced the number of false-positive responses. Water sources having a low background response could be monitored with either method alone. When a concentration step, with disposable C18 extraction columns, was used, the limit of sensitivity was 5 micrograms/L. Method A was the more sensitive of the 2 methods with a limit of detection of 10 micrograms/L without the concentration step.     

Multivariate analysis of FTIR and ion chromatographic data for the quality control of tequila

Principal component analysis (PCA) was applied to the chromatographic and spectroscopic data of authentic Mexican tequilas (n = 14) and commercially available samples purchased in Mexico and Germany (n = 24). The scores scatter plot of the first two principal components (PC) of the anions chloride, nitrate, sulfate, acetate, and oxalate accounting for 78% of the variability allowed a classification between tequilas bottled in Mexico and overseas; however, no discrimination between tequila categories was possible. Mexican products had a significantly (p = 0.0014) lower inorganic anion concentration (range = 1.5-5.1 mg/L; mean = 2.5 mg/L) than the products bottled in the importing countries (range = 3.3-62.6 mg/L; mean = 26.3 mg/L). FTIR allowed a rapid screening of density and ethanol as well as the volatile compounds methanol, ethyl acetate, propanol-1, isobutanol, and 2-/3-methyl-1-butanol using partial least-squares regression (precisions = 5.3-29.3%). Using PCA of the volatile compounds, a differentiation between tequila derived from "100% agave" (Agave tequilana Weber var. azul, Agavaceae) and tequila produced with other fermentable sugars ("mixed"tequila) was possible. The first two PCs describe 89% of the total variability of the data. Methanol and isobutanol influenced the variability in PC1, which led to discrimination. The concentrations of methanol and isobutanol were significantly higher (methanol, p = 0.004; isobutanol, p = 0.005) in the 100% agave (methanol, 297.9 +/- 49.5; isobutanol, 251.3 +/- 34.9) than in the mixed tequilas (methanol, 197.8 +/- 118.5; isobutanol, 151.4 +/- 52.8).     

Flame absorption spectroscopic determination of cadmium, copper, iron, lead, and zinc in mussels

A rapid method is proposed for determination of Cd, Cu, Fe, Pb, and Zn in mussel samples. The elements are extracted with concentrated nitric acid in borosilicate glass tubes at 90 degrees C for 1 h, and determined by flame atomic absorption spectroscopy. Detection limits for a 300 mg sample corresponded to 0.3 microgram Cd/g, 0.7 microgram Cu/g, 33 microgram Fe/g, 0.7 microgram Pb/g, and 6 micrograms Zn/g. The coefficient of variation for 20 independent analyses was 7% for Cd, 7% for Cu, 6% for Fe, 14% for Pb, and 8% for Zn. Recoveries were 107 +/- 3% for Cd, 90 +/- 3% for Cu, 94 +/- 1% for Fe, 90 +/- 5% for Pb, and 96 +/- 3% for Zn. The accuracy of the method was determined by analyzing NBS Oyster Tissues.     

Isolation and gas chromatographic determination of chlorsulfuron in milk

A method for the isolation and gas chromatographic determination of chlorsulfuron in milk is presented. Blank or chlorsulfuron-spiked milk samples were blended into C-18 (octadecylsilyl derivatized silica, ODS) packing material. A column made from the C-18/milk matrix was washed with hexane after which chlorsulfuron was eluted with dichloromethane (DCM). The DCM eluate contained chlorsulfuron which was free from interfering co-extractants when analyzed by gas chromatography utilizing a nitrogen/phosphorus detector. Chlorsulfuron was found to undergo a thermally induced decomposition to give 2-amino-4-methoxy-6-methyl-1,3,5-triazine, which was detected and quantitated by this method. Standard curves for these analyses were linear (r = 0.992 +/- 0.004, n = 5), with an average percentage recovery of 91.6 +/- 10.8%, over the concentration range examined (62.5-2000 ng/mL). The inter- and intra-assay variabilities were 11.6 +/- 7.5% and 6.2%, respectively.     

Simple, rapid, and inexpensive cleanup method for quantitation of aflatoxins in important agricultural products by HPLC

A chemical cleanup procedure for low-level quantitative determination of aflatoxins in major economically important agricultural commodities using HPLC has been developed. Aflatoxins were extracted from a ground sample with MeOH/H2O (80:20, v/v), and after a cleanup step on a minicolumn packed with Florisil, aflatoxins were quantified by HPLC equipped with a C18 column, a photochemical reactor, and a fluorescence detector. Water/MeOH (63:37, v/v) served as the mobile phase. Recoveries of aflatoxins B1, B2, G1, and G2 from peanuts spiked at 5, 1.7, 5, and 1.7 ng/g were 89.5+/-2.2, 94.7+/-2.5, 90.4+/-1.0, and 98.2+/-1.1, respectively (mean+/-SD, %, n=3). Similar recoveries, precision, and accuracy were achieved for corn, brown and white rice, cottonseed, almonds, Brazil nuts, pistachios, walnuts, and hazelnuts. The quantitation limits for aflatoxins in peanuts were 50 pg/g for aflatoxin B1 and 17 pg/g for aflatoxin B2. The minimal cost of the minicolumn allows for substantial savings compared with available commercial aflatoxin cleanup devices.     

Capillary column gas chromatographic determination of dicamba in water, including mass spectrometric confirmation

A sensitive method is described for determining dicamba at low micrograms/L levels in ground waters by capillary column gas chromatography with electron-capture detection (GC-EC); compound identity is confirmed by gas chromatography-mass spectrometry (GC-MS) using selected ion monitoring. Dicamba residue is hydrolyzed in KOH to form the potassium salt. The sample is then extracted with ethyl ether which is discarded. The aqueous phase is acidified to pH less than 1 and extracted twice with ethyl ether. The combined ethyl ether extracts are concentrated, and the residue is methylated using diazomethane to form the corresponding dicamba ester. The derivatized sample is cleaned up on a deactivated silica gel column. The methylated dicamba is separated on an SE-30 capillary column and quantitated by electron-capture or mass spectrometric detection. Average recoveries (X +/- SD) for ground water samples fortified with 0.40 microgram/L of dicamba are 86 +/- 5% by GC-EC and 97 +/- 7% by GC-MS detections. The EDL (estimated detection limit) for this method is 0.1 microgram dicamba/L water (ppb).     

Metabolism of metamitron in goat following a single oral administration of a nontoxic dose level: a continued study

Disposition kinetic behavior and metabolism studies of metamitron and its metabolite in terms of the parent compound were carried out in black Bengal goats after a single oral administration of a nontoxic oral dose at 30 mg kg(-1) of body weight. Metamitron was detected in the blood sample at 5 min (2.23 +/- 0.04 microg mL(-1)), maximum at 1 h (3.43 +/- 0.02 microg mL(-1)) and minimum at 12 h (0.41 +/- 0.01 microg mL(-1)), after a single oral administration. Metabolite [3-methyl-6-phenyl-1,2,4-triazin-5(4H)-one] in terms of the parent compound was detected in the blood sample at 5 min (0.47 +/- 0.006 microg mL(-1)), maximum at 6 h (5.12 +/- 0.02 microg mL(-1)) and minimum at 96 h (1.06 +/- 0.016 microg mL(-1)), after a single oral administration. The t(1/2 K) and Cl(B) values of metamitron were 3.63 +/- 0.05 h and 1.36 +/- 0.016 L kg(-1) h(-1), respectively, whereas the t(1/2K)(m) and Cl(B)(m) values of the metabolite were 38.15 +/- 0.37 h and 0.091 +/- 0.001 L kg(-1) h(-1), respectively, which suggested long persistence of the metabolite in blood and tissues of goat. Metamitron was excreted through feces and urine for up to 48 and 72 h, whereas the metabolite was excreted for up to 168 and 144 h, respectively. Metabolite alone contributed to 96 and 67% of combined recovery percentage of metamitron and metabolite against the administered dose in feces and urine of goat, respectively. All of the goat tissues except lung, adrenal gland, ovary, testis, and mammary gland retained the metabolite residue for up to 6 days after administration.     

Improved malonaldehyde assay using headspace solid-phase microextraction and its application to the measurement of the antioxidant activity of phytochemicals

A modified malonaldehyde (MA) assay for antioxidant activity, which involves derivatization and headspace solid-phase microextraction (HS-SPME) was developed and validated. The recovery of MA as 1-methylpyrazole (product of MA and N-methylhydrazine) from a headspace of an aqueous solution containing MA, buffer, surfactant, and cod liver oil using HS-SPME with a PDMS/DVB fiber was 91.3 +/- 3.38%. MA was analyzed by a gas chromatograph with a nitrogen-phosphorus detector, and its detection limit was 0.0103 nmol/mL. The antioxidant activities of natural compounds were determined as the percentage inhibition of MA formed from cod liver oil oxidized by Fenton's reagents in the above aqueous solution. Sesamol inhibited MA formation most (86.1%), followed by eugenol (84.4%), capsaicin (80.7%), ethylvanillin (45.3%), and vanillin (31.6%) at a level of 50 microg/mL. This method did not require any organic solvents and is a simple, fast, and a highly sensitive method for MA determination.     

Quantification of heterocyclic aromatic amines in fried meat by HPTLC/UV-FLD and HPLC/UV-FLD: a comparison of two methods

A recently developed HPTLC/UV-FLD method was compared to the routinely used HPLC/UV-FLD method for the quantification of heterocyclic aromatic amines (HAA) formed at trace levels during the heating process of meat. For formation of these process contaminants under normal cooking conditions, beef patties were fried in a double-contact grill at 230 degrees C for five different frying times and extracted by solid-phase extraction. The HAAs most frequently found, that is, 2-amino-1-methyl-6-phenylimidazo[4,5- b]pyridine (PhIP), 2-amino-3,8-dimethylimidazo[4,5- f]quinoxaline (MeIQx), 2-amino-3,4,8-trimethylimidazo[4,5- f]quinoxaline (4,8-DiMeIQx), 9 H-pyrido[3,4- b]indole (norharman), and 1-methyl-9 H-pyrido[3,4- b]indole (harman), were quantified by two chromatographic methods, which were orthogonal to each other (normal versus reversed phase system). Both methods showed a similar performance and good correlation of the results ( R (2) between 0.8875 and 0.9751). The comparison of running costs and run time in routine analysis proved HPTLC/UV-FLD to be more economical (factor of 3) and faster (factor of 4) due to its capability of parallel chromatography. The HAA findings calculated by standard addition increased with the heating time from <1 to 33 microg/kg related to 3-6 min of frying time. The precision (RSD) was between 7 and 49% (HPTLC) and between 5 and 38% (HPLC) at these very low HAA levels formed.     

Sensitive and rapid reversed-phase liquid chromatography-fluorescence method for determining bisphenol A diglycidyl ether in aqueous-based food simulants.

A method has been developed for determination of bisphenol A diglycidyl ether (BADGE) in 3 aqueous-based food simulants: water, 15% (v/v) ethanol, and 3% (w/v) acetic acid. BADGE is extracted with C18 cartridges and the extract is concentrated under a stream of nitrogen. BADGE is quantitated by reversed-phase liquid chromatography with fluorescence detection. Relative precision at 200 micrograms/L was 3.4%, the detection limit of the method was 0.1 micrograms/L, and recoveries of spiking concentrations from 1 to 8 micrograms/L were nearly 100%. Relative standard deviations for the method ranged from 3.5 to 5.9%, depending on the identity of the spiked aqueous-based food simulant.     

Pharmacokinetics of cycloalliin, an organosulfur compound found in garlic and onion, in rats

Cycloalliin, an organosulfur compound found in garlic and onion, has been reported to exert several biological activities and also to remain stable during storage and processing. In this study, we investigated the pharmacokinetics of cycloalliin in rats after intravenous or oral administration. Cycloalliin and its metabolite, (3R,5S)-5-methyl-1,4-thiazane-3-carboxylic acid, in plasma, urine, feces, and organs was determined by a validated liquid chromatography-mass spectrometry method. When administered intravenously at 50 mg/kg, cycloalliin was rapidly eliminated from blood and excreted into urine, and its total recovery in urine was 97.8% +/- 1.3% in 48 h. After oral administration, cycloalliin appeared rapidly in plasma, with a tmax of 0.47 +/- 0.03 h at 25 mg/kg and 0.67 +/- 0.14 h at 50 mg/kg. Orally administered cycloalliin was distributed in heart, lung, liver, spleen, and especially kidney. The Cmax and AUC0-inf values of cycloalliin at 50 mg/kg were approximately 5 times those at 25 mg/kg. When administered orally at 50 mg/kg, cycloalliin was excreted into urine (17.6% +/- 4.2%) but not feces. However, the total fecal excretion of (3R,5S)-5-methyl-1,4-thiazane-3-carboxylic acid was 67.3% +/- 5.9% (value corrected for cycloalliin equivalents). In addition, no (3R,5S)-5-methyl-1,4-thiazane-3-carboxylic acid was detected in plasma (<0.1 microg/mL), and negligible amounts (1.0% +/- 0.3%) were excreted into urine. In in vitro experiments, cycloalliin was reduced to (3R,5S)-5-methyl-1,4-thiazane-3-carboxylic acid during anaerobic incubation with cecal contents of rats. These data indicated that the low bioavailability (3.73% and 9.65% at 25 and 50 mg/kg, respectively) of cycloalliin was due mainly to reduction to (3R,5S)-5-methyl-1,4-thiazane-3-carboxylic acid by the intestinal flora and also poor absorption in the upper gastrointestinal tract. These findings are helpful for understanding the biological effects of cycloalliin.