Capillary electrophoresis-mass spectrometry - a fast and reliable tool for the monitoring of milk adulteration

The development of a rapid, simple and accurate analytical method aimed at the detection and quantification of bovine milk in either ovine or caprine milk samples by means of CE-MS analyses of whey proteins with high-ionic strength and presence of acidic running buffer is described. The high-ionic strength buffer was used in order to minimize the problems with the adsorption of the proteins onto the fused-silica capillary wall. The acidic running electrolyte, pH 1.9, was used to support the production of positive ions in electrospray. Highly linear dependences of the ratio of the sum of non-bovine beta-lactoglobulins (ovine or caprine) to the total beta-lactoglobulins in milk mixture (bovine plus ovine or bovine plus caprine) vs. the volume percentage of added bovine milk in ovine (or caprine) milk were obtained. This technique allowed the fast and reliable evaluation of milk adulteration. The amount of bovine milk added into the "non-bovine" ones can be well within the concentration range of 5-95%.     

Species-specific identification of ruminant components contaminating industrial crude porcine heparin using real-time fluorescent qualitative and quantitative PCR

Ever since the emergence of bovine spongiform encephalopathy, the source of pharmaceutical heparin has been restricted to porcine intestinal mucosa. In this project, two real-time fluorescent PCR methods were developed to assist with quality control analysis. The first is a qualitative method which relies on SYBR Green I chemistry to confirm the porcine origin of industrial crude porcine heparin (ICPH), identify any ruminant contaminants, and generally control purity. The second is based on TaqMan chemistry and is able to quantitatively identify porcine, bovine, caprine, and ovine components and contaminants in ICPH. By targeting mitochondrial DNA, both PCR systems showed a detection limit of 1 pg DNA and amplification efficiencies ranging between 96% and 102%. Moreover, quantitative PCR showed a detection limit of 0.02 ppm in samples comprising porcine, bovine, caprine, and ovine DNA. The results of qualitative PCR over 27 ICPH samples showed that all samples were porcine in origin and that 17 had ruminant contaminants. The results of quantitative PCR further showed that out of all 17 samples with ruminant contaminants, seven samples had bovine, ovine, and caprine contaminants, two samples had bovine and ovine contaminants, and eight samples had only ovine contaminants. In conclusion, the qualitative PCR system was found to be a relatively inexpensive, rapid, and flexible method of identifying the porcine origin of and ruminant contaminants in ICPH, while the quantitative PCR was found suitable to accurately analyze the components and contaminants in detail. Both methods are suitable for routine control assays for the evaluation of ICPH purity and origins of contaminants.     

Toward milk speciation through the monitoring of casein proteotypic peptides

The possibility of detecting extraneous milk in singles species cheese‐milk has been explored. A mass spectrometry (MS)‐based procedure has been developed to detect 'signature peptides', corresponding to the predefined subset of 'proteotypic peptides', as matchless analytical surrogates of the parent caseins. Tryptic digests of skimmed milk samples from four species were analyzed by matrix‐assisted laser desorption/ionization time‐of‐flight (MALDI‐TOF) MS. Amongst the candidate signature peptides that are able to differentiate milks from the four species, the α_s1‐casein (CN) f8‐22 peptide was selected as a convenient marker for bovine, ovine and water buffalo milk while the f4‐22 peptide was selected as a marker for the two caprine α_s1‐CN A and B variants, which differ by a Pro¹⁶ (B)‐>Leu¹⁶ (A) substitution. MALDI analysis of the digest allowed the detection of α_s1‐CN f8‐22 and caprine α_s1‐CN f4‐22. The accurate evaluation of caprine milk in a quaternary mixture required the development of a liquid chromatography/electrospray ionization (LC/ESI)‐MS procedure. Five synthetic signature peptide analogues, which differed from their natural counterparts by a single amino acid substitution, were used as internal standards to quantify the α_s1‐CN, which was chosen as a reference milk protein, from the different species. The limits of detection were 0.5% (1% for caprine) for either the MALDI or the LC/ESI‐MS method. The isotopic‐label‐free quantification of isoform‐ or variant‐specific signature peptides has disclosed a convenient approach for targeting proteins in complex mixtures. Copyright © 2010 John Wiley & Sons, Ltd.     

Isolation and identification of O- and N-linked glycoproteins in milk from different mammalian species and their roles in biological pathways which support infant growth

Milk serves as the sole nutrition for newborns, as well as a medium for the transfer of immunological components from the mother to the baby. This study reveals different glycoprotein profiles obtained from human, bovine, and caprine milk and their potential roles in supporting infant growth. Proteins from these three milk samples are separated and analyzed using two-dimensional gel electrophoresis (2-DE). Glycosylated proteins from all samples are enriched by affinity chromatography using lectins from the seeds of Artocarpus integer before analysis using LC/MS-QTOF. The glycoproteome profiling demonstrates that glycosylated proteins are higher in caprine milk compared to other samples. Analysis using LC/MS-QTOF identified 42 O-glycosylated and 56 N-glycosylated proteins, respectively. Among those identified, human milk has 17 glycoproteins, which are both O- and N-glycosylated, whereas caprine and bovine have 10 and 1, respectively. Only glycoproteins from human milk have shown positive matching to important human biological pathways, such as vesicle-mediated transport, immune system and hemostasis pathways. Human milk remains unique for human babies with the presence of antibodies in the form of immunoglobulins that are lacking in ruminant milk proteomes.     

High resolution two-dimensional electrophoresis and sodium dodecyl sulphate-polyacrylamide gel electrophoresis of cheese proteins after rapid solubilisation

The proteins of cheese are rapidly solubilised by heating to 95 degrees C in buffered 2% sodium dodecyl sulfate, 5% 2-mercaptoethanol. Electrophoretic analysis of the solubilised proteins by either one-dimensional sodium dodecyl sulphate-polyacrylamide gel electrophoresis or high resolution two-dimensional electrophoresis yields reproducible patterns characteristic of an individual cheese and its extent of ripening. The patterns reveal (i) the residual amounts of milk casein and whey proteins, and (ii) the appearance of casein degradation products, including pink-violet components as detected by Coomassie Blue staining.     

Peptidomic approach based on combined capillary isoelectric focusing and mass spectrometry for the characterization of the plasmin primary products from bovine and water buffalo beta-casein

The main peptides produced by hydrolysis of water buffalo beta-casein with plasmin were characterized by capillary electrophoresis and mass spectrometry and compared with their bovine homologous. A novel breakdown product arising from the hydrolysis of water buffalo beta-casein, originated by the presence of a plasmin-sensitive Lys bond at position 68 was identified, which was not present in bovine beta-casein. On the basis of this evidence, an improved procedure for the detection and the differentiation of the products of plasmin hydrolysis of bovine and water buffalo beta-casein by capillary isoelectric focusing was set-up. In the experimental conditions, the gamma-casein from the two species was efficiently separated. Comparison of the capillary electropherograms with those obtained by ultra-thin-layer isoelectric focusing, the reference method for routine analysis of plasmin digests of casein, suggests that capillary electrophoresis isoelectric focusing may constitute a successful alternative to the traditional slab gel electrophoresis analysis of plasmin digests of casein either for basic structural studies or for applications in the quality assessment of dairy products.     

Quantification of bovine casein fractions by direct chromatographic analysis of milk. Approaching the application to a real production context

The ability to quantify the casein content by an exact and cost-effective approach represents an issue of crucial importance in the dairy industry as the natural variations in milk protein concentration can markedly affect the yield of the cheesemaking processes, thus causing a direct and significant economic impact on the producers. In this work, the separation and quantification of α_s1-, α_s2-, κ- and β-casein was carried out by direct RP-HPLC analysis of milk. The identification of each casein was established by electrospray ionization mass spectrometry. The data show that this method is able to effectively separate the bovine casein fractions, it provides simplified analytical conditions (with special regard to mobile phase composition and gradient profile) and faster separation while ensuring adequate precision to achieve reliable quantifications in milk samples from dairy production.     

Fluoroquinolone antibiotic determination in bovine,ovine and caprine milk using solid-phase extraction and high-performance liquid chromatography-fluorescence detection with ionic liquids as mobile phase additives

This paper describes the use of 1-ethyl-3-methylimidazolium tetrafluoroborate (EMIm-BF₄) as mobile phase additive for the analysis by high-performance liquid chromatography with fluorescence detection of a group of seven basic fluoroquinolone antibiotics (i.e. fleroxacin, ciprofloxacin, lomefloxacin, danofloxacin, enrofloxacin, sarafloxacin and difloxacin) in different milk samples. EMIm-BF₄ was found superior to 1-butyl-3-methylimidazolium tetrafluoroborate for the separation of the analytes from chromatographic interferences of the sample matrix. The optimized method was applied to the analysis of ovine, caprine and bovine milk, in the last case in either skimmed, semi-skimmed and full-cream milk after suitable acidic deproteination followed by a solid-phase extraction procedure. Recovery values between 73% and 113% were obtained for the three types of bovine milk samples, as well as for ovine and caprine milk (RSDs below 16% in all cases), which clearly demonstrates the applicability of the method to the three types of milk irrespective of the fat content of the samples. Limits of detection were in the range of 0.5–8.1 μg/L (approximately 0.5–25.9 μg/kg), well below the maximum residue limits established for these compounds by the current European legislation. A screening study of 24 different milk samples was also developed. In none of the samples, residues of the selected antibiotics were found.     

Preparative separation of immunoglobulins from bovine colostrum by continuous divergent-flow electrophoresis

Immunoglobulins in bovine colostrum were separated and fractionated from other proteins using the method and instrumentation developed in our laboratory. The proposed separation was based on bidirectional isotachophoresis/moving boundary electrophoresis with electrofocusing of the analytes in a pH gradient from 3.9 to 10.1. The preparative instrumentation included the trapezoidal non-woven fabric that served as separation space with divergent continuous flow. The defatted and casein precipitate-free colostrum supernatant was loaded directly into the instrument without any additional colostrum pre-preparation. Immunoglobulin G was fractionated from other immune proteins such as bovine serum albumin, β-lactoglobulin, and α-lactalbumin, and was continuously collected in separated fractions over 3 h. The fractions were further processed, and isolated immunoglobulin G in the liquid fractions was confirmed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and by re-focusing in gel isoelectric focusing. Separated immunoglobulin G was detected in seven fractions by sodium dodecyl sulfate-polyacrylamide gel electrophoresis with a gradually decreased concentration in the fractions. Re-focusing of the proteins in the fractions by gel isoelectric focusing revealed multiple separated zones of immunoglobulin G with the isoelectric point values covering the range from 5.4 to 7.2. Each fraction contained distinct zones with gradually increased isoelectric point values and decreased concentrations from fraction to fraction.     

Validation of capillary electrophoresis in the analysis of ewe's milk casein

Recent investigations have shown that capillary electrophoresis (CE) can be an alternative to other techniques such as polyacrylamide gel electrophoresis (PAGE) or sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) in the qualitative analysis and separation of the different casein fractions in cow's and ewe's milk. However, past work has not yet clarified whether that method can achieve good quantifications. The present study has used a commercial whole ovine casein standard and a mixture of the standard and whole casein extracted from ewe's milk cheese to test the reliability of the technique. The results show that CE was able to quantify the ewe's milk caseins. The areas under four of the most representative peaks on the electrophoretogram for two alpha and two beta-caseins (designated alpha-casein1CE, alpha-casein2CE, beta-casein1CE, and beta-casein2CE in order of elution) were used to validate the method. In relation to linearity, coefficient of determination (r2) values greater than 99% were obtained for the regressions of each of the caseins. Moreover, each casein yielded response factors with a relative standard deviation (R.S.D.) of less than or equal to 5. The coefficients obtained in the day-to-day reproducibility analysis were higher than those for the same-day repeatability, but all the values were within acceptable limits. In the study of accuracy, the percentage recovery rates for the alpha-casein fractions were higher than those for the beta-casein fractions, hence quantification of the latter using this technique would appear to be more accurate under the conditions employed.     

Detection of Ovine or Bovine Milk Components in Commercial Camel Milk Powder Using a PCR-Based Method

Food ingredient adulteration, especially the adulteration of milk and dairy products, is one of the important issues of food safety. The large price difference between camel milk powder, ovine, and bovine milk powder may be an incentive for the incorporation of ovine and bovine derived foods in camel milk products. This study evaluated the use of ordinary PCR and real-time PCR for the detection of camel milk powder adulteration based on the presence of ovine and bovine milk components. DNA was extracted from camel, ovine, and bovine milk powder using a deep-processed product column DNA extraction kit. The quality of the extracted DNA was detected by amplifying the target sequence from the mitochondrial Cytb gene, and the extracted DNA was used for the identification of milk powder based on PCR analysis. In addition, PCR-based methods (both ordinary PCR and real-time PCR) were used to detect laboratory adulteration models of milk powder using primers targeting mitochondrial genes. The results show that the ordinary PCR method had better sensitivity and could qualitatively detect ovine and bovine milk components in the range of 1% to 100% in camel milk powder. The commercial camel milk powder was used to verify the practicability of this method. The real-time PCR normalization system has a good exponential correlation (R² = 0.9822 and 0.9923) between ovine or bovine content and Ct ratio (specific/internal reference gene) and allows for the quantitative determination of ovine or bovine milk contents in adulterated camel milk powder samples. Accuracy was effectively validated using simulated adulterated samples, with recoveries ranging from 80% to 110% with a coefficient of variation of less than 7%, exhibiting sufficient parameters of trueness. The ordinary PCR qualitative detection and real-time PCR quantitative detection method established in this study proved to be a specific, sensitive, and effective technology, which is expected to be used for market detection.     

Critical reactions in ripening of cheeses

A mathematical model has been developed for the key reactions taking place during cheese ripening. It includes growth and lysis of cells in the cheese matrix, cell-wall bound proteinases and intracellular peptidases that are released into cheese upon cell lysis, and the production of peptides and amino acids from casein in cheese. The model parameters have been estimated using published experimental data for cheddar cheese, and model simulations have been conducted to suggest effective means of reducing ripening times of cheeses. The time required for ripening of cheeses can be significantly reduced by carefully controlling the cell numbers at the beginning of cheese ripening and their proteinase and peptidase activities.     

Determination of vitamin B12 in milk products and selected foods by optical biosensor protein-binding assay: method comparison

Biomolecular interaction analysis was evaluated for the automated determination of vitamin B12 in a range of foods. The analytical technique was configured as a biosensor-based, nonlabeled inhibition protein-binding assay using nonintrinsic R-protein. Sample extraction conditions were optimized, and both ligand specificity and nonspecific binding considerations were evaluated. Performance parameters included a quantitation range of 0.08-2.40 ng/mL, recoveries of 89-106%, agreement against assigned reference values for 3 independent certified food reference materials, and a mean between-laboratory reproducibility relative standard deviation of 4.9%. The proposed method was compared with reference microbiological and radioisotope protein-binding methods for a range of food samples. A wide selection of milks, infant formulas, meats, and liver were evaluated for their vitamin B12 content. The influence of season was studied in herd milk, early lactation was followed for a single animal, and the cobalamin content of bovine, caprine, and ovine milks was compared.     

Identification of novel angiotensin-converting enzyme-inhibitory peptides from ovine milk proteins by CE-MS and chromatographic techniques

In this report, we present the use of CE-MS as complement to RP separation for the identification of novel angiotensin-converting enzyme-inhibitory (ACEI) peptides from a complex milk protein hydrolysate. As preliminary step, fast protein LC (FPLC) was used to isolate the different casein fractions from raw ovine milk. Enzymatic hydrolysis of these fractions was performed by using proteolytic enzymes of gastrointestinal origin. The most active hydrolysate, corresponding to kappa-casein hydrolyzed with pepsin, chymotrypsin, and trypsin, was fractionated by RP-HPLC and the peptides contained in the active fractions were sequenced by CE coupled to IT-MS (CE-MS). The use of CE-MS allowed the identification of short peptides with ACEI activity included in the scarcely retained fraction obtained by semipreparative RP-HPLC. Among the identified peptides, those with hydrophobic or positively charged residues at the C-terminal tripeptide were chemically synthesized to determine their ACEI activity. This procedure allowed us to identify four novel potent ACEI peptides from kappa-casein with sequences IAK, YQQRPVA, WQVLPNAVPAK, and HPHPHLSF. These active sequences could be obtained by enzymatic hydrolysis either of the individual kappa-casein fraction or the total casein fraction from ovine milk.     

Phosphoprotein electrophoresis in the presence of Fe(III) ions

Preparation of affinity polyacrylamide gels containing immobilized Fe(III) ions for the separation of proteins exhibiting metal ion binding properties is described. The presented method enables uniform distribution of immobilized metal ions in the affinity part of the polyacrylamide separating gel. Affinity gels prepared by this way are suitable to follow the effect of different concentrations of metal ions immobilized in polyacrylamide gel on a protein electrophoretic behavior. Polyacrylamide gels containing immobilized Fe(III) ions were used to study the electrophoretic behavior of two model proteins differing in their phosphate group content: chicken ovalbumin and bovine α‐casein. For the electrophoretic separation, both the native and the denaturating conditions were used.     

Rapid analysis of bovine milk proteins by fast protein liquid chromatography

The separation of bovine milk proteins by fast protein liquid chromatography has been investigated by ion-exchange chromatography on Mono Q and Mono S columns and by gel filtration on a column of Superose 12. The four major casein components (alpha s1, alpha s2, beta and kappa) as well as the minor gamma-caseins were generally well separated on the Mono S column with urea-containing buffers at pH 3.8 in as short a time as 7 min, although there was considerable overlap between alpha s1- and alpha s2-casein peaks. Peak area measurements indicated that the four caseins alpha s1, alpha s2, beta and kappa were present in total casein in the approximate proportions of 3.0:0.5:3.4:0.9, in good agreement with other literature values. Whey proteins were not separated on the Mono S column, but were all well resolved by rapid analysis on the Mono Q column at pH values between 6 and 8 in buffers free of urea or 2-mercaptoethanol. Both urea and 2-mercaptoethanol were required for casein analyses on the Mono Q column, but all the casein components were then separable over a broad pH range (5.0-11.0). While urea levels of 4.5-8.0 M and pH values of 7.0 to 8.0 were most generally useful, the resolution of some components was affected by urea concentration or pH, so conditions may have to be modified for specific analysis problems. The caseins were too similar in size to be separated on the Superose 12 column but high-speed gel filtration in as little as 15 min separated all the whey proteins well, molecular weight values obtained being in good agreement with literature values.     

Bank of reference samples of blank urine from livestock

From three species of livestock, bovine, ovine and porcine, samples of urine from each 10–20 different animals were collected. The animals originated from a governmental experimental farm and differed in race, age, sex and were fed with different types of feed. Fifty different samples of urine were lyophilised in units of 5 mL in sealed amber glass bottles. After lyophilisation of the samples, several quality control tests were performed. The variation in net weight of the units of urine and the lyophilisates ranged from 0.01%–1.7% and 0.1%–10%, respectively. Most values did not exceed 0.1% for urines and 1% for lyophilisates. The residual water content of the lyophilisates was determined by the Karl-Fisher titration method. The average percentage of residual water ranged from 1.0%–7.0% for bovine, from 1.3%–3.0% for porcine and from 1.6%–5.5% for ovine urine lyophilisates. Further, all different samples were analysed for the presence of anabolic compounds with a multi-residue procedure developed at the CRL. In two samples of bovine urine the presence of α-boldenone was detected and confirmed. In most porcine urines the endogenous steroids α-testosterone and α-estradiol were detected. In a number of porcine and ovine urine samples α- and β-zearalenol were detected. The origin of these compounds is the f2-toxin produced by a Fusarium fungus. In ovine samples α-testosterone and α-estradiol were detected.     

Aggregation and gelation of micellar casein particles

Micellar casein particles (submicelles) are formed by removing calcium phosphate from native casein. The submicelles aggregate and eventually form a gel with a rate that increases strongly with increasing temperature and casein concentration. At low casein concentrations the gel is very weak and collapses under its own weight so that a precipitate is formed. The structure of the aggregates is studied using light scattering and cryo-electron microscopy. It is found that the aggregates have a self-similar structure with fractal dimension 2. The viscoelastic properties of the gel are studied by frequency scans of the loss and storage moduli during the gelation process. The bonds between the submicelles probably involve calcium phosphate complexes.     

Subunits of bovine lutropin. Hormonal and immunological activity after separation with reversed-phase high-performance liquid chromatography

By means of reversed-phase high-performance liquid chromatography, we have fractionated bovine lutropin (LH) standard preparations. The highly purified NIAMDD-bLH-4 was fractionated into two components, while the less pure NIH-LH-B9 revealed three distinct peaks. The eluted material was further characterized by in vitro bioassay and by homologous radioimmunoassay for bovine LH, ovine LH-alpha and ovine LH-beta subunits. The material with the shortest retention time possessed almost no LH-activity and showed a displacement curve nearly identical with that of the ovine LH-alpha subunit. The material corresponding to the second peak exhibited 6% of the original LH-activity, and its immunoreactivity was equal to that of the ovine LH-beta subunit. Furthermore, the fractions supposed to contain the alpha and the beta subunits were rechromatographed and their aminoacid contents analyzed. The results show close similarities between the rechromatographed fractions and the pure subunits.     

Isolation and structural studies of porcine, ovine and murine thymosin beta 4 by high-performance liquid chromatography

Rapid high-performance liquid chromatographic (HPLC) procedures have been used to isolate and characterize thymosin beta 4 from different species. Crude extracts termed thymosin fraction 5A were prepared from porcine and ovine thymus glands as well as murine spleen. Each fraction 5A preparation was then fractionated by HPLC on a muBondapak C18 reversed-phase column. Porcine and ovine thymus fraction 5A, and murine spleen 5A, each yields a predominant peak at a retention time similar to that of bovine thymosin beta 4. Amino acid analysis as well as HPLC tryptic peptide mapping of these peaks indicate that they have homologous sequences to bovine thymosin beta 4. Chromatographic analysis of fresh murine thymus and spleen tissues also revealed protein peaks at the position of bovine beta 4, suggesting that thymosin beta 4 is the native protein present in the animal tissues.