ABC transporters harness the energy of ATP to escort drugs, metabolites, and other substrates across the plasma membrane. Among the seven subfamilies (ABCA to ABCG), three contain proteins that have been shown to pump Aβ across the blood-brain barrier. An in-vitro study (Lam et al., 2001) identifying P-glycoprotein (aka ABCB1) as one such Aβ transporter prompted Pahnke and colleagues to look at ABCB1 levels in patient brains. The researchers found, indeed, that seniors with less ABCB1 in their brain epithelium had more plaque (Vogelgesang et al., 2002) and more vascular Aβ (Vogelgesang et al., 2004). A similar trend showed up in AD transgenic mice (see ARF related news story on Cirrito et al., 2005). ABCA1 (Koldamova et al., 2005) and ABCG2 (Xiong et al., 2009) can also export Aβ. ABCA7 came up as a hit in a recent AD genomewide association study (ARF related news story on Hollingworth et al., 2011), and currently sits in the fourth position in the AlzGene Top Results.

In the current paper, researchers led by joint first authors Markus Krohn and Cathleen Lange generated APP/PS1 mice deficient in ABCB1, ABCG2, or ABCC1 to study how these transport proteins regulate Aβ clearance. Because the researchers had wanted to focus on ABCB1, the latter two strains were meant as controls, “but early on we noticed that the mice with no C1 transporter had more brain amyloid than the other two lines,” Pahnke told ARF. Comparing the three strains by Aβ immunochemistry, ABCC1-deficient AD mice had the largest plaques and, by ELISA, 12- to 14-fold more brain Aβ than control AD mice. Brain amyloid levels were up two- to fourfold in ABCB1-deficient mice, while ABCG2-deficient animals looked comparable to controls. The dramatic changes seen in ABCC1-/- AD mice did not stem from abnormal processing of amyloid precursor protein (APP) or microglial activation. Moreover, the researchers saw similar effects when they crossed ABCC1-knockout mice with a less aggressive strain (APP Dutch) that develops amyloidosis around 24 months of age, as opposed to six to eight weeks for APP/PS1 mice.

Measuring Aβ42 transport across mouse capillary endothelial cells in vitro, the scientists determined that APP/PS1 mice lacking ABCC1 were 60 percent less efficient at exporting Aβ compared to their ABCC1+/+ counterparts. Considering this defect alongside other factors that affect rates of synthesis and removal of Aβ, Pahnke’s team came up with a mathematical model that predicted an 11 percent overall reduction in Aβ clearance in ABCC1-/- AD mice. Last year, scientists measured real-time Aβ turnover in the cerebrospinal fluid of AD patients and found they get rid of brain Aβ about 30 percent more slowly than age-matched controls (ARF related news story on Mawuenyega et al., 2010).

ABCC1 is also expressed at the choroids plexus, suggesting that in addition to playing a role at the blood-brain barrier, it may actively clear Aβ from the CSF as well, noted John Cirrito, Washington University School of Medicine, St. Louis, Missouri, in an e-mail to ARF (see comment below). Kwasi Mawuenyega, a Washington University scientist who was first author of the 2010 study on real-time Aβ dynamics in people, called the new data “exciting” and said the paper provides “a valuable clue for proving causality of impaired Aβ clearance for AD (see comment below).”

Next, Pahnke and colleagues explored whether boosting the enzyme’s function would relieve the amyloid burden in the brains of APP/PS1 mice. They treated the animals with an FDA-approved anti-nausea drug (thiethylperazine) that drives up ABCC1 transport activity in vitro by 69 percent. They tested preventive and therapeutic strategies. The former involved 30 days of twice-daily intramuscular drug injections begun at 45 days of age, before this strain develops plaques; for the latter, a 25-day oral treatment started when disease is underway at 75 days of age. “Both paradigms were able to reduce plaques and soluble Aβ in the brain,” Pahnke said. He and colleagues did not test whether the drug crosses the blood-brain barrier, or how it is metabolized, in mice. “For the effect we want, it doesn’t need to get into the brain,” he told ARF. “We think this drug circulates in the blood, activates the transporter, and by doing so reduces the amount of soluble Aβ in the brain. By reducing the soluble forms (i.e., dimers, monomers), it prevents the growth of plaques.”

Pahnke and colleagues are now recruiting elderly couples for a genetic study to identify ABCC1 polymorphisms, and for a Phase 1 trial of thiethylperazine set to begin next year. In addition, the researchers are trying to develop a brain imaging assay for ABCC1 function to see if declining activity might portend neurodegenerative disease. Toward that end, they are planning a positron emission tomography (PET) study that will use a radiolabeled ABCC1 inhibitor to track transport activity in the brain. PET studies have found decreased ABCB1 function in disease-specific brain regions of people with Parkinson’s disease, progressive supranuclear palsy, and corticobasal degeneration (Kortekaas et al., 2005; Bartels et al., 2008; Bartels et al., 2000; see also review by Bartels, 2011). “The hypothesis is that decreased transport in specific brain areas leads to accumulation of proteins, thus to neurodegenerative disease, not only AD,” Pahnke said.—Esther Landhuis

Reference:
Krohn M, Lange C, Hofrichter J, Scheffler K, Stenzel J, Steffen J, Schumacher T, Brüning T, Plath AS, Alfen F, Schmidt A, Winter F, Rateitschak K, Wree A, Gsponer J, Walker LC, Pahnke J. Cerebral amyloid-β proteostasis is regulated by the membrane transport protein ABCC1 in mice. J Clin Invest. 2011 Sep 1. Abstract