Substantial evidence implicates Aß peptides in the etiology of Alzheimer's disease (AD). Aß is produced by the proteolytic cleavage of the amyloid precursor protein by b - and g -secretase enzymes. Thus, b -secretase (BACE1) inhibition is a promising disease modifying treatment for AD. Utilizing transgenic CRND8 (TgCRND8) mice, a model of AD-related amyloid accumulation and deposition, we assessed the effects of 4 months of chronic in diet, oral administration of a potent and selective small molecule iminoheterocycle BACE inhibitor on amyloid-related pathology when treatment began well after amyloid plaques had been deposited. This paradigm more closely mimics the state of Aß accumulation and deposition that is likely to be present even at the earliest stages of AD. Since it was reported that constitutive BACE1 knockout mice show 1) reduced peripheral nerve myelination, 2) reduced prepulse inhibition (PPI, a model of auditory gating often associated with schizophrenia) and 3) in certain instances impaired cognitive performance, we tested the mice in PPI and the Morris water maze and quantified sciatic nerve myelination after 4 months of chronic BACE inhibition. Chronic BACE inhibition in post-plaque TgCRND8 mice dramatically reduced Aß in plasma and cerebrospinal fluid and halted further age-related increases in brain Aß and amyloid deposition. Furthermore, the burden of mature plaques was reduced below that observed at baseline, suggesting amyloid plaque regression following chronic BACE inhibition. There was no effect of chronic BACE inhibition on performance in PPI or the water maze, or on peripheral nerve myelin thickness, suggesting that chronic pharmacological inhibition of BACE in adult mice does not produce the same effects observed in constitutive BACE1 knockout mice. Thus, chronic administration of our potent BACE inhibitor in post-plaque TgCRND8 mice prevents further progression of amyloid-related pathology. In addition, there were no adverse effects observed following chronic pharmacological inhibition of BACE, including no effects on PPI, cognition or myelination. These data suggest that even in patients with substantial amyloid burden, BACE1 inhibition may halt further progression or promote regression of amyloid-related pathology in the clinic.

One of the neuropathological hallmarks of Alzheimer's disease is the presence of β-amyloid (Aβ)-containing neuritic plaques. Aβ is produced by the proteolytic cleavage of the amyloid precursor protein (APP) by β- and γ-secretase enzymes. Thus, inhibition of β-secretase (BACE1) is a promising disease modifying approach for Alzheimer's disease. Several high affinity BACE inhibitors have been described that reduce plasma Aβ in vivo. Although this may be sufficient to provide clinical benefit (e.g., via peripheral sink), our goal was to identify BACE inhibitors that substantially reduced (>25%) Aβ in the brain. To support ongoing medicinal chemistry SAR efforts to identify BACE inhibitors that would reduce Aβ in the brain, we used lowering of Aβ1-40 in cerebrospinal fluid (CSF) as a marker of CNS BACE1 inhibition. We also switched our in vivo screening model from transgenic CRND8 mice expressing the Swedish and Indiana mutations in the human APP gene to non-transgenic rats and developed a sensitive rat Aβ1-40 immunoassay assay to support the in vivo screen. Our ex vivo binding studies and studies with heterozygous BACE1 mice supported the conclusion that BACE is not the rate limiting step for Aβ synthesis in the rodent brain and thus high occupancy of the BACE enzyme is likely required for brain efficacy. After focusing our SAR efforts around robust Aβ-lowering in rat CSF, several novel, potent, high affinity, orally available, brain penetrant BACE inhibitors were identified that substantially lowered CSF Aβ and more importantly, inhibited Aβ in the rat cortex. Based on our experience, brain exposures well in excess of inhibitor cell potency alone did not always result in substantial lowering of CSF or cortical Aβ, but that high brain exposure combined with sustained and substantial CSF efficacy was needed to observe Aβ-lowering in the cortex. Thus, we have identified several novel, potent BACE inhibitors with excellent Aβ-lowering activity in CSF and the brain which may be of use as disease modifying Alzheimer's disease therapeutics.

BACE1 null mice exhibit reduced peripheral nerve myelination similar to what is observed in NRG1 Type III het mice. BACE1 null mice also exhibit reduced processing of Neuregulin 1 (NRG1) type III β1a in vivo. Therefore, BACE1 processing of NRG1 type III β1a was proposed to be required for proper peripheral nerve myelination. Since BACE1 inhibition is the most attractive therapeutic strategy for Alzheimer's Disease (AD) it is important to determine if BACE1 inhibitors impact the processing of NRG1 type III β1a and myelination in vivo. Previous biochemical studies with neuregulins relied on rat cDNAs and the human NRG1 type III β1a cDNA has not been reported. As a first step to determining the impact of BACE1 inhibitors on NRG1 type III β1a processing, we cloned human NRG1 type III β1a from a human fetal brain cDNA library and expressed it in a mammalian cell system. Mutations in NRG1 type III β1a were identified which altered its subcellular localization. And in addition, similar to APP, co-expression of BACE1 also changed the subcellular localization of NRG1 type III β1a. Comparisons between BACE1 processing of APP and NRG1 type III β1a and the effects of inhibitors will also be reported.