The fascination in amyloid precursor protein (APP), a type I transmembrane protein enriched in neurons, rests primarily on one of its proteolytic products, amyloid β-protein (Aβ), which is the principal component of senile plaques found in brains of Alzheimer's disease (AD) individuals. To date, research in AD pathogenesis has largely been driven by the amyloid cascade hypothesis, which posits that Aβ is the initiator of cellular changes resulting in AD. While supported by a large body of literature, this hypothesis has skeptics, as it fails to address a number of important questions, such as the lack of correlation between senile plaques and cognition, and the failure of APP transgenic animal models to recapitulate the full disease spectrum. Not surprisingly, a number of laboratories have asked whether perturbations of normal physiological functions of APP – beyond Aβ generation – may contribute to AD pathophysiology. Single knockout (KO) studies of APP and its orthologs in various species have implicated APP in facilitating transcriptional activities, cell adhesion, axonal transport, neuronal migration, and synaptogenesis 1. Double and triple KO studies in mice of APP and the other two members of the gene family, APLP1 and APLP2, yielded early postnatal lethality phenotypes, suggesting that although functional redundancy exists among APP family members, certain specificity in APLP2 is revealed when APP or APLP1 is absent 2. Of note, sequence similarities between the APP family members are in the N- and C-termini. And within the C-terminus lies the –YENPTY– motif that is both a signal for clathrin-mediated endocytosis and a binding site for numerous cytosolic adaptor proteins. The roles of this motif in normal or pathological functions are still unclear, but clues can be gleaned by comparing KO animals and single mutation of the Tyr682 residue. In particular, APP single KO animals showed diminished brain and body weight, reduced locomotor functions, depressed long-term potentiation, and loss of synapses, which are associated with neurobehavioral deficits (2). However, APP knockin (KI) animals deleted in the last 15 amino acid residues – which encompass the YENPTY domain – were able to rescue a variety of deficits seen in APP KO mice, suggesting that the YENPTY domain is dispensable for certain physiological functions of APP. On the other hand, the Y682G KI mice showed decreased synaptic density and impaired neuromuscular functions. Curiously, when this Y682G mutation was introduced into an APLP2−/− background, the resultant mouse exhibited early postnatal lethality highly reminiscent of the APP/APLP2 double-KO mice. This interesting finding thus places primary importance on the Y682 residue in APP functions, and conflicts with the aforementioned result obtained from a knockin mouse line missing the C-terminal 15 amino acids of APP. The existence of Tyr682 allows the interaction of various adaptor proteins through Tyr phosphorylation and dephosphorylation, creating many possibilities for the system to be modulated. Further, some of the APP binding proteins are phosphorylation state-independent, but their binding is nonetheless affected by other adaptor molecules that are phosphorylation-dependent. This suggests that the YENPTY domain might regulate APP signal transduction functions that are variably dependent on the phosphorylation state of Tyr682. Given this, one explanation of the divergent results from the two knockin mice lines may be that Tyr682 can act as a “biochemical switch,” as proposed by Matrone, and accordingly, complete loss of the APP C-terminus might disrupt multiple pathways that cancel each other's cellular effects 3. Nonetheless, the findings from the Y682G mice are both intriguing and noteworthy, as they point to the physiological importance of the YENPTY domain. Hence the versatile functions of APP continue to provide fodder for interesting studies. However, whether APP plays a role in AD pathophysiology in an Aβ-independent manner is a crucial question that remains to be answered.
Read full abstract