. It is therefore more consistent to
refer to transient colonization or persistence of probiotics
rather than colonization. The correlation between mucosal
adhesion and transient colonization also appears to be less
clear than expected. In experimental animals, a clear
correlation between in vitro and in vivo adhesion has been
observed for six strains of propionic acid bacteria (12). In
the case of humans, however, this is less clear. Hautefort
and co-workers (13) observed that both an adherent and a
low-adherent strain of Lactobacillus fermentum were able to
colonize transiently in vivo. On the other hand, Crociani
and co-workers (14) reported that a good correlation exists
between the in vitro adhesion and in vivo persistence of
three Bifidobacterium strains. Unfortunately, the persistence
of the strains administered was not determined, only the
level of faecal bifidobacteria. Likewise, Jacobsen and coworkers
(8) observed a good correlation between the in vitro
adhesion to Caco-2 tissue culture cells and persistence of
five Lactobacillus strains in faeces. The lowest adhering
strain tested was detected in 3 of 12 volunteers, the highest
in 10 of 12. Although the results are indicative, they do not
allow us to conclude that that there is a correlation between
in vitro adhesion and in vivo colonization, since different
adhesive and low-adhesive strains have been used. To our
knowledge only Cesena and co-workers (15) have reported
that an adherent L. curvatus strain could be isolated from
faeces and colonic biopsies of volunteers, whereas the lowadhesive
mutant was not detected. As both strains evinced
similar resistance to bile, low pH and pancreatic enzymes,
this difference in recovery is not likely to be due to
differences in survival. It thus remains to be determined
whether in vitro adhesion is indeed associated with
prolonged persistence. Only isogenic strains differing in
their adhesive properties can provide information on the
importance of adhesion to the intestinal mucosa. Studies
ORIGINAL ARTICLE
# Taylor & Francis 2003. ISSN 0891-060X Microbial Ecology in Health and Disease
DOI: 10.1080/08910600310019886
such as that described above (15) are therefore urgently
needed