The resulting simulation parameters are shown in Figure 1 and described in the Material and Methods section. During the experimental stage of this study,

Sumeri et al. [9] developed a similar system to evaluate Lactobacillus sp. in a stomach-intestine passage simulation. The software package “”Lucullus”" was an excellent tool to control the pH and the process according to the developed simulation. Selecting the medium in the bioreactor was simplified by choosing the corresponding growth medium for the strains, supplemented with skim milk, functioning as a simulated food matrix. Afterwards, it was acidified to the starting pH and supplemented with enzyme solutions as described in Materials and Methods. The simulations were carried out serially, one per day. The results are shown in Figure 6. The strains used for the simulation are listed in table 3 (only Bifidobacterium GW2580 in vitro dentium was excluded) and were standardized to an OD650 of 1.5 prior to inoculation. Figure 6 Development of 7 Bifidobacterium strains during stomach-intestinal passage simulation for 7 h. Dashed line shows the time of addition of bile salts and pancreatic juice. Numbers in the bacterial names are the strain numbers in the FAM-database of ALP. Table 3 Strains tested in the simulation. Name Identification number of ALP strain collection Bifidobacterium adolescentis FAM-14377 Bifidobacterium breve FAM-14398

Bifidobacterium longum subsp. this website infantis FAM-14390 Bifidobacterium animalis subsp. Lactis FAM-14403 Bifidobacterium dentium FAM-14396 Bifidobacterium longum FAM-14382, -14383, -14406 Lactobacillus gasseri K7 FAM-14459 Bifidobacterium adolescentis was inoculated as described above at an initial concentration of 107 cfu ml-1 and decreased almost linearly to below 104 cfu ml-1 after 5 hours. B. breve and B. longum strains had an initial concentration between 107 and 108 cfu ml-1 and diminished to below 102 cfu ml-1 within the first 30 minutes. B. animalis subsp. lactis 14403 survived to approximately 15% of the initial average cfu of 5 × 108 cfu ml-1. There was a rapid decrease

in survival of B. longum subsp. infantis over the first 30 min. Afterwards Endonuclease the survival decreased only slowly from 105 to 104 cfu ml-1. In a later phase, Lactobacillus gasseri K7 was included in the study since several projects were running at this time at our institute with this strain. Lactobacillus gasseri K7 was inoculated at 2.2 × 107 cfu ml-1 and after 7 h simulation a concentration of 105 cfu ml-1 living cells was still present in the culture media (Figure 7, curve for 250 ml pre-culture). The highest reduction in survival was within the first 2 hours and began immediately after the addition of gastric juice and bile salts. Within this time, there was a reduction of living cells by log 2. During the rest of the simulation time, there was only a log 1 reduction of living cells.