New capillary electrophoresis method for the determination of furosine in dairy products

A new capillary electrophoresis (CE) method was established for the quantitative determination of furosine in dairy products. Sample preparation and suitable electrophoretic conditions allowed accurate and reproducible quantitation of furosine in dairy products. Sample preparation consisted of drying hydrolyzed samples, redissolving them in 0.2 M NaOH, and purifying them by solid-phase extraction. The electrophoretic separation was carried out in an uncoated capillary maintained at 30 degrees C using 0.1 M phosphate buffer containing the additive hexadecyl trimethylammonium bromide (HDTAB, 1.2 mM) (pH 7.0) under 10 kV voltage and reverse polarity. Coefficients of variation of less than 2.25% for migration time and 5.80% for peak areas indicated that the technique was reproducible. The calibration curve followed a linear relationship with a highly significant (p < 0.01) coefficient of multiple determination (R (2) = 0.997). The limit of quantitation was 0.5 ppm, a concentration that corresponds to 4.5 mg/100 g of protein in milk samples. Furosine concentration (mg/100 g of protein) ranges of different dairy products (raw, pasteurized, UHT, and evaporated milks and yogurt) agreed with ranges previously reported. Therefore, the CE method presented is a suitable technique for the routine assessment of furosine in dairy products.     

Chemical indicators of heat treatment in fortified and special milks

Carbohydrate and furosine contents in 12 commercial fortified and special milk samples (pasteurized goat's and ewe's milks; ultrahigh-temperature (UHT) goat's milk, UHT milks fortified with calcium, magnesium, fiber, or royal jelly and honey; and lactose-hydrolyzed milks) were analyzed. Except for lactose-hydrolyzed milks, furosine, lactose, lactulose, galactose, glucose, N-acetylgalactosamine, N-acetylglucosamine, and myo-inositol contents were similar to the previously reported values for UHT or pasteurized milk samples. In lactose-hydrolyzed milks, lactulose was not detectable and lactose was present in low amount; high levels of glucose, galactose, fructose, tagatose, and furosine were also detected in this type of milk. Results found in commercial milks were compared to those obtained in laboratory-prepared UHT milks with lactose hydrolyzed prior to heating. Hydrolysis of lactose before thermal treatments promoted elevated accumulation of reducing sugars (galactose and glucose) that could be partially converted to the corresponding isomers (tagatose and fructose) during heating. In addition, the reducing sugars could also react with the amino groups of proteins, giving rise to the corresponding Amadori compound. According to the obtained results, heating prior to hydrolysis of lactose is suggested to avoid a considerable loss of available lysine.     

Generation of furosine and color in infant/enteral formula-resembling systems

The extent of the Maillard reaction was studied by measuring furosine and color formation in infant and enteral formula-resembling model systems prepared by mixing calcium caseinate, laboratory-obtained or commercial whey protein with lactose or dextrinomaltose (ingredients similar to those used in infant and enteral formula manufacture) and heating the mixture at 100, 120, or 140 degrees C for 0-30 min. The furosine determination was performed by HPLC and the color determination by measuring colorimetric parameters L, a, and b in a reflection photometer. The first steps of the Maillard reaction could be followed by furosine determination when initial ingredients had low thermal damage. Hence, furosine may be an indicator of low thermal damage in ingredients with <100 mg/100 g of protein. At the concentrations used in these model systems, similar to those in infant and enteral formulas, furosine values (indirect measure of lysine losses) were higher in lactose than in dextrinomaltose systems, in which only glucose, maltose, maltotriose, and maltotetraose among all of the sugars present showed reactivity with casein. Finally, the advanced steps could be followed by color determination when the initial ingredients had high thermal damage or the model systems were heated at high temperature or for a long time. Among the parameters assayed, b was the most sensitive.     

The fluorescence of advanced Maillard products is a good indicator of lysine damage during the Maillard reaction

The objective of this study was to determine whether in heat-treated milk-resembling models or milk there is a lag phase, before lactulosyllysine (LL) is converted into advanced Maillard products (AMP), and if there is a step during the heat treatment where LL is actively degraded into AMP. For that purpose, a low temperature (60-85 degrees C) and a long heat treatment (15-90 h) were chosen. We observe that the heat treatment first induces a parallel increase in furosine and AMP fluorescence, confirming that AMP are produced very early during the heat treatment. At this step, both indicators are correlated with each other and precisely reflect the lysine damage. After a time, however, furosine reaches a steady-state concentration, whereas AMP fluorescence still increases, remaining correlated with the lysine blockage. Nevertheless, heat treatment applied to milk does not reach this step so that AMP fluorescence appears as a rapid alternative to furosine quantification.     

Furosine as indicator of maillard reaction in jams and fruit-based infant foods

To investigate the presence of furosine in commercial samples of jams and fruit-based infant foods a simple method by ion-pair reversed-phase liquid chromatography is described. The yield of furosine during the hydrolysis with hydrochloric acid was optimized. The reproducibility in the repeatability and recovery of the method, expressed in relative standard deviation percentages, proved to be in the ranges of 4.1-8.3% and 1.0-4.4%, respectively. The recovery percentages of furosine varied between 86.7 and 95.3%. The obtained results support the suitability of the method. Furosine was detected in all studied samples. Although a high variability in the content of furosine was noticed, in general terms, the lowest levels of furosine were observed in samples of fruit-based infant foods and the highest were observed in jams of more than 60% sugar. These results could be due to different heat treatment, storage conditions, and/or differences in the values of water activity (a(w)) and amounts of sugar. The results obtained in the present paper point out the usefulness of furosine as an indicator of Maillard reaction for jams and fruit-based infant foods.     

Ratio of maltose to maltulose and furosine as quality parameters for infant formula

Nonenzymic browning reactions in commercial infant formulas were evaluated through their furosine content as well as the isomeric disaccharides formed during processing. Lactulose was observed only in samples containing appreciable amounts of lactose, whereas maltulose was present in all samples due to the isomerization of maltose. Because formation of maltulose depends on the initial amount of maltose present, the ratio maltose/maltulose was used for comparative purposes. The ratio maltose/maltulose varied within a wide range, 27-167; therefore, low values in maltose/maltulose ratio may indicate severe processing conditions during manufacture, whereas high values may indicate mild processing conditions. Variable amounts of furosine content in samples with similar maltose/maltulose ratios may be attributed to different conditions used during storage. Levels of furosine higher than those reported for milk powder were detected in most studied samples. Determination of both furosine and maltose/maltulose ratio would yield information retrospectively about the heat treatment applied during processing and the storage conditions of commercial infant formula.     

Storage stability assessment of freeze-dried royal jelly by furosine determination

The effect of freeze-drying and the assessment of the storage stability of freeze-dried royal jelly (RJ) were investigated by the determination of furosine and blocked lysine. The level of furosine in the RJ samples collected from cells at different times (1, 2, and 3 days after grafting) showed that the Maillard reaction had already occurred in the hive as indicated by the increase in furosine: from 9.6 to 20.8 mg/100 g of protein. Freeze-dried RJ was more prone to the early stage of the Maillard reaction than fresh RJ, as confirmed by the significantly higher furosine values found after 12 months, both at 4 degrees C (253.4 versus 54.9 mg/100 g of protein) and at room temperature (884.3 versus 332.5 mg/100 g of protein). After 18 months at room temperature, the lyophilized samples reached a furosine level of 1440.4 mg/100 g of protein, which corresponded to the blocked lysine levels, amounting to 24% of total lysine.     

Quantitation of major choline fractions in milk and dietary supplements using a phospholipase d bioreactor coupled to a choline amperometric biosensor

Current analytical methods lack the capacity of simultaneous determination of the content of free choline and phosphatidyl-bound choline, mainly phosphatidyl choline, in raw milk. Quantitative determination of total, free, and phosphatidyl-bound choline in milk and a dietary supplement is described using a phospholipase D packed bioreactor coupled to a choline oxidase-based amperometric biosensor. The response for choline and phosphatidyl choline was linear up to 0.5 mM and 1 mM, respectively, and the detection limits were 0.02 and 0.03 mM, respectively. The conversion efficiency of phosphatidyl choline to choline was 50% at 0.2 mL min(-1). The within days coefficient of variation for choline and phosphatidyl choline determination in milk samples was 2.8% and 3.2%, respectively. With the addition of an acid hydrolysis step, the method can quantify the concentrations of total, free, phosphatidyl-bound, and non-phosphatidyl-bound choline esters, thus permitting determination of major choline fractions in a complex matrix.     

Studies on N-terminal glycation of peptides in hypoallergenic infant formulas: quantification of alpha-N-(2-furoylmethyl) amino acids

To obtain information about the extent of the early Maillard reaction between the N-termini of peptides and lactose, alpha-N-(2-furoylmethyl) amino acids (FMAAs) were quantified together with epsilon-N-(2-furoylmethyl)lysine (furosine) in acid hydrolyzates of hypoallergenic infant formulas, conventional infant formulas, and human milk samples using RP-HPLC with UV-detection. FMAAs are formed during acid hydrolysis of peptide-bound N-terminal Amadori products (APs), and furosine is formed from the Amadori products of peptide-bound lysine. Unambiguous identification was achieved by means of LC/MS and UV-spectroscopy using independently prepared reference material. The extent of acid-induced conversion of APs to FMAAs was studied by RP-HPLC with chemiluminescent nitrogen detection (CLND). Depending on the corresponding alpha-N-lactulosyl amino acid, between 6.0% and 18.1% of FMAAs were formed during hydrolysis for 23 h at 110 degrees C in 8 N HCl. From epsilon-N-lactulosyllysine, 50% furosine is formed under these conditions. Whereas furosine was detectable in all assayed samples, five different FMAAs, alpha-FM-Lys, alpha-FM-Ala, alpha-FM-Val, alpha-FM-Ile, and alpha-FM-Leu, were exclusively detected in acid hydrolyzates of hypoallergenic infant formulas in amounts ranging from 35 to 396 mumol/100 g protein. Taking the conversion factors into account, modification of N-terminal amino acids in peptides by reducing carbohydrates was between 0.3% and 8.4%. This has to be considered within the discussion concerning the nutritional quality of peptide-containing foods.     

Assessment of initial stages of Maillard reaction in dehydrated onion and garlic samples

The initial steps of the Maillard reaction in freshly laboratory-freeze-dried and commercial dehydrated onion and garlic samples have been assessed by quantitative determination of 2-furoylmethylamino acids, obtained after acid hydrolysis of the corresponding Amadori compound. In freshly prepared samples, hardly any presence of 2-furoylmethylamino acids was detected, whereas in commercial samples, onion contained much more important levels of 2-furoylmethylamino acids as compared to garlic species. 2-Furoylmethyl-gamma-aminobutyric acid (1), 2-furoylmethyl-lysine (furosine; 2), and 2-furoylmethylarginine (3) were identified in all commercial dehydrated onion samples, with compound 3 being the most abundant. All garlic samples presented slightly higher levels of 2 than 3 with no presence of 1. The observed differences between onion and garlic commercial samples may be due to their very different content of reducing sugars. Moreover, some variations found in 2-furoylmethyl derivatives within both onion and garlic species could be also attributed to different processing and storage conditions during the manufacture of these products. The findings of this study show the first evidence of important levels of Amadori compounds in dehydrated garlic and onion samples, as well as the usefulness of 2-furoylmethyl derivatives as quality indicators for the early detection of the Maillard reaction in onion and garlic products.     

Chemical and physical changes in milk protein concentrate (MPC80) powder during storage

The solubility and chemical changes due to the Maillard reaction were investigated in milk protein concentrate powder containing 80% protein (MPC80) during storage at temperatures and relative humidities in the ranges of 25-40 °C and 44-84%, respectively. The Maillard reaction was studied by measuring furosine (a product of lactosylated protein after digestion with acid) and free hydroxymethylfurfural (HMF) contents by HPLC and L*, a*, b* values with a color-meter. Furosine, free HMF, and browning in MPC80 increased during storage, whereas the solubility decreased. The correlation between the Maillard reaction and solubility loss was explored in modified MPC80 to which glucose was added to enhance the rate of the Maillard reaction. More furosine and brown pigments were observed in the glucose-containing MPC80 than in MPC80 with added lactose. The opposite trend occurred for solubility, suggesting that the Maillard reaction may be a cause of solubility loss in MPC powder.     

Effects of ultra-high-pressure homogenization treatment on the lipolysis and lipid oxidation of milk during refrigerated storage

Free fatty acid (FFA) release and quantification and lipid oxidation extent of ultra-high-pressure homogenized (UHPH) milk samples were evaluated to assess the effect of UHPH on the susceptibility of milk lipids to lipolysis and oxidation. Milk was UHPH-treated at 200 and 300 MPa with inlet temperatures of 30 and 40 degrees C. UHPH-treated samples were compared to high-pasteurized milk (PA; 90 degrees C, 15 s). Results showed that all FFA increased significantly during storage only in 200 MPa samples. Lipid oxidation was measured as an accumulation of lipid hydroperoxides as the primary oxidation product and malondialdehyde and hexanal as the secondary oxidation products. Samples treated at 300 MPa presented higher malondialdehyde and hexanal content compared to 200 MPa treated-samples and to PA milk.     

Furosine: a suitable marker for assessing the freshness of royal jelly

Fifteen commercial samples of royal jelly, consisting of 10 imported samples, and 5 samples of known origin obtained freshly harvested from beekeepers, were analyzed for protein, lysine, and furosine content. In addition, a commercial sample of royal jelly, at the beginning of its commercial shelf life, was stored for 10 months both at 4 degrees C and at room temperature in order to assess the development of the Maillard reaction (furosine) and relative nutritional damage (blocked lysine). The commercial royal jelly products contained different amounts of furosine, ranging from 37.1 to 113.3 mg/100 g protein, evidence of different storage times and conditions. The average furosine content of the royal jelly samples of known origin and harvesting was significantly lower than that of the imported samples (41.7 versus 73.6 mg/100 g protein, respectively). With regard to shelf life, furosine content increased significantly from 72.0 mg/100 g protein to 500.8 mg/100 g protein after 10 months of storage at room temperature, while it increased to a much lower level (100.5 mg/100 g protein) when the royal jelly was stored at 4 degrees C. However, nutritional damage, expressed as blocked lysine (calculated indirectly from the furosine content), was minor or negligible, 11.9 and 2.3% of total lysine, in samples stored at room temperature and at 4 degrees C, respectively. Lysine was determined by an innovative procedure based on high-performance anion-exchange chromatography with pulsed amperometric detection (HPAEC-PAD). The results showed that furosine is a suitable index for assessing the quality and freshness of royal jelly.     

Lactoperoxidase-induced protein oxidation in milk

The reaction between lactoperoxidase (LPO) and H(2)O(2) in the presence of bovine serum albumin (BSA), beta-lactoglobulin, or casein was investigated for the formation of protein radicals by freeze-quench electron spin resonance (ESR) and by the formation of the protein oxidation product, dityrosine. The presence of BSA resulted in a dramatic change after 1 min of reaction in the obtained ESR spectrum compared with the spectrum obtained for LPO and H(2)O(2) alone. Furthermore, experiments employing BSA or beta-lactoglobulin resulted in the formation of long-lived protein radicals detectable 10 min after initiation of the reaction. The presence of casein resulted in a minor change in the fine structure of the ESR spectrum after 1 min of reaction compared with LPO and H(2)O(2) alone, but no difference between the two reaction mixtures could be observed after 10 min of reaction. The formation of dityrosine could be detected in reaction mixtures containing LPO and H(2)O(2) after 1 and 10 min of incubation at 25 degrees C both in the absence and in the presence of BSA, beta-lactoglobulin, or casein. The presence of casein resulted in an increased dityrosine concentration compared with the reaction with LPO and H(2)O(2) alone. Endogenous LPO in unpasteurized milk was activated at 25 degrees C by adding 1 mM H(2)O(2). Radical species could be detected directly in the milk by freeze-quench ESR during the initial phase of the reaction, and dityrosine could be measured after 4 h of incubation. The role of LPO activity in the formation of ESR detectable radical species and dityrosine in milk was further verified in ultrahigh temperature (UHT) milk with no endogenous enzyme activity, as the formation of ESR detectable radical species and dityrosine took place in UHT milk only upon the addition of both H(2)O(2) and exogenous LPO.     

Determination of choline in milk, milk powder, and soy lecithin hydrolysates by flow injection analysis and amperometric detection with a choline oxidase based biosensor

A fast-response and interference-free amperometric biosensor based on choline oxidase immobilized onto an electropolymerized polypyrrole film for flow injection determination of choline in milk, milk powder, and soy lecithin hydrolysates is described. The sensor displayed an Imax value of 1.9 +/- 0.2 microA and an apparent Michaelis-Menten constant, k'M, equal to 1.75 +/- 0.07 mM. Detection limits of 0.12 microM could be obtained. Because even a slight deterioration of the anti-interference membrane can adversely affect measurement accuracy, a real time monitoring of the biosensor selectivity has been achieved by a dual Pt electrode flow-through cell where the enzyme modified electrode is coupled to an enzyme-free electrode in a parallel configuration. Finally, bracketing technique (alternate injections of sample and standards) allows a two-point calibration to be performed in real-time, correcting for any drift in sensor response.     

Maillard reaction and protein cross-linking in relation to the solubility of milk powders

Protein changes in relation to solubility, Maillard reaction (MR), and protein cross-linking in whole milk powder (WMP), skim milk powder (SMP), and whey protein concentrate (WPC) stored at different relative humidities (RHs) were investigated by chemical and electrophoretic methods. WMP and SMP reached minimum solubility rapidly, while WPC showed no change in solubility. The loss of solubility corresponded with development of high-molecular-weight protein complexes observed by two-dimensional electrophoresis. The maximal MR rate occurred at 66% RH for WMP and SMP (high lactose/protein ratios) and 84% RH for WPC (low lactose/protein ratios) based on the furosine and hydroxymethylfurfural contents. However, browning was greatest at 84% RH in all powders. The minimum solubility corresponded with the casein and fat contents. The retention of solubility and minimal protein cross-linking of WPC compared to casein-containing powders suggest that the casein content and cross-linking strongly influence the decrease in the solubility of milk powder.     

Nondestructive quantification of organic compounds in whole milk without pretreatment by two-dimensional NMR spectroscopy

In this study, various organic compounds in commercial whole milk were quantified simultaneously by 1H 1D and 1H - 13C HSQC 2D NMR spectra without any pretreatment. 2D NMR spectroscopy was applied to quantification of milk compounds for the first time. Milk fat content was easily determined to be 3.6 +/- 0.1%, and the lactose content was 47.8 +/- 1.0 mg/mL by 1H NMR spectra. From 1H-13C HSQC spectra, the concentrations of citrate, N-acetylcarbohydrates, and trimethylamine were determined to be 3.2 +/- 0.2, 2.9 +/- 0.1, and 4.0 +/- 0.6 mM, respectively. The latter two compounds were quantified in milk for the first time. Butyric acid, total monounsaturated fatty acids, and total polyunsaturated fatty acids of triacylglycerols were 6.2 +/- 0.5, 9.1 +/- 0.9, and 2.9 +/- 0.3 mM, respectively. The fatty acid compositions (mol %) of triacylglycerols were then calculated and were observed to be in good agreement with reference values. The results indicated that 1H 1D and 1H-13C HSQC 2D NMR spectroscopy is useful for the rapid and nondestructive determination of various compounds in milk.     

Hydroxymethylfurfural and furosine reaction kinetics in tomato products

The reaction kinetics of two heat damage indices, HMF and furosine, were examined in four tomato products with different dry matter contents (10.2, 25.5, 28.6, and 34.5%) over a temperature-time range of 80-120 degrees C and 0-255 min. The reactions followed pseudo-zero order kinetics. E(a) and z-value were, respectively, 139. 9 kJ/mol and 19.2 degrees C for HMF, and 93.9 kJ/mol and 28.4 degrees C for furosine. The analyses of both indices in several samples of commercial and industrial tomato products showed very low levels of HMF (from 1 to 42 ppm) and a lack of correlation between HMF and furosine mainly because of the different evolution of the two indices during storage. The HMF level of a tomato paste sample stored at 25 degrees C decreased from 609 to 17 ppm after 98 days, while furosine increased from 458 to 550 mg/100 g of protein.     

Properties and stability of oil-in-water emulsions stabilized by coconut skim milk proteins

Protein fractions were isolated from coconut: coconut skim milk protein isolate (CSPI) and coconut skim milk protein concentrate (CSPC). The ability of these proteins to form and stabilize oil-in-water emulsions was compared with that of whey protein isolate (WPI). The solubility of the proteins in CSPI, CSPC, and WPI was determined in aqueous solutions containing 0, 100, and 200 mM NaCl from pH 3 to 8. In the absence of salt, the minimum protein solubility occurred between pH 4 and 5 for CSPI and CSPC and around pH 5 for WPI. In the presence of salt (100 and 200 mM NaCl), all proteins had a higher solubility than in distilled water. Corn oil-in-water emulsions (10 wt %) with relatively small droplet diameters (d32 approximately 0.46, 1.0, and 0.5 mum for CSPI, CSPC, and WPI, respectively) could be produced using 0.2 wt % protein fraction. Emulsions were prepared with different pH values (3-8), salt concentrations (0-500 mM NaCl), and thermal treatments (30-90 degrees C for 30 min), and the mean particle diameter, particle size distribution, zeta-potential, and creaming stability were measured. Considerable droplet flocculation occurred in the emulsions near the isoelectric point of the proteins: CSPI, pH approximately 4.0; CSPC, pH approximately 4.5; WPI, pH approximately 4.8. Emulsions with monomodal particle size distributions, small mean droplet diameters, and good creaming stability could be produced at pH 7 for CSPI and WPI, whereas CSPC produced bimodal distributions. The CSPI and WPI emulsions remained relatively stable to droplet aggregation and creaming at NaCl concentrations of < or =50 and < or =100 mM, respectively. In the absence salt, the CSPI and WPI emulsions were also stable to thermal treatments at < or =80 and < or =90 degrees C for 30 min, respectively. These results suggest that CSPI may be suitable for use as an emulsifier in the food industry.     

Portable surface plasmon resonance immunosensor for the detection of fluoroquinolone antibiotic residues in milk

An inexpensive and portable surface plasmon resonance (SPR) sensor, SPReeta Evaluation Kit SPR3, has been used to develop a biosensor for the determination of fluoroquinolone antibiotics (FQs) and to demonstrate its performance analyzing FQ residues in milk samples. The SPReeta three-channel gold chips were activated with a mixed self-assembled monolayer (m-SAM) and functionalized with a FQ haptenized protein. Binding of the antibody produced a concentration-dependent increase of the SPR signal as a result of the change in the refraction index. Similarly, the presence of the FQ produced a dose-dependent decrease of the response, which allowed a good limit of detection (LOD) to be obtained (1.0 ± 0.4 μg L(-1) for enrofloxacin in buffer). The response was reproducible in all three channels, on different injections and days, and also between chips. Milk samples could be analyzed after a simple sample treatment involving fat removal by centrifugation and dilution with water. Under these conditions calibration curves were obtained showing that FQ residues can be analyzed in milk samples with an IC(50) value of 26.4 ± 7.2 μg L(-1) and a LOD of 2.0 ± 0.2 μg L(-1) (for enrofloxacin), far below the European Union regulations for this antibiotic family in this matrix. Finally, the paper also demonstrates that the biosensor is able to selectively detect the presence of FQs in milk samples, even in the presence of other antibiotics. Enrofloxacin, ciprofloxacin, and norfloxacin residues were detected in blind samples supplied by Nestle? Co.