Purpose. The effect of P-glycoprotein (Pgp) on brain distribution using mdrla (-/-) mice was investigated.

Methods. Fluorescein (Flu) and FD-4 were used to check whether blood-brain barrier (BBB) integrity was maintained in mdrla (-/-) mice. The Pgp substrate rhodamine-123 (R123) was infused and total brain, blood and brain microdialysate concentrations in mdrla (-/-) mice and wild-type mice were compared.

Results. Maintenance of BBB integrity was indicated by equal total brain/blood ratios of Flu and FD-4 in both mice types. R123 concentrations in brain after i.v. infusion were about 4-fold higher in mdrla (-/-) than in wild-type mice (P < 0.05), without changes in blood levels. After microdialysis experiments the same results were found, excluding artifacts in the interpretation of Pgp functionality by the use of this technique. However the 4-fold ratio in brain was not reflected in corresponding microdialysates. No local differences of R123 in the brain were found. By the no-net-flux method in vivo recovery appeared to 4.6-fold lower in mdrla (-/-) mice compared with wild-type mice.

Conclusions. Pgp plays an important role in R123 distribution into the brain. Using intracerebral microdialysis, changes in in vivo recovery by the absence or inhibition of Pgp (or active efflux in general) need to be considered carefully.