Commentary Amyloid-β, BACE, and oxidative stress in Alzheimer's disease, a commentary on “The different aggregation state of beta-amyloid 1-42 mediates different effects on oxidative stress, neurodegeneration and BACE-1 expression” Hyoung-gon Lee a , George Perry a,b , Xiongwei Zhu a , Akihiko Nunomura c , Mark A. Smith a, ⁎ a Department of Pathology, Case Western Reserve University, 2103 Cornell Road, Cleveland, OH 44106, USA b College of Sciences, University of Texas at San Antonio, San Antonio, TX 78249-0661, USA c Department of Psychiatry and Neurology, Asahikawa Medical College, Asahikawa 078-8510, Japan Received 7 March 2006; accepted 9 March 2006 Available online 4 May 2006 The amyloid cascade hypothesis, positing amyloid-β (Aβ) as the cause of Alzheimer disease (AD), has been the dominant theory in the AD research field for over a decade [1,2]. In the original version of the hypothesis, the fibrilized form of Aβ (fAβ), the main component of senile plaques, was positioned as the primary culprit of the disease [1] and found support from many in vitro and in vivo experiments showing cytotoxicity of fAβ [3]. However, since many aspects of AD could not be explained by fAβ [4,5], the amyloid cascade hypothesis [1] was modified to claim that oligomeric Aβ, rather than fAβ, was key [2]. Oligomeric Aβ, like fibrillar Aβ, is toxic to neurons in in vitro and in vivo paradigms [6]. Amazingly, in this new paradigm, the previously toxic fAβ entity is now cast as moot [7]. In this issue of Free Radical Biology & Medicine, Tamagno and colleagues provide an important insight into the fAβ/ oligomer Aβ saga by showing differential effects of Aβ that are dependent on aggregation state [8]. Specifically, they demon- strate that prefibrillar oligomeric Aβ species, while increasing 4- hydroxynonenal (HNE) and H 2 O 2 more than fAβ and being more toxic than fAβ, have no effect on BACE-1 expression and activity. Conversely, fAβ is less toxic but increases BACE-1 expression and activity. Based on these data, the authors hypo- thesize that Aβ acts via a biphasic neurotoxic mechanism that is conformation dependent such that oligomeric Aβ exerts toxic effects by inducing oxidative stress and leads to fAβ formation from oligomeric Aβ. Although fAβ has less toxicity than oli- gomeric Aβ, fAβ increases the accumulation of Aβ by inducing BACE-1 expression and activity and this process may contribute further to the toxicity of fAβ. In other words, the biphasic action of Aβ sets up a vicious cycle. While intriguing, the major unresolved question is whether the proposed biphasic mechanism of Aβ toxicity applies to in vivo studies. On the one hand, the evolution of oligomeric Aβ in the amyloid cascade hypothesis [2] has found considerable support [6,9] and the development of in situ methods to detect these oligomers [10] shows an increase in AD brain. However, on the other hand, and detracting from the importance of oli- gomeric Aβ, even in a transgenic mouse model (Tg2576) that is driven by Aβ overproduction, oligomeric Aβ and oxidative stress appear at approximately the same age [11,12] and most importantly do not result in cell death [13–15]. Furthermore, in human and animal studies, the chronological relationship bet- ween oligomeric Aβ and oxidative stress in vivo is ill defined, and it is unclear whether oligomeric Aβ induces oxidative stress or vice versa. While further study is definitely required to dissect the relationship between oxidative stress and oligomeric Aβ, it is important to remember that oxidative stress leads to the increased production of Aβ [16] by modulating BACE activity [17,18] and that Aβ is associated with decreased levels of oxidative stress [19,20]. These findings implicate Aβ as a protective response to oxidative stress rather than a causative factor in disease pathogenesis [4,5,21–23]. References [1] Hardy, J. A.; Higgins, G. A. Alzheimer's disease: the amyloid cascade hypothesis. Science 256:184–185; 1992. Free Radical Biology & Medicine 41 (2006) 188 – 189 www.elsevier.com/locate/freeradbiomed Abbreviations: Aβ, amyloid-β; AD, Alzheimer disease; fAβ, fibrilized form of Aβ. PII of original article: S0891-5849(06)00064-5. ⁎ Corresponding author. Fax: +1 216 368 8964. E-mail address: mark.smith@case.edu (M.A. Smith). 0891-5849/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.freeradbiomed.2006.03.014