Antibodies directed against amyloid beta protein (AssP) have been suggested to be effective in the treatment of Alzheimer's disease (AD). Here, we used in vivo and in vitro models to test some of the mechanisms by which antibodies may produce their effects. We found that the blood-to-brain uptake of murine AssP1-42 was significantly reduced when co-injected peripherally with an antibody known to reverse cognitive defects in the SAMP8, an mouse model of AD. This antibody was not effective when tested against the more slowly transported human AssP1-42. Antibody given by intracerebroventricular (icv) injection did not improve the clearance of murine AssP1-42 from the brains of young healthy mice, which already rapidly clear AssP by saturable and non-saturable mechanisms. Antibody given icv also did not improve the clearance of human AssP1-42 from the brains of aged SAMP8 mice, a combination in which the AssP is only poorly cleared from brain. IV antibody also did not affect retention of murine AssP in young mice. In vitro transwell studies with monolayers of mouse brain endothelial cells (MBEC) found no evidence that antibody in the vascular chamber would retard the reuptake of AssP which had been effluxed from the brain-side chamber. A statistical trend suggested that antibody might decrease the association of AssP with brain vasculature. In conclusion, we found that icv administration of antibody was not effective in aiding clearance of AssP already in brain, but that blood-borne antibody can inhibit the entry of AssP into brain and might prevent AssP from associating with the brain vasculature.