“I think it’s an interesting and novel approach,” Ralph Nixon at NYU Langone Medical Center, New York City, told Alzforum. He was not involved in the work. “There is growing evidence that a methodology that would lower β-CTF would have potential therapeutic value in its own right.”

The secretase BACE1 cleaves APP to release a soluble N-terminal extracellular domain and β-CTF, which γ-secretase then snips to generate Aβ peptides. This pathway remains a major focus for drug development. Though failures in clinical trials have effectively scuttled γ-secretase inhibitors (see ARF related news story and ARF news story), γ-secretase modulators remain under active development (see ARF related news story; ARF news story), and several BACE1 inhibitors are in Phase 2 trials (see ARF related news story; ARF news story). Researchers have also uncovered indirect ways to potentially target Aβ production, such as inhibiting γ-secretase activating protein (GSAP) (see ARF related news story) or the scaffolding protein β-arrestin 2, as was suggested just this month (see ARF related news story), but those have not yet led to drug candidates.

Haass and colleagues wanted to find alternative targets as well. First author Richard Page performed a genomewide RNA interference screen to look for genes that affected Aβ secretion in fruit fly cells that express human β-CTF. The authors found 41 genes that specifically raised or lowered Aβ secretion when knocked down. Silencing 20 of these in a human cell line had the same effect. Not counting the four components of γ-secretase, the gene that most strongly put the brakes on Aβ production when silenced goes by the moniker platelet activating factor acetylhydrolase, isoform 1B, subunit 2 (PAFAH1B2). This serine esterase may play several biological roles, including in subcellular trafficking and microtubule dynamics (see Bechler et al., 2011; Zhang et al., 2007). It is highly expressed in brain.

To confirm the findings, the authors turned to PAFAH1B2 knockout mice, which are healthy except for defects in spermatogenesis. Embryonic fibroblasts from the knockouts secreted about 50 percent less Aβ than did wild-type cells, matching results from the cell lines. Curiously, knocking out just the catalytic activity of PAFAH1B2 had no effect on Aβ levels. This left the authors searching for non-catalytic mechanisms by which PAFAH1B2 enhances Aβ production. Looking at APP processing, they found that the knockout cells had less β-CTF than wild-type. Since β-CTF cleavage releases Aβ, low levels of the precursor might explain why Aβ levels fall.

The authors wondered if the drop in β-CTF could be due to enhanced degradation. Production of β-CTF occurs in endocytic vesicles, from where the peptide travels either to lysosomes for degradation, or to early endosomes where it encounters γ-secretase. Could deleting PAFAH1B2 increase lysosomal trafficking of β-CTF? In support of this, Page and colleagues found that treating PAFAH1B2 knockout cells with a lysosomal protease inhibitor restored β-CTF to wild-type levels.

To further test the trafficking hypothesis, the authors deleted the APP C-terminal domain, which drives internalization of APP and sorting of the β-CTF fragment to intracellular compartments. PAFAH1B2 binds to this internalization sequence, Haass told Alzforum. In a cell line expressing the C-terminal truncated APP, PAFAH1B2 knockdown did not lower Aβ levels, suggesting that APP has to be internalized for PAFAH1B2 to enhance Aβ production, supporting the trafficking hypothesis. Importantly, PAFAH1B2 deletion did not affect other γ-secretase substrates such as Notch and cadherin, which do not possess the internalization sequence. This hints that PAFAH1B2 knockdown could specifically target Aβ without harming Notch, a key issue for therapy.

Other researchers found the data intriguing, but noted that many questions remain about the mechanism. Nixon wondered whether PAFAH1B2 knockout cells chew up only more β-CTF, or whether they upregulate lysosomal function in general. Sanjay Pimplikar at the Cleveland Clinic, Ohio, had the same question, noting that the paper shows enlarged lysosomes in PAFAH1B2 knockout cells, suggesting a dysregulation of this pathway.

Haass and colleagues plan to investigate the mechanism behind Aβ lowering in more detail, and they also want to confirm the findings in vivo by crossing APP/PS1 mice with PAFAH1B2 knockouts. Haass believes the pathway has therapeutic potential. Studies of BACE1 inhibitors have shown that even small reductions in Aβ production result in fewer amyloid deposits. Haass suggested that a 50 percent drop in Aβ might be enough to delay AD onset by several years, pointing out that a protective APP mutation that reduces lifetime Aβ production by about 20 percent significantly lowers AD risk (see ARF related news story). It is not immediately obvious how PAFAH1B2 could be targeted in people, as the authors’ findings show that inhibitors of its catalytic activity would have no effect on Aβ. One possibility would be to use antisense oligonucleotides to knock down protein levels, Haass said. Antisense therapeutic approaches are in Phase 1 trials for amyotrophic lateral sclerosis (also see Neurology Today article), but the technique remains unproven.––Madolyn Bowman Rogers.

Reference:
Page RM, Münch A, Horn T, Kuhn PH, Colombo A, Reiner O, Boutros M, Steiner H, Lichtenthaler SF, Haass C. Loss of PAFAH1B2 reduces amyloid-β generation by promoting the degradation of amyloid precursor protein C-terminal fragments. J Neurosci. 2012 Dec 12;32(50):18204-14. Abstract