27 October 2005. Keeping brain amyloid-β (Aβ) levels in check means controlling both the means of production and the method of disposal. On the disposal side, a few years ago researchers recognized for the first time that Aβ is actively cleared from the brain via the LDL receptor-related protein-1 (LRP1) (Shibata et al., 2000). Now, David Holtzman and colleagues from Washington University in St. Louis have identified a second pump. In a paper that appeared online in the Journal of Clinical Investigation on October 20, the scientists, working with collaborators at Eli Lilly and at the University of Rochester Medical Center, New York, report that P-glycoprotein (Pgp), best known as the membrane transporter that confers multidrug resistance on tumor cells, also contributes to Aβ efflux from brain. Their work shows that blocking Pgp function enhances amyloid deposition in a mouse model of AD. Since genetic polymorphisms and many common drugs modulate Pgp activity, both positively and negatively, understanding the role of this protein could lead to the identification of new risk factors, or protective mechanisms, for AD in people.
The Pgp transporter is an important component of the capillary endothelial cell blood-brain barrier. There, it acts like a sump pump, that is, as fast as substrate drugs and peptides pour in, the pump escorts them back out. Hints that Pgp could also be an Aβ exporter included in-vitro work (Lam et al., 2001), as well as histological studies showing that Pgp expression inversely mirrored the distribution of Aβ in human brain (Vogelgesang et al., 2002). These intriguing results led Holtzman and colleagues to directly measure the contribution of Pgp to Aβ efflux in vivo.
Using different experimental approaches, first author John Cirrito and colleagues built the case for Pgp’s involvement in Aβ clearance. First, they measured efflux in wild-type versus Pgp knockout mice by following the fate of radioactive Aβ after injection into the brain. When they looked at what remained 30 minutes after injection, they found that about twice as much Aβ40 or Aβ42 cleared the blood-brain barrier in the wild-type mice as in Pgp knockout mice. In another experiment, they crossed APPsw AD transgenic mice with the Pgp knockouts. At one year of age, the Pgp-null offspring had larger plaque areas, more thioflavin S reactivity, and double the levels of Aβ42 in the hippocampus compared to AD mice with normal Pgp.
Though suggestive, these experiments were complicated by the observation that Pgp-null mice also had much lower levels of the other Aβ transporter, LRP1, on capillary endothelial cells. To sort out the contribution of Pgp alone to Aβ efflux, the investigators turned to a pharmacological inhibitor to acutely reduce Pgp activity. After administering the inhibitor XR9576 intravenously, they followed Aβ levels in brain interstitial fluid for 10 hours by microdialysis. During this period, they saw Aβ levels climb, with no change in LRP1 protein levels. Taking these results together with previous studies showing the role of LRP1 in Aβ clearance, the authors speculate that the two efflux proteins may actually act synergistically, with LRP1 functioning on the basolateral surface of brain endothelial cells, and Pgp on the luminal surface.
The implication of Pgp in Aβ clearance and AD pathology brings up some pressing questions. First, Pgp polymorphisms that affect drug handling are well known. If the changes also affect Aβ clearance, they could represent genetic risk or protection for AD. Because of this, the authors stress the need for a detailed genetic analysis of Pgp polymorphisms and AD risk.
Second, many common drugs alter Pgp function, and conceivably, their use could carry an increased or decreased risk of AD. Among that group is the antibiotic rifampin, which upregulates Pgp. Rifampin was shown in a clinical trial to lessen cognitive decline in people with mild to moderate AD after one year of treatment (Loeb et al., 2004). These new results raise the possibility that the drug could be beneficial because it increases clearance of Aβ. On the flip side, many drugs inhibit Pgp (indeed, some cancer treatments have been designed to do exactly that). Whether the use of such agents will be associated with an increased risk of AD needs to be investigated.—Pat McCaffrey.
Cirrito JR, Deane R, Fagan AM, Spinner ML, Parsadanian M, Finn MB, Jiang H, Prior JL, Sagare A, Bales KR, Paul SM, Zlokovic BV, Piwnica-Worms D, Holtzman DM. P-glycoprotein deficiency at the blood-brain barrier increases amyloid-beta deposition in an Alzheimer disease mouse model. J Clin Invest. 2005 Oct 20; [Epub ahead of print] Abstract