27 Feb 2001

The most promising target for an Alzheimer's therapy in the short term may be the enzyme β-secretase, suggest two papers in the March issue of Nature Neuroscience. Two separate groups have found that mice lacking the BACE1 isoform of β-secretase do not produce amyloid-β. Perhaps more significantly, these transgenic mice grow to seemingly normal adulthood, suggesting that interfering with the enzyme in human Alzheimer's patients might not lead to unacceptable side effects.

Most of the attention in this arena has been paid to γ-secretase, which is also involved in the production of the presumed cellular assassin, amyloid-β (Aβ). Both γ- and β-secretase (as well as α-secretase) cleave the amyloid precursor protein (APP). Philip Wong, Huaibin Cai, and their associates at Johns Hopkins, working with cultured embryonic day 16 neurons from BACE1 deficient mice, report that BACE1 cleaves APP at the +1 and +11 sites, ultimately resulting in the toxic Aβ varieties 1-40/42 and 11-40/42. Their transgenic mice lacking BACE1 failed to produce these Aβ varieties, secreting only small amounts of the Aβ17-40 species, even when human Swedish variant APP, associated with familial Alzheimer's, was introduced.

These results are echoed by Robert Vassar, Yi Luo, and associates at Amgen, who determined that brain extracts from two- to three-month-old BACE1-deficient mice lacked any appreciable Aβ of either the x-40 or x-42 varieties, despite the fact that these mice had been genetically altered to overproduce APP. Even so, three- to four-month-old mice revealed no abnormalities on a battery of pathology tests of neural and nonneural tissues, as well as clinical chemistry analyses. "We found the viability of BACE1(-/-) mice to be very surprising, because BACE1 is expressed at low levels in most tissues, and is highly expressed in brain and pancreas," wrote the authors. This last point is especially newsworthy because γ-secretase, another prime target for an Alzheimer's therapy, may be too important to other physiological functions to allow interference for the sake of abolishing Aβ production in humans. The fact that BACE1-deficient mice appear normal up to one year of age is likely to encourage the development of BACE1 inhibitors for trials in Alzheimer's patients.—Hakon Heimer