Evidence that tacrolimus augments the bioavailability of mycophenolate mofetil through the inhibition of mycophenolic acid glucuronidation

We previously reported an unexpected augmentation of mycophenolic acid (MPA ) levels (trough and AUC(0-12)) in patients receiving mycophenolate mofetil (MMF) in combination with tacrolimus versus patients receiving the same do se of MMF in combination with cyclosporin A (CsA). This finding was accompa nied by a corresponding reduction of the inactive glucuronide metabolite of MPA (MPAG) in patients, suggesting that tacrolimus may effect the conversi on of MPA to MPAG by the enzyme UDP-glucuronosyltransferase (UDPGT). To inv estigate this possibility directly, UDPGT was extracted from human liver an d kidney tissue and its activity was characterized using MPA as a substrate in vitro, assessing the conversion of MPA to MPAG using analysis by high-p erformance liquid chromatography. With crude microsomal preparations, amoun ts of UDPGT at least 100 times higher in specific activity (i.e., units to milligrams of protein) could be extracted per gram of tissue from kidney as opposed to liver. This result did not appear to be related to the coextrac tion of a liver-specific UDPGT inhibitor because initial enzyme kinetic val ues (V-max and k(m)) were identical for kidney and liver extracts, and furt her purification of the liver enzyme did not enhance activity (as is seen w hen inhibitors are removed during purification). With further UDPGT purific ation (similar to 200-fold) from kidney extracts using a combination of amm onium sulfate precipitation, followed by anion exchange, hydroxyapatite, an d size exclusion chromatography, the enzyme was more than 80% pure when ass essed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Initial enzyme kinetic analysis of this purified product showed a k(m) value for M PA of 35.4 +/- 5.7 mu g/mL and a V-max of 2.87 +/- 0.31 MPAG produced per h our (n = 7). The addition of clinically relevant concentrations of CsA (200 -1,000 ng/mL) or tacrolimus (10-25 ng/mL) resulted in a dose-dependent inhi bition of the UDPGT enzyme by both agents with tacrolimus, which was approx imately 60-fold more efficient as an inhibitor. The calculated inhibition c onstants (K-I) of tacrolimus and CsA for the purified UDPGT were 27.3 +/- 5 .6 ng/ml and 2,518 +/- 1473 ng/ml, respectively. Both agents displayed an i nhibition profile characteristic of a competitive inhibitor (substrate) tha t could be demonstrated in a reciprocal experiment with CsA as a substrate, but not with tacrolimus. This finding suggested that the significantly mor e efficient inhibition of UDPGT by tacrolimus may occur by a more complicat ed mechanism that is yet to be determined.