Clostridia in the milk can lead to late blowing, a cheese defect. Clostridia are ubiquitous, deriving from both the farm environment and the feed ingested by the cows, and are transferred into the milk through faecal contamination. Our aim was to investigate the effect of different in-parlour practices on the content of anaerobic spore-forming bacteria in milk, and to monitor the variation in spore content in the feed and environment. The experiment, conducted in an experimental dairy during autumn, was repeated in exactly the same way for two consecutive years. The experimental design applied three different milking routines in three consecutive 7-d periods: forestripping alone (F); forestripping and post-dipping (F+Post); pre-dipping, wiping, forestripping and post-dipping (Pre+F+Post). Teat skin swabs and samples of feed, faeces, bedding materials and milk were collected for microbiological analyses. The dietary forage of the lactating cows included maize silage, which, in both years, was found to have the highest level of clostridial spore contamination. Pre-dipping with a detergent/emollient solution, and drying with a disposable paper towel, proved much more efficient in reducing spore contamination than forestripping alone, both on the teats (1·30 vs. 2·20 log10 MPN/swab; P < 0·001) and in the milk (1·82 vs. 2·47 log10 MPN/L, P < 0·02), while post-dipping had little influence on spore count. The standard plate count in milk was significantly lower with Pre+F+Post treatment than with F (3·80 vs. 4·51 log10 CFU/mL, P < 0·01). The teat preparation procedure did not influence the lactic acid bacterial levels in the milk, which is very positive in that decreased lactic acid bacterial content can lessen raw milk cheese quality.

In addition to their use in cheese technology, dairy propionibacteria have been identified as potential probiotics. However, to have a probiotic effect, propionibacteria have to survive and to remain metabolically active in the digestive tract. The aim of the present study was to investigate the survival and metabolic activity of Propionibacterium freudenreichii within the gastrointestinal tract of human microbiota-associated rats, and its influence on intestinal microbiota composition and metabolism. Twenty-five dairy Propionibacterium strains were screened for their tolerance towards digestive stresses and their ability to produce propionate in a medium mimicking the content of the human colon. Three strains were selected and a daily dose of 2 × 1010 colony-forming units was fed to groups of human microbiota-associated rats for 20 d before microbiological, biochemical and molecular investigations being carried out. These strains all reached 8-log values per g faeces, showing their ability to survive in the gastrointestinal tract. Transcriptional activity within the intestine was demonstrated by the presence of P. freudenreichii-specific transcarboxylase mRNA. The probiotic efficacy of propionibacteria was yet species- and strain-dependent. Indeed, two of the strains, namely TL133 and TL1348, altered the faecal microbiota composition, TL133 also increasing the caecal concentration of acetate, propionate and butyrate, while the third strain, TL3, did not have similar effects. Such alterations may have an impact on gut health and will thus be taken into consideration for further in vivo investigations on probiotic potentialities of P. freudenreichii.

To determine the effect of feeding propionibacteria on metabolic indicators during lactation, multiparous and primiparous Holstein cows were fed one of three dietary treatments in a 2×3 factorial design from 2 weeks prepartum to 30 weeks post partum: (1) Control (primiparous n=5, multiparous n=8) fed a total mixed ration (TMR); (2) high-dose group (primiparous n=6, multiparous n=5) fed TMR plus 6×1011cfu/head daily (high-dose P169) of propionibacterium strain P169; or (3) low-dose group (primiparous n=8, multiparous n=6) fed TMR plus 6×1010 cfu/head daily (low-dose P169) of P169. Blood samples were collected weekly and analysed for plasma concentrations of glucose, insulin, insulin-like growth factor-I (IGF-I), leptin, nonesterified fatty acids (NEFA) and cholesterol. Between weeks 25 and 30, all groups received bovine somatotropin (bST) every 2 weeks. Low-dose P169 multiparous cows had lower (P<0·05) plasma insulin and glucose concentrations than high-dose P169 multiparous cows, whereas high-dose P169 primiparous cows had lower glucose but greater insulin concentartions than low-dose P169 primiparous cows (P<0·05). Plasma insulin[ratio ]glucose molar ratios were 13–18% lower (P<0·05) in low-dose P169 cows than in control or high-dose P169 cows. Plasma IGF-I, NEFA and leptin levels did not differ among diet groups between weeks 1 and 25. Low-dose P169 multiparous cows had 25% greater plasma cholesterol levels than high-dose P169 and control multiparous cows, but cholesterol levels in primiparous cows did not differ. During bST treatment, high-dose P169 multiparous cows and low-dose P169 primiparous cows had lower IGF-I levels than their respective controls and, regardless of parity, high-dose P169 cows had greater NEFA than control cows. Although supplemental feeding of P169 altered plasma hormones and metabolites, the particular effects were dependent on dose of P169 and parity of cows.