Mattsson and colleagues looked for signs of BACE1 inhibition in several different cell lines, some of which expressed wild-type human APP, some human APP with the Swedish mutation, and some another form of mutant human APP (APP751 with the V717F familial mutation). The authors treated the cultures with two different BACE1 inhibitors: the commercially available BACE1 inhibitor IV and, in collaboration with pharmaceutical company AstraZeneca, Södertälje, Sweden, the inhibitor AZ-20. Normally, BACE1 cuts APP at position 1 of the Aβ peptide, and γ-secretase then cleaves the protein somewhere around position 40-49 to create the common Aβ fragments. As expected, with BACE1 inhibition, levels of Aβ40 and Aβ42, as well as those of a host of other Aβ peptides, dropped in the cells. Notably, in most assays the relative decline in the minor peptide Aβ1-34 was much greater than that of the other peptides. Aβ1-34 is cut at both ends by BACE1, Mattsson said, which may make it particularly vulnerable to the lack of the enzyme. The results were consistent across cell lines and treatments.

Conversely, BACE1 inhibition strikingly and consistently increased levels of the rare Aβ5-40 in all cell lines. This suggests the existence of another enzyme, unmasked by BACE1 inhibition, that cuts at position 5, although its identity remains a mystery. Some previous studies have found Aβ5-40 in amyloid plaques in human AD brains (see Takeda et al., 2004; Portelius et al., 2010). Further research on this rare peptide may yield more insights into APP processing, Mattsson suggested, adding that it is not yet known if it plays a role in AD.

To study BACE1 inhibition in vivo, the authors used dogs. Canine Aβ is identical to human, and dogs release similar Aβ cleavage products into their CSF (see Portelius et al., 2010). The researchers collaborated with pharmaceutical company Novartis in Basel, Switzerland, to treat the animals with two experimental BACE1 inhibitors. A separate group of dogs received BACE1-inhibitor S from collaborators at Janssen Research and Development, Beerse, Belgium. In agreement with the cell culture results, levels of Aβ5-40 soared in the CSF, while most other Aβ peptides dropped, with the decline in Aβ1-34 particularly steep. Overall, the authors found that an increased ratio of CSF Aβ5-40/Aβ1-34 is a more sensitive marker of BACE1 inhibition than are declines in Aβ40 or Aβ42, and can completely separate placebo and treatment groups.

“The reported data are amazing, and demonstrate that simplistic views of proteolytic processing of APP are all but untenable,” Fred Van Leuven at KULeuven, Belgium, wrote to ARF. For his part, Mattsson said the data further reinforce that it is worth looking at peptides other than Aβ40 and Aβ42.

The paper by Vassar and colleagues focused on potential side effects of BACE1 inhibition. The enzyme has other substrates besides APP, and recent studies have turned up several axon guidance molecules as possible targets (see ARF related news story on Hemming et al., 2009). To investigate further, first author Tharinda Rajapaksha examined the axons of olfactory sensory neurons in BACE1 knockout mice. Each axon expresses a particular odorant receptor (out of a suite of around 1,000), and needs to hook up to a specific, corresponding glomerulus in the olfactory bulb for odors to be relayed correctly to the cortex. Rajapaksha and colleagues found that olfactory bulb glomeruli in the knockouts were misshapen and smaller than those in wild-type mice, hinting that many olfactory axons failed to reach their targets. Using fluorescent tags, the authors looked at the behavior of axons containing specific odorant receptors and discovered that at least some of these hooked up to the wrong glomeruli. This kind of miswiring might affect the mouse’s ability to smell that particular odor, Vassar noted, adding that one of the next steps will be to test this.

Vassar plans to look at whether BACE1 aids axon guidance in other neuronal cell types as well. He also wants to find the BACE1 substrates involved and dissect the molecular mechanisms. It is not yet clear how the defects in knockout mice might apply to BACE1 inhibitor therapy in adult humans. However, Vassar pointed out that some neuronal populations, such as olfactory and hippocampal neurons, regenerate throughout life, implying new wiring could be affected by BACE1 inhibition. Vassar told ARF he remains enthusiastic about BACE1 inhibition as a therapeutic strategy, but he also believes scientists should proceed with caution. “The drug companies need to take this [the data] into consideration in their clinical trials and be on the lookout for these types of problems,” he suggested.

Johan Lundkvist at AstraZeneca, who is a coauthor on the PLoS One paper, notes that Vassar’s paper provides “novel and important insights,” but more data are needed on how it might translate to trials of BACE1 inhibitors. “We have learned that only a limited decrease in BACE activity (BACE+/- mice) is sufficient to affect the development of amyloid pathology,” Lundkvist wrote to ARF. “It will therefore be interesting to learn…whether small molecule-mediated inhibition of BACE, at doses efficacious in lowering Aβ production, has any impact on the sensory neuronal network of the olfactory system.” (See full comment below.)—Madolyn Bowman Rogers.

References:
Mattsson N, Rajendran L, Zetterberg H, Gustavsson M, Andreasson U, Olsson M, Brinkmalm G, Lundkvist J, Jacobson LH, Perrot L, Neumann U, Borghys H, Mercken M, Dhuyvetter D, Jeppsson F, Blennow K, Portelius E. BACE1 inhibition induces a specific cerebrospinal fluid β-amyloid pattern that identifies drug effects in the central nervous system. PLoS One. 2012. Abstract

Rajapaksha TW, Eimer WA, Bozza TC, Vassar R. The Alzheimer’s β-secretase enzyme BACE1 is required for accurate axon guidance of olfactory sensory neurons and normal glomerulus formation in the olfactory bulb. Mol Neurodegener. 2011 Dec 28;6(1):88. Abstract