A study of the vitamin D3 (D3) 25-hydroxylase was undertaken in an in vivo-in vitro model. [3H]-D3 (0.7, 1.0, 10, or 100 nmol/100 g of body weight) was injected into the portal vein and the liver was excised 18 seconds later. The liver homogenate was then submitted to differential centrifugation and the amount of [3H]-D3 incorporated in the subcellular fractions was evaluated. The microsomal fraction was also incubated in vitro and the appearance of [3H]-25-hydroxyvitamin D3 [25(OH)D3] was determined by high performance liquid chromatography (HPLC). Results showed that the fractional liver [3H]-D3 uptake varied between 37 percent and 48 percent of the dose injected. The intracellular distribution of [3H]-D3 showed that most of the vitamin was incorporated into the microsomal fraction (45% to 50% of the intracellular [3H]-D3) except at the highest dose of [3H]-D3 where the cytosolic fraction contained the highest amount (56.4%) of the incorporated vitamin. Mathematical analysis of the intracellular [3H]-D3 distribution showed that the microsomal fraction was the only subcellular fraction that was found to incorporate [3H]-D3 in relation to the total liver uptake of the vitamin. The apparent Michaelis-Menten kinetics of the [3H]-D3-25-hydroxylase showed that with substrate concentration of up to 88.5 nM, the apparent Km and Vmax were 28.2 nM and 25.8 fentomoles (fmol) X min-1 X mg microsomal pro-1, respectively, but the reaction lost considerable efficiency with higher substrate concentrations. With the in vivo-in vitro model used, the cytosolic fraction was not essential for the optimal C-25 hydroxylation of D3. These results show that the endoplasmic reticulum of rat hepatocytes possess a high capacity for D3 incorporation.(ABSTRACT TRUNCATED AT 250 WORDS)