This work extends our prior findings, which suggest that aggregate flux may occur in the setting of intracellular pathology, raising the possibility of therapies that can assist in aggregate clearance by targeting

extracellular mTOR activation species. This work has important implications for the design of therapeutic antibodies and suggests that targeting seeding activity in particular may produce the most effective agents. Several prior active and passive peripheral immunotherapy approaches against tau have also reduced tau pathology and improved behavioral deficits, but the underlying rationale for antibody choice was based either on a phospho-epitope, reactivity with neurofibrillary tangles, or was not stated (Asuni et al., 2007, Bi et al., 2011, Boimel et al., 2010, Boutajangout et al., 2010, Boutajangout et al., 2011, Chai et al., 2011 and Troquier et al., 2012). One tau immunization study, performed by vaccinating mice with full-length tau, induced pathology in wild-type mice (Rosenmann et al., 2006). However, subsequent active immunization approaches with phospho-tau peptides in tau transgenic models reduced tau pathology (Bi et al.,

2011 and Boimel et al., 2010) and showed behavioral improvement (Asuni et al., 2007, Boutajangout et al., 2010 and Troquier et al., INCB018424 2012). In a passive immunization study, JNPL3 tau transgenic mice were administered the PHF1 antibody intraperitoneally at 2–3 months of age, prior to the onset of tauopathy. PHF-1 targets a pathological form of abnormally phosphorylated tau (Otvos et al., 1994). Treatment reduced tau pathology and improved behavior (Boutajangout et al., 2011). However, while it decreased insoluble phosphorylated tau, next total insoluble tau

did not change. In another passive immunization study, JNPL3 and P301S mice (at age 2–3 months, prior to the onset of tauopathy) were peripherally administered the PHF1 or MC1 antibody, which targets an aggregate-associated epitope (Jicha et al., 1999). Both treatments improved tau pathology and delayed the onset of motor dysfunction (Chai et al., 2011). In these prior studies, the mechanism of action of the antibodies was not clear, and none was explicitly tested. Indeed, some proposed an intracellular mechanism (Sigurdsson, 2009). Moreover, no study appears to have produced the magnitude of reduction in tau pathology that we observed here, with the caveats that we infused antibodies into the CNS, while the other studies utilized peripheral infusion and different animal models were utilized. We designed this study explicitly to test a prediction that extracellular tau seeds are a key component of pathogenesis. We began with a selection process to pick antibodies capable of blocking tau seeding in vitro, purposely testing agents with a range of predicted activities.