, 2010). In the present work, we explored the subset of extracellular proteins produced by a panel of LAB and bifidobacteria frequently found in foods or that are normal inhabitants of the human GIT. We aimed to detect changes in the production of extracellular proteins as affected by the presence of cecum extract in

the culture medium. A panel of food/probiotic bacteria was used, among which representative strains for dairy starters, adjunct dairy cultures, commensal species inhabiting the human GIT, and probiotic strains were chosen. In addition, the strain B. animalis ssp. lactis R2, a strain producing a ropy exopolysaccharide that may be relevant for the food industry (Ruas-Madiedo & de los Reyes-Gavilán, 2005), was also included (Table 1) (Gasson, 1983). In our RXDX-106 nmr experimental design, different subinhibitory concentrations of cecum extract obtained from the pooled cecum contents of four healthy donors were added to the

growth culture media. The highest amount of extracellular proteins was recovered from the supernatants of bacteria cultured to stationary phase of growth. Therefore, we used GS 1101 extracellular proteins isolated in this phase for obtaining preliminary electrophoretic profiles. In general, the extracellular protein profiles of the cultures of selected bacteria were affected qualitatively by the presence and concentration of cecum extract initially added to the growth medium. Many of the new mafosfamide bands were identified as components of the cecum extract (Fig. 1, see Supporting Information, Table S1), but two of them were shown to be highly upregulated bacterial proteins: surface antigen (Imp11; accession number ZP_00121020) from Bifidobacterium longum and a small extracellular protein of unknown function (Imp23; accession number YP_193019) produced by L. acidophilus (Fig. 2a). After their identification, we could further demonstrate the induction of

the corresponding genes: the expression level of imp11 remained at a twofold higher level in the presence of cecal content along the growth curve, whereas imp23 was considerably more induced in exponential than in stationary phase (Fig. 2b). It is known that intestinal bacteria are able to react to the GIT environment by activating certain genes, normally under the control of inducible promoters (Gueimonde et al., 2009; Rivera-Amill et al., 2001; Sleator et al., 2005). Our results suggest that the expression of certain genes, whose products could be relevant for the physiology of the bacterium in the GIT, may be up- or down-regulated in conditions used in the laboratory, thus escaping analysis. In contrast, the actual relevance regarding bacteria–host interaction of proteins produced at higher amounts in nonconditioned media with respect to simulated GIT conditions should be carefully addressed. For instance, S-layer protein A from L.