We have characterized and purified the bile salt hydrolase from Lactobacillus sp. strain 100-100. Bile salt hydrolase from cells of the strain was purified with column and high-performance liquid chromatography. The activity was assayed in whole cells and cell-free extracts with either a radiochemical assay involving [14C]taurocholic acid or a nonradioactive assay involving trinitrobenzene sulfonate. The activity was detectable only in stationary-phase cells. Within 20 min after conjugated bile acids were added to stationary-phase cultures of strain 100-100, the activity in whole cells increased to levels three- to fivefold higher than in cells from cultures grown in medium free of bile salts. In cell-free extracts, however, the activity was about equal, 1.41 and 1.53 mumol/min per mg of protein, respectively, whether or not the cells have been grown with bile salts present. When supernatant solutions from cultures grown in medium containing taurocholic acid were used to suspend cells grown in medium free of the bile salt, the bile salt hydrolase activity detected in whole cells increased two- to threefold. Two forms of the hydrolase were purified from the cells and designated hydrolases A and B. They eluted from anion-exchange high-performance liquid chromatography in two sets of fractions, A at 0.15 M NaCl and B at 0.18 M NaCl. Their apparent molecular weights in nondenaturing polyacrylamide gel electrophoresis were 115,000 and 105,000, respectively. However, discrepancies existed in the apparent molecular weights and number of peptides detected in sodium dodecyl sulfate-polyacrylamide gel electrophoresis of the two forms. Both had similar substrate specificities, highest on taurodeoxycholic and glycocholic acid, and pH optima between 3.8 and 4.5. The kinetic properties were also similar, with Vmaxs of 17 and 53 micromoles/min per mg of protein and Kms of 0.76 and 0.95 mM taurocholic acid for A and B, respectively. Therefore, whether the enzyme exists in two forms in the cells remains to be determined.