Ivermectin (IVM) and moxidectin (MXD) are broad-spectrum endectocides belonging to the avermectin/milbemycin class of antiparasitic drugs not approved for use in dairy sheep. However, these compounds are widely used extra-label to control endo- and ecto-parasites in lactating dairy sheep. Effects of the route of administration on the pattern of IVM and MXD excretion in milk were comparatively characterized in lactating dairy sheep. The relationship between the milk and plasma disposition kinetics after subcutaneous (s.c.) and oral administration at 200 μg/kg body weight was also evaluated. IVM and MXD concentration profiles were measured in milk and plasma using a specific HPLC-based methodology. IVM and MXD were extensively distributed from the bloodstream to the mammary gland and large quantities, particularly for MXD, were excreted in milk. Residual concentrations of IVM were recovered in milk up to 11 d (oral treatment) or 25 d (s.c. treatment) post treatment. However, high MXD concentrations were detected in milk between 1 h and 35 d after its oral and subcutaneous administration. MXD concentrations as high as 3·77 ng/ml (oral) and 30·3 ng/ml (s.c.) were measured in milk at day 35 post administration. A higher MXD excretion in milk, compared with that of IVM, was obtained for both administration routes. An extensive plasma to milk distribution pattern was observed, being the area under the concentration-time curve of MXD obtained in milk up to 14-fold higher than that measured in the bloodstream. The total fraction of the administered dose excreted in milk for MXD was significantly higher than that for IVM, which agrees with the well known higher MXD lipophilicity. The long persistence of milk residual concentrations of MXD and IVM in lactating dairy sheep should be seriously considered before their extra-label use is recommended.

The study was carried out in a commercial flock on 286 Churra breed ewes (566 half-udders) assigned to three lots depending on the type of antibiotic dry therapy received in the lactation previous to the one studied. One-hundred-and-four ewes were given complete therapy in all udders, 103 received selective therapy of infected half-udders, and 79 received no therapy at all. Half-udders of all animals were sampled for bacteriological study at ≤72 h (lambing), 60 d, 120 d, and 155 d (drying-off) post partum. Dry therapy, parity number, lactation stage and therapy×parity interaction contributed significantly to variation in intramammary infection prevalence. Antibiotic dry therapy had the most significant effect. Prevalence during the whole of the subsequent lactation was significantly lower in lots receiving complete (18·8%) and selective (15·6%) dry therapy than in the untreated control lot (48·3%). Coagulase-negative staphylococci and streptococci (in particular Streptococcus agalactiae) were the organisms most significantly affected by dry therapy. In untreated ewes, prevalence increased noticeably from the 2nd to the 6th and subsequent lactations, but no significant changes were observed in the treated lots. Milk yield in the dry treated lots was 6·9% higher that in the untreated one. It was concluded that complete and selective treatments of ewes at drying-off were efficient and comparable methods of reducing the intramammary infection prevalence, improving bacteriological quality of milk, and increasing milk yield.

A total of 1304 goat udder halves were sampled monthly during an entire lactation (6262 samples) with the aim of identifying factors affecting milk somatic cell count (SCC). Bacteriological analyses for identification of mastitis pathogens were carried out on all samples and SCC was also determined. All animals were examined for infection by caprine arthritis-encephalitis virus (CAEV) using a commercial ELISA test kit. Results obtained were arranged in two databases (whole-lactation average half-udder database and monthly half-udder database) and two mixed models were applied. Random effects of half udder nested into flock and fixed effects of flock, intramammary infection (IMI) status, number of kids born, length of lactation and interaction of parity with IMI status were significant for the first database. CAEV infection and its interaction with IMI status was not significant. Milk SCC was significantly increased for infected udder halves and milk from udder halves infected with minor pathogens had lower SCC than udder halves infected with major pathogens. For healthy udder halves, SCC was higher in older animals but this effect was not evident in halves with IMI. Multiple birth and short-duration lactation were factors associated with elevated milk SCC. The second mixed model considered repeated measures in time for consecutive samplings throughout lactation (stage of lactation) which was also a significant factor with increasing stage of lactation. The influence of all these factors should be taken into account in the establishment of more reliable diagnostic SCC thresholds for IMI.

Earlier studies with temporarily isolated rumen of heifers show saturation kinetics of Mg efflux across the rumen wall. Therefore, we hypothesized that high Mg intakes would not further increase the rate of Mg absorption in cows. To test our hypothesis, six ruminally fistulated non-pregnant dry cows were given diets with different Mg concentrations in a 6×6 Latin square design. Desired concentrations of Mg were attained by adding MgO to the basal diet and the Mg concentrations in the total rations were 3·8, 6·4, 9·1, 11·8, 14·1 and 17·3 g Mg/kg dry matter, which provided Mg intakes of 27·1, 44·6, 64·6, 83·5, 100·4 and 124·3 g/d, respectively. Increasing Mg intakes were associated with increased (P<0·001) faecal Mg excretion. However, apparent Mg absorption expressed as g/d was not significantly different for Mg intakes from 100·4 to 124·3 g/d while Mg absorption expressed as a proportion of intake was not significantly different for Mg intakes ranging from 64·6 to 124·3 g/d. Mg concentrations in rumen fluid after feeding increased (P<0·001) with increasing Mg intakes. Apparent absorption of Mg appeared to become saturated at a ruminal Mg concentration of 17·5 mM (Mg intake of 83·5 g/d). Group-mean post-feeding concentrations of Mg and Na in rumen fluid were significantly correlated (Pearson's r=−0·96; P=0·003, n=6). This study showed that under conditions of practical dairy cow feeding, Mg absorption was maximal at Mg intakes [ges ]84 g/d.

Effects of ventilation regimen on the quality of ewes' milk and on proteolysis in Canestrato Pugliese cheese during ripening were studied. Cheeses were manufactured from the bulk milk of Comisana ewes subjected to three different ventilation regimens, which were designated low (LOV, 23 m3/h per ewe), moderate (MOV, 47 m3/h per ewe) and programmed ventilation regimen (PROV, 73 m3/h per ewe; fan set to maintain 70% relative humidity). Bulk milk was analysed for chemical and microbial composition, renneting parameters and plasmin-plasminogen activities. At 1, 15, 30 and 45 d of ripening, the cheeses were analysed for gross chemical composition, nitrogen fractions, and plasmin and plasminogen activities. The pH 4·6-insoluble nitrogen fractions were analysed by urea-PAGE. Free amino acid content was determined at the end of ripening. Lower concentrations of bulk milk somatic cell count (BMSCC) and of mesophilic bacteria were found in the MOV group than in the LOV and the PROV groups. A lower plasminogen (PG) to plasmin (PL) ratio (PG/PL) was observed in the MOV and PROV than in the LOV cheeses. Irrespective of treatment, PL activity in cheeses was higher at 15 d of ripening, while a sudden decrease of PL and PG activities was observed at 30 d, which was associated with a marked increase in non-protein nitrogen. The peptide profile characterized in the urea-PAGE showed a greater intensity of α- and β-CN hydrolysis in the MOV than in the PROV and LOV cheeses. The results provide evidence that a proper ventilation regimen is critical for optimizing the hygienic quality of milk and the proteolysis of Canestrato Pugliese cheese during ripening.

Effects of six different milking intervals on the distribution of milk between cistern and alveoli were studied in a randomized, incomplete Latin Square experiment with four lactating Holstein cows. Cisternal and alveolar milk was measured by udder quarter at 4, 8, 12, 16, 20 and 24-h intervals with a 3-d interperiod of regular milking. Cisternal milk was evacuated using a cannula after injection of an oxytocin-receptor blocking agent, followed by an injection of oxytocin to remove the alveolar fraction. Milk samples from each fraction and quarter were collected for analysis. Cisternal and alveolar milk increased with milking interval and represented on average 30 and 70% of the milk stored in the udder, respectively. Fat content in alveolar milk remained constant during the first 16 h, increasing rapidly thereafter, reaching its maximum at 24 h (6·95%). Fat content in cisternal milk decreased with milking interval and reached its minimum at 24 h (0·96%). Total fat yield tended to increase for cisternal milk with longer milking intervals, but it increased markedly for alveolar milk, showing that fat globules did not pass freely from alveoli to cistern between milkings. Milk protein content was greater in rear quarters than in front quarters for both milk fractions. Milk protein content increased in the cisternal milk fraction and tended to increase in the alveolar milk fraction with longer milking intervals, but values did not differ between cisternal and alveolar fractions or between front and rear quarters. Total protein yield increased with milking interval in both fractions, indicating that casein micelles passed more freely than fat globules from the alveolar to the cisternal compartment. In conclusion, the short-term effects of milking intervals in milk composition were explained by the changes observed in alveolar and cisternal milk ratio.

A study was undertaken in cattle to evaluate changes in milk L-lactate in relation to mastitis. A healthy, rear quarter of the udder of each of ten cows in mid-lactation was infused with 1000 colony-forming units (cfu) of Streptococcus uberis following an afternoon milking. Foremilk samples were taken at each milking from control and treated quarters and antibiotic treatment was applied following the onset of clinical mastitis or after 72 h. One cow did not become infected. Six quarters showed clinical symptoms of mastitis within 24–40 h and this was associated with a more than 30-fold increase in milk L-lactate (to 3·3 mM) and an increase in somatic cell count (SCC) from 4·5×103 to 1×107 cells/ml. Three cows were subclinical, with cell counts ranging from 1·5×106 to 1×107 cells/ml. In these animals, milk lactate ranged from 0·7 to 1·5 mM in the infected quarters up to 40 h post-infection, compared with less than 0·1 mM in control quarters. Milk was examined from 137 cows in mid-lactation which were known to have mastitis. Foremilk samples were taken aseptically from control and infected quarters of cows on commercial farms. Mean milk L-lactate concentrations and SCC were 0·14±0·02 mM and 1·85±0·3×105 cells/ml, respectively, in control (bacteriologically negative) samples. However, L-lactate concentrations exceeded 2·5 mM in the presence of some types of infection, the level of the lactate response being closely related to the impact of the infection on SCC. L-Lactate concentrations were relatively elevated in milk samples taken post partum, declining from 0·8 to 0·14 mM over the first few days of lactation. In conclusion, milk L-lactate has potential as an indicator of clinical and subclinical mastitis in dairy cows.

The fate of ivermectin (IVM) residues was studied throughout the processing of daily bulk milk from 30 ewes (taken up to 33 d following subcutaneous administration of 0·2 mg IVM/kg b.w.) in the following milk products: yoghurt made from raw and pasteurized milk; cheese after pressing; 30- and 60-day ripened cheese; and whey, secondary whey and whey proteins obtained after cheese–making (albumin cheese). The concentration of the H2B1a component of IVM was analysed in these dairy products using an HPLC method with fluorescence detection. The mean recovery of the method was, depending on the matrix, between 87 and 100%. Limits of detection in the order of only 0·1 μg H2B1a/kg of product were achieved. Maximum concentrations of IVM were detected mostly at 2 d after drug administration to the ewes. The highest concentration of IVM was found on day 2 in 60-day ripened cheese (96 μg H2B1a/kg cheese). Secondary whey was the matrix with the lowest concentration of IVM (<0·6 μg H2B1a/kg). Residue levels fell below the limits of detection between day 5 (for secondary whey) and day 25 (for all cheese samples). In the matrices investigated, linear correlations between daily concentrations of IVM, milk fat and solid content were evident. During yoghurt production, fermentation and thermal stability of IVM was observed. During cheese production, approximately 35% of the IVM, present in the raw (bulk) milk samples, was lost. From the results it was concluded that the processing of ewes' milk did not eliminate the drug residues under investigation. The consequences of IVM in the human diet were discussed. Milk from treated animals should be excluded from production of fat products like cheese for longer after treatment with IVM than for lower fat products.

A total of 118 enterococcal strains isolated from artisanal Fiore Sardo cheese were characterized technologically and genetically. The presence of potential virulence factors was also investigated. Strains were classified as Ec. faecium (84 strains), Ec. durans (24 strains) and Ec. faecalis (10 strains). RAPD-PCR analysis with two different primers (M13 and XD9) confirmed species identification and proved useful for the detection of interstrain variations, especially among Ec. faecium isolates. Most strains could hydrolyse casein and had weak acidifying activity in milk. None of the isolates produced lipolytic reactions. Gelatinase activity was observed in two strains of Ec. faecalis and one strain of Ec. durans. β-Haemolysis on horses' blood was never detected in any of the strains, independently of species. Most strains produced tyramine from tyrosine but none decarboxylated lysine, histidine or ornithine. Overall, a wide spectrum of resistance was observed. Almost all strains were resistant to the aminoglycosides, gentamycin, kanamycin, streptomycin, neomycin, and to the semisynthetic penicillin, oxacillin, but resistance to vancomycin was not widespread among our strains: only one Ec. faecium and one Ec. durans strain were found to be vancomycin resistant. Our results show a certain diversity in technological traits of the enterococcal strains isolated from artisanal Fiore Sardo, together with a low incidence of some potentially pathogenic traits of health concern.

Milk from cows fed a low-fat diet high in cereals designed to stimulate fat synthesis de novo was lower in unsaturated fatty acids (21·3%) than milk from cows fed a diet high in fat, mainly from roasted soy beans (41·3% unsaturated fatty acids). Buttermilk from the more unsaturated milk was less oxidatively stable during storage (at 4 °C, followed for 11 d) than buttermilk from the more saturated milk, as monitored both by primary lipid oxidation products (lipid hydroperoxides) and by the secondary lipid oxidation product, hexanal. Fat-soluble antioxidants, β-carotene and α-tocopherol, analysed by HPLC, were not consumed during storage for either of the two types of buttermilk. In contrast, the antioxidative capacity of the serum phase decreased during storage as evaluated in a radical scavenging assay based on the semi-stable water-soluble radical nitrosodisulphonate (Fremy's salt). The time course for the decrease in water-soluble antioxidants was very similar for the two types of buttermilk suggesting that oxidation is initiated in the serum phase independently of fatty acid composition.

As a result of multiple QTL-mapping projects in recent years, a quantitative trait locus for milk fat percentage and milk yield has been described on BTA14. Recent reports name the acyl-CoA[ratio ]diacylglycerol acyltransferase (DGAT1) gene on BTA14 as a potential candidate gene, with a nonconservative substitution of lysine by alanine (K232A) producing a major effect on milk composition and yield. DGAT1K appears to be the ancestral allele and the K232A substitution probably occurred after the divergence of the Bos indicus and Bos taurus lineages. These findings prompted us to genotype 1748 DNA samples of 38 different Bos taurus and Bos indicus cattle breeds from 13 countries on five continents (Europe, Africa, Asia, North America and South America), to examine the occurrence of the DGAT1 polymorphism and characterize the K232A substitution in cattle breeds of different origins and selected for different purposes (e.g., beef, dairy and dual purpose). Calculating pairwise FST values for pooled subpopulations showed least divergence for Bos indicus breeds with high milk fat percentage. Fixation of DGAT1A was found in some Bos taurus breeds and fixation of DGAT1K in one Bos indicus breed. Breeds of no known organized breeding background from the Near East domestication centre of Bos taurus and taurine African N'Dama cattle were found to possess intermediate frequencies of DGAT1K. While beef breeds tended to harbour higher DGAT1A levels, dairy cattle showed everything from very low levels of DGAT1K to unexpectedly high frequencies of this allele.

The study was aimed at identifying the pathogens causing subclinical udder infections in representative Israeli dairy goat herds and determining their effect on milk quality. Five hundred goats in ten flocks of various breeds and crossbreeds were surveyed. Of the 500 goats, 13·4% were in their first lactation, 36·4% were in their second lactation and 50·2% were in their third or higher lactation. Percentages of udder halves with subclinical intramammary infection in the flocks ranged from 35 to 71%. The effect of the bacteriological infection on somatic cells count (SCC) was significant (P<0·001). Various species of coagulase-negative staphylococci (CNS), mainly Staphylococcus caprae and Staphylococcus epidermidis, were the main pathogens in infected udder halves. Lactation number did not significantly influence either infection rate of udder halves or SCC, although the percentage of udder halves with no bacteriological findings was higher at the first lactation than at the third lactation. Milk composition (fat, protein and lactose) varied among flocks, with lower mean total protein in uninfected halves than in infected ones and higher lactose in uninfected than infected halves.

We assessed polymorphisms in exon IV of the κ-casein gene (CSN3) in ten different breeds of domestic goat (Capra hircus) from three continents and in three related wild caprine taxa (Capra <ibex> ibex, Capra <ibex> sibirica and Capra aegagrus). Thirty-five DNA samples were sequenced within a 558 bp fragment of exon IV. Nine polymorphic sites were identified in domestic goat, including four new polymorphisms. In addition to four previously described polymorphic positions, a total of 13 polymorphisms allowed the identification of 13 DNA variants, corresponding to 10 protein variants. Because of conflicting nomenclature of these variants, we propose a standardized allele designation. CSN3*A, CSN3*B, and CSN3*D were found as widely distributed alleles in European goat breeds. Within Capra ibex we identified three variants and showed that the sequence of Capra aegagrus is identical to the most common Capra hircus variant, consistent with Capra aegagrus being the wild progenitor of domestic goats. A dendrogram was drawn to represent the molecular network between the caprine CSN3 variants.

Twenty-nine multiparous cows of each of the Jersey and Friesian breeds, all κ-casein AB phenotype, were grazed together and managed identically. On three occasions during 10 d in spring (early lactation), milk was collected from all cows at four consecutive milkings and bulked according to breed. On a separate occasion, milk samples were also collected from each cow at consecutive a.m. and p.m. milkings to form one daily sample per cow. The bulked milks (800–1000 l per breed on each occasion) were standardized to a protein[ratio ]fat (P[ratio ]F) ratio of 0·80, and 350 l from each breed was made into Cheddar cheese. The solids content of the remaining Friesian milk was then increased by ultrafiltration to a solids concentration equal to that of the Jersey milk. This solids-standardized Friesian milk and a replicate batch of P[ratio ]F standardized Jersey milk were made into two further batches of Cheddar cheese in 350-l vats. Compared with Friesian milk, Jersey milk had higher concentrations of most milk components measured, including protein, casein and fat. There were few difference in milk protein composition between breeds, but there were differences in fat composition. Friesian milk fat had more conjugated linoleic acid (CLA) than Jersey milk fat. Jersey milk coagulated faster and formed firmer curd than Friesian milk. Concentrations of some milk components were correlated with coagulation parameters, but relationships did not allow prediction of cheesemaking potential. Jersey milk yielded 10% more cheese per kg than Friesian milk using P[ratio ]F standardized milk, but for milks with the same solids concentration there were no differences in cheese yield. No differences in cheese composition between breeds were detected. Differences in cheesemaking properties of milk from Jerseys and Friesians were entirely related to the concentrations of solids in the original milk.

The aim was to select potentially probiotic lactobacilli from 88 strains isolated from unpasteurized milk and cheese products, and to incorporate these bacteria in a viable state into a soft cheese, without changing its quality. The survival of these bacteria was assessed in acidic and bile conditions, after freezing at −80 °C. Four strains from unpasteurized Camembert – two Lactobacillus plantarum strains and two Lb. paracasei/casei strains – were identified and typed by PCR and PFGE and were found to display potentially probiotic characteristics in addition to resistance to low pH and bile. These characteristics were resistance to lysozyme, adhesion to CACO-2 cells, antimicrobial effects against common foodborne pathogens (Listeria monocytogenes, Staphylococcus aureus, Salmonella spp., Escherichia coli), innocuity following the ingestion of high doses by mice and appropriate antibiotic susceptibility profiles. The potential of Lb. plantarum strain UCMA 3037 for incorporation into a soft cheese (Pont-l'Eveque registered designation of origin (RDO)) was investigated. This strain grew well and survived in sufficient numbers (more than 107 cfu/g throughout the shelf-life of the product) in the cheese. This strain did not change the quality score of the product until the best before date (75 days after manufacture). Thus, unpasteurized Camembert is a natural source of potentially probiotic lactobacilli, which could be used as an additive in the development of potentially probiotic soft cheeses. Further work is required to demonstrate the persistence and efficacy of these strains in the human host upon ingestion.

We examined the relationship between physicochemical indicators and somatic cells in the milk of dairy cows during experimentally induced mastitis and their significance as indicators for use in controlling udder health. We were concerned particularly with the effect of alveolar milk ejection on the sensitivity of these indicators. In Expt 1, Escherichia coli lipopolysaccharide (Esch. coli LPS) was injected into the left rear quarter to induce an inflammatory reaction in one quarter in each of six cows. The contralateral control quarter was injected with a solution of NaCl (9 g/l). Nine milk samples were taken from both quarters until 60 h after injection. In Expt 2, repeated milk samples were taken every 20 s from one quarter during a 120-s teat stimulation in 20 cows with different somatic cell counts (SCC). Quarters were clustered for low (<5·0 log cells/ml), mid (5·0–5·7 log cells/ml) and high (>5·7 log cells/ml) SCC of the sample taken at t=0 s. Samples were analysed for SCC, electrical conductivity (EC) and Na+ and Cl− concentrations. During the experimental inflammation SCC, EC, Na+ and Cl− peaked at 12 h from LPS administration and values in treated quarters (T) at this time were elevated to 7900, 157, 501 and 169% of the values in untreated quarters, respectively. In Expt 2, SCC, EC, Na+ and Cl− in high SCC quarters were 2520, 121, 283 and 141% of low SCC quarters at the start of stimulation (t=0 s), respectively. Highly significant (P<0·001) differences in EC, Na+ and Cl− between high and low SCC quarters disappeared owing to the onset of alveolar milk ejection 100 s after the first contact with the teat. In conclusion, SCC in cows' milk provided the strongest amplitude in the case of an intramammary inflammation. EC, Na+ or Cl− were useful tools only if the measurements were performed in cisternal milk before the start of alveolar milk ejection.

Two monoclonal antibodies (MAb) raised against bovine κ-casein were developed and applied in an automated optical biosensor (Biacore 3000) to create easy and fast direct and inhibition biosensor immunoassays (BIA) for the detection of cows' milk in the milk of ewes and goats. With both assay formats, low limits of detection (<0·1%) and fast run times (around 5 min) were obtained. For sample preparation, milk was diluted in buffer (direct assay) or in an antibody-containing buffer (inhibition assay) only. For quantitative analysis, calibrants of cows' milk in ewes' or goats' milk were used. Advantages of the direct BIA are: the single reagent format (biosensor chip immobilized antibodies only); the use of small amounts of antibodies (2 μg for >350 tests); and the wide measurement range (0·1 to 10% cows' milk). Despite these advantages, the inhibition BIA (using κ-casein immobilized on the chip) was preferred because of the possible application of non-purified Mab, the higher responses, the higher sensitivity at relevant low percentages of cows' milk and its robustness (>800 cycles per chip).

Aminotransferase (AT) activity against 18 amino acids was studied in ten strains of three species of Lactobacillus. A method for permeabilisation of cells was developed using toluene and ethanol combined with mechanical treatment. It was found that the AT activities in the washed permeabilised cells (W-PC) corresponded well to that in cell-free extracts (CFE). The AT specificity pattern was species as well as strain dependant. Strains of Lb. helveticus had high specificity for aromatic amino acids (ArAA) and lower activity against branched-chain amino acids (BcAA) and Asp, while strains of Lb. paracasei subsp. paracasei degraded BcAA and Asp, but had a lower and variable specificity against ArAA. One of the Lb. paracasei strains was characterised by having very high AT activity against all three BcAA (Ile, Leu, Val) compared with any of the other Lb. paracasei strains tested. Strains of Lb. danicus, which is a newly discovered Lactobacillus species isolated from cheese, had up to about 20 times higher AT activity against Leu than Lb. paracasei and Lb. helveticus. The permeabilised cells of Lb. danicus had also considerably higher AT activity against ArAA than Lb. paracasei and Lb. helveticus strains, and also higher AT activity against Asp. All Lactobacillus strains tested had AT activity against Met, but at a much lower rate than against other amino acids. Results of this study also demonstrated a chemical reaction between α-ketoglutaric acid and Asp that was catalysed by pyridoxal-5-phosphate without any AT present.

Effects of including silage in the diet on volatile compound profiles and their modification during ripening of Montasio cheeses were examined. Twelve farms were selected and grouped according the type of forage in the ration: hay-based diets (four farms); hay and corn silage-based diets (four farms); and diets based on hay, corn silage and grass silages (four farms). For cheesemaking, 1000 kg of milk collected from two consecutive milkings was sampled from each farm and processed in the same cheese factory. Cheeses were ripened in the same cellar, in controlled humidity (78–85% relative humidity) and temperature (9–12°C), until analysis. After 68 (62–74), 200 (194–206) and 360 (354–366) d of ripening, a cheese from each batch was analysed for moisture, protein and fat, volatile fatty acids (VFA) and volatile compounds. These latter were analysed by dynamic reverse carrier gas headspace gas chromatographic mass spectrometry (GC-MS) technique, scanning from m/z 29 to m/z 300 at 0·5 s cycle time. Total and individual VFA contents of cheeses did not differ between the different dietary treatments and increased linearly (P<0·01) during ripening, indicating that there were no appreciable defects of fermentation. Sixty-two volatile compounds were identified in the cheese samples: 12 aldehydes, 9 ketones, 16 alcohols, 17 esters, 2 hydrocarbons, 4 sulphur-containing compounds and 2 terpenes. Diet significantly affected the amount of total alcohols, ethanol, isobutanol, 1-penten-3-ol, 2-methyl-1-butanol and the sum of all the volatile compounds. Significant variations of 33 volatile compounds (9 aldehydes, 4 ketones, 6 alcohols, 10 esters, 1 hydrocarbons, 2 sulphur containing compounds, and 1 terpene) were observed during ripening. These results suggest that the influence of diet composition on volatile compounds of matured cheeses are related more to effects on microbial and chemical fermentations in cheese during ripening rather than to a direct transfer of molecules from milk.

The link between oxidation and increased proteolysis in raw milk was studied. To accelerate oxidation, H2O2 (1 mM) was added to raw milk, resulting in enhanced proteolysis by up to 11·2% after 24 h incubation at 5 °C. Addition of Cu2+ (10 μM) to milk or exposure of milk to light (60 min) likewise increased proteolysis. To explain the mechanism responsible for increased proteolysis as a result of oxidation, the effect of lipid oxidation products on plasmin-induced proteolysis was tested. Addition of malondialdehyde to skim milk increased the formation of γ-caseins, a proteolysis product from plasmin hydrolysis of β-casein. The same observation was made in a model system containing 4·5 g β-casein/l sodium tetraborate buffer at pH 8 and plasmin. Addition of a plasmin inhibitor blocked the formation of γ-casein. The results indicate that aldehydes accumulated from lipid oxidation can modify β-casein and thereby increase susceptibility of the proteins to proteolysis. Furthermore, the data suggest that proteolysis in raw milk may be connected to oxidative processes.