Phosphopeptide probes in raw milk and derived products

Bioactive peptides can be produced in vitro by fermentation of milk inoculated with starter or non-starter cultures or by digestion of milk proteins with one or more proteolytic enzymes. In vitro casein hydrolysis by gastrointestinal enzymes is also capable of generating bioactive peptides. Our study has focused on the fraction of bioactive peptides called casein phosphopeptides (CPPs). The negatively charged side chains of phosphoserines make the phosphorylated residues capable to chelate macroelements, such as Ca, Mg and Fe, as well as oligoelements, such as Zn, Ba, Cr, Ni, Co and Se, as suitable metal ion carriers from which derive their biological activity. CPPs are also resulted very good markers of raw material quality, process and authenticity of dairy products beyond their bioactivity. Native and partially dephosphorylated CPPs are markers of: 1) Raw milk from infected subjects with high somatic cell count (SCC); 2) Heat treatment intensity of commercial milks; 3) Parmigiano Reggiano (PR) cheese at different age of ripening: younger cheeses (up to 15 months) were distinguished from those older than 26 month; 4) Adulterated water buffalo milk (or ovine and caprine) with cheaper bovine milk used for producing cheese made from single-species milk. In milk with high SCC, the majority of the peptides identified by LC-ESI-Q-TOF-MS originated from ?s1-, ?s2-, ?- and ?-casein (CN), as a result of degradation by plasmin (PL) and enzymes of somatic cells. Moreover, seventeen native and twenty partially dephosphorylated CPPs were identified through selective enrichment on hydroxyapatite (HA) and matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS) analysis. The degree of phosphorylation was well correlated with the alkaline phosphatase (ALP) activity of milk. The CPP enrichment on HA of heat-treated milk samples has allowed detection of CPPs, ordinarily suppressed by the co-existence of non-phosphorylated peptides. Most CPPs had C-terminal lysine probably derived from the proteolytic activity of endogenous PL present in milk. ?s1-, ?s2- and ?-CN are preferentially hydrolyzed based on their affinity for PL, according to the decreasing order: ?- > ?s2- > ?s1-CN. Several phosphopeptides belonging to the ?-CN family and to the ?s2-CN were detected in raw and heat treated milk. A synthetic peptide analogue of ?-CN (f1-28) 4P was used as internal standard to measure the rate of proteolysis of ?-CN. ?-CN was found to have released ~4.3% of the ?-CN (f1-28) 4P, in raw milk. Moreover, CPPs underwent partial lactosylation by milk heating or drying. Lactosylated CPPs, ?-CN (f1-29) 4P, (f1-28) 4P, (f1-27) 4P and ?s2-CN (f1-21) 4P, (f1-24) 4P, were observed especially in liquid and powder milk. Raw and thermized/pasteurized milk did not contain lactosylated peptides/proteins whereas they were specific of intensely heated milks. The limit of detection for spiked raw or pasteurized milk with Ultra High Temperature milk was ~10%. Using MALDI-TOF, pH 4.6-soluble CPPs of different cheese varieties, including PR, selectively enriched on HA were identified. The rate of proteolysis and dephosphorylation phenomenon is resulted to change in cheeses from raw or pasteurized milk. Heat treatments modified the peptide profile by increasing the content of larger peptides whereas the shorter CPPs, containing the multi-phosphorylated motif -SerP-SerP-SerP-Glu-Glu- more resistant to proteolytic enzymes, accumulated in long-ripened cheeses made from raw milk. The ?-, ?s1- and ?s2-CN CPPs were isolated from PR cheese samples aged 5, 15, 26 and 35 months (mo). ?-CN was hydrolyzed within the first 15 mo of ripening faster than ?s1- and ?s2-CN did. In contrast, ?s1-CN was more hydrolyzed in 26- and 35-mo-old cheeses than ?-CN, and ?s1-CN (f67-79) 3P, ?s1-CN (f71-79) 1P, ?s1-CN (f64-74) 2P, ?s1-CN (f66-74) 3P and ?s1-CN (f67-74) 2P were detected just in these older samples. There was a significant hydrolysis of ?s1-CN at 26 mo of ripening; afterwards, the increase slowed down with time. On the other hand, the shorter CPPs, ?s1-CN (f65-74) 3P and ?s1-CN (f64-74) 3P, accumulated in the 35-mo-old PR cheese. Unlike cheese-milk, proteolysis proceeded in the PR cheese sequentially in the decreasing order: ?- > ?s1- > ?s2-CN. On the basis of CPP monitoring, the findings have allowed to categorize PR cheese according to the age, i.e. young, within 5- and 15-mo-old, and old, aged more than 26 mo. The MALDI analysis of proteotypic CPPs for bovine (B) and water buffalo (WB) milk has allowed to set up a proteomics methodology for the false-positive detection of B milk in genuine WB milk and derived cheese. Tryptic phosphopeptides ?-CN (f1-25) 4P and ?-CN (f1-28) 4P, selectively bound to HA, were used as markers of B and WB species, respectively. The actual percentage of adulterating B milk was calculated using a calibration plot with a detection limit of 0.5% via MALDI-TOF-MS analysis of species markers. The detection of B ?-CN (f1-25) 4P in CN digests of WB Mozzarella cheese samples have allowed to measure the percentage of B adulterant. The excellent specificity of proteotypic peptides opens the possibility to broaden the application to other species' milk or to any food protein of which is known the protein sequence.