gasseri strains were digested with SmaI, SacII, and ApaI with same PFGE profiling. Four of these strains are shown in Figure 5. All of these L. gasseri strains showed banding patterns identical to those of TMC0356 with all three restriction enzymes. However, following ApaI digestion, a band of 113.5 kb was confirmed for TMC0356 but not for TMC0356-F100. A band of 108.3 kb was confirmed for TMC0356-F100 but not for TMC0356. Lactobacillus gasseri was originally classified into the L. acidophilus group based on biochemical, enzymatic, physiological and other phenotypic characteristics (19).

It was reclassified as L. gasseri on the basis of genomic characterization techniques such as DNA homology studies. Phylogenetically, L. gasseri remains closely related to other species in the L. acidophilus group. Like them, L. gasseri is also a natural resident of the human intestine, and currently available find more methods have not been able to discriminate TMC0356 from the other original residents of L. gasseri. In our previous studies, the number of lactobacilli species, including L. gasseri, was shown to increase significantly in the intestines of subjects after oral administration of TMC0356 (12). Such increases are considered a possible underlying mechanism for the observed improvement of allergic symptoms among subjects taking lactobacilli orally (3). However, it has remained unclear whether

ingested TMC0356 would increase in fecal samples. Lactobacilli may reliably be distinguished at the strain level by DNA-based techniques. Genomic methods used Kinase Inhibitor Library cell assay Sodium butyrate for typing include randomly amplified polymorphic DNA analysis, ribotyping, and PFGE (18). PFGE allows the use of rare-cutting restriction enzymes, which enable the separation of large fragments of

genomic DNA. The DNA fingerprint obtained by this method typically consists of 5–20 large well-resolved fragments ranging in size from 10 to 800 kb. It is a highly discriminatory and reproducible method, and has been used to differentiate strains of important probiotic bacteria (20). Björkroth reported that PFGE patterns had the greatest discriminatory power for revealing genetic variation in the main group of ropy slime-producing L. sake strains, and for distinguishing all non-ropy strains from slime-producing ones (21). In the present study, total genomic DNA was isolated from 15 L. gasseri strains (including the probiotic strain TMC0356 and 14 reference strains from JCM) and analyzed by PFGE after treatment with three restriction enzymes—SmaI, SacII, and ApaI. TMC0356 showed a banding pattern similar to these of JCM 1031 and JCM 1131 but different from those of the other strains. TMC0356 differed from JCM1031 and JCM 1131 by a 42.9 kb band formed after digestion with SmaI and SacII. In the present study, the PFGE profiles of chromosomal DNA of the dominant L. gasseri strains isolated from the feces of subjects who had ingested TMC0356 were identical to those of cultured TMC0356.