Tarja Suomalainen

Improving the quality of European hard-cheeses by controlling of interactions between lactic acid bacteria and propionibacteria

Abstract This work aimed to improve the knowledge and the control of interactions between lactic acid bacteria (LAB) and propionibacteria (PAB) in order to control the propionic acid fermentation in European hard cheeses, and therefore their quality. Among more than 50 couples of strains LAB/PAB used in cheese making, some pairs of strains were chosen in function of quality results on cheeses (sensorial analysis). A whey model system and an alternative conductimetric method, with good reproducibility were used to study interactions. The lipolytic activity of PAB strains varied by a factor 4 to 5. The aroma of ripened Emmental cheese was linked to the level and the type of lipolysis in the cheese, correlated with lipolytic activity of PAB. The main pathways of the utilisation of lactate by PAB, studied by 13 C RMN, were useful to understand the balance between final products: acetic, propionic acid, CO 2 .

Antimicrobial activity of 2-pyrrolidone-5-carboxylic acid produced by lactic acid bacteria

Thirteen Lactobacillus and five Pediococcus strains were shown to produce an antimicrobial agent, 2-pyrrolidone-5-carboxylic acid (PCA). PCA inhibited many spoilage bacteria, particularly Enterobacter cloacae 1575, Pseudomonas fluorescens KJLG, and P. putida 1560-2. The antimicrobial activity of PCA did not change at higher temperatures. However, the activity was destroyed rapidly by neutralization with ammonium hydroxide. PCA showed slightly lower antimicrobial activity than lactic acid.

Effect of prebiotic supplementation on a probiotic bacteria mixture: comparison between a rat model and clinical trials

The effects of a probiotic mixture (PRO), supplemented with either galacto-oligosaccharide (GOS) or polydextrose (PDX), on cell numbers of lactic acid bacteria (LAB) and bifidobacteria (BIF) were studied in conventional rats and healthy human subjects. In rats the baseline BIF cell numbers were below the detection limit and were increased by the 2-week GOSPRO intervention. In contrast baseline LAB numbers in rats were high and not affected by the treatments. The human study consisted of two independent but concurrent trials; both started with PRO followed by GOSPRO or PDXPRO periods. In the human subjects variation in numbers of BIF and LAB were high. The GOSPRO group exhibited high counts of faecal LAB and BIF at the start and showed little or no effects of the interventions. In contrast, the PDX group had low faecal LAB and BIF numbers at the start and clearly increased cell numbers of BIF after the PDXPRO period, and LAB after the PRO and PDXPRO period, compared with the run-in period. We propose here that responses to pro- and prebiotics are dependent on baseline numbers of LAB and/or BIF, and that the conventional rat model does not predict well the treatment responses in humans. The survival of PRO was presumably enhanced by the use of prebiotic supplementation and advocates the use of particular combinations of pro- and prebiotics.