br Experimental procedures br Results

Experimental procedures
Results
Discussion Our biophysical and cellular analyses of mutant Aβ1–42 peptides support a role of the N-terminus of Aβ in peptide aggregation and toxicity. We demonstrate that double mutations constructed to humanize the rodent Aβ1–42 sequence result in a significantly reduced efficiency of fibril formation, as determined by kinetic ThT aggregation assays and further confirmed with our electron micrographic studies, which revealed sparse fibrils that were morphologically shorter and thinner than both hAβ1–42 and rAβ1–42. Interestingly, the mutants were readily able to aggregate into oligomers, but each mutant formed oligomers that were predominantly larger than those comprised of hAβ1–42. Cell viability deduced from MTT and LDH assays, showed a rank order of oligomer toxicity of hAβ1–42 > rAβ1–42 ≫ mutant Aβ1–42, demonstrating that toxicity can be influenced by N-terminal Aβ1–42 mutations via reduction of fibril formation and/or alteration of oligomer size. There is evidence that the efficiency of Aβ fibril formation does not predict toxicity, and the effect of N-terminal Aβ mutation can further depend on whether the construct being used is Aβ1–40 or Aβ1–42. Peptides containing the single mutation H13R (created within human Aβ1–40) (Poduslo et al., 2010) or Y10F (also created within human Aβ1–40) (Dai et al., 2012) were more likely than hAβ1–40 to aggregate and yet had lower toxicity. By contrast, studies of naked mole rat Aβ, which is the equivalent to an H13R mutation within hAβ1–42, revealed a disconnect between aggregation propensity and toxicity, with the peptide aggregating more slowly yet having the same toxicity as hAβ1–42 at 10 µM (Edrey et al., 2013). Rather than propensity for fibril formation, oligomer characteristics such as size may be more relevant to the endpoint measure of cell viability. It is well known that levels of oligomers correlate more with the symptoms of AD (Lue et al., 1999, McLean et al., 1999, Wang et al., 1999, Näslund et al., 2000) and are more toxic in vitro (Benilova et al., 2012). Aβ lacking residue 22 causes AD and, while it is unable to form fibrils in vitro, it can produce oligomers that inhibit long-term potentiation (Tomiyama et al., 2008). The N-terminal mutation A2V causes AD, possibly via an altered pathway of oligomerization (Messa et al., 2014), and in a study of D-phenylalanine substitution in Aβ, toxicity was correlated with oligomer size, with the presence of very large NU 7026 associated with reduced toxicity (Kumar et al., 2015). A study of the H13R mutation in hAβ1–42 also demonstrated an inverse relationship between oligomer size and toxicity, with H13R peptides forming larger oligomers with less effect on cell viability, as measured by MTT (Roychaudhuri et al., 2015). Many different types of oligomers can be generated in vitro or isolated from AD-affected brains (Benilova et al., 2012). It is now well established that Aβ peptide exists in multiple forms including monomers, dimers, trimers and oligomers to protofibrils and fibrils that range in size from 4 kD to more than 100 kD which vary in morphology and conformation (Rushworth and Hooper, 2010, Jarosz-Griffiths et al., 2016). Soluble Aβ oligomers appear to be the most neurotoxic species, triggering various processes that underlie AD including synaptic dysfunction, impairment of long-term potentiation, Ca+2 dysregulation, mitochondrial dysfunction, ER stress, lysosomal breakdown and activation of pro-apoptotic pathways leading to cell death (Walsh and Selkoe, 2007, Ferreira and Klein, 2011, Benilova et al., 2012, Thal et al., 2015). Although several experiments using primary neurons or neuronal cell lines have shown that cytotoxicity induced by Aβ peptide correlates with its β-sheet structure and fibrillar state, the underlying mechanism by which extracellular Aβ damages neurons remains unclear (Iversen et al., 1995, Xia et al., 2016). There is evidence that Aβ peptide can bind to the cell membrane and form ion channels or pores that induce membrane disruption followed by neuronal damage. In fact, some studies have reported pore-like structures of Aβ under in vitro conditions as well as in cell membranes of AD brains and mice (Bhatia et al., 2000, Inoue, 2008, Kawahara et al., 2009). Additionally, soluble Aβ oligomers, but not monomers or fibrils, have been shown to increase membrane permeability and thus dysregulate Ca+2 signals associated with neurotoxicity (Demuro et al., 2005). Other lines of experimental evidence indicate that Aβ binding to neurons may involve single or multi-protein cell surface receptor complex which can subsequently trigger a variety of downstream signaling pathway leading to cell toxicity. The cell surface protein/receptors that can regulate Aβ-mediated toxicity include cellular prion protein, receptor for advanced glycation end products, the α7 nicotinic acetylcholine receptor, the p75 neurotrophin receptor, the β2 adrenergic receptor, the low-density lipoprotein receptors, the amylin 3 receptors, Fcγ receptor II-b (FcγRIIb), scavenger receptors, the Eph receptors and the paired immunoglobulin-like receptor B (Yan et al., 1996, Wang et al., 2000, Husemann et al., 2001, Hashimoto et al., 2004, De Felice et al., 2007, Lauren et al., 2009, Wang et al., 2010, Cisse et al., 2011, Basak et al., 2012, Fu et al., 2012, Kim et al., 2013, Kam et al., 2014, Jarosz-Griffiths et al., 2016, Xia et al., 2016). However, the role of several of these receptors in mediating Aβ toxicity is somewhat controversial or yet to be reproduced and considering the heterogeneity and dynamic nature of Aβ peptide, it is possible that different receptors may interact with different species of Aβ to trigger a specific signaling cascade. Many of the signaling pathways initiated by these ligand-receptor interactions then converge into a common downstream target that is ultimately responsible for neurotoxicity and cell death (Kam et al., 2014, Jarosz-Griffiths et al., 2016).