This is an important study that may usher in a new perspective on the normal function of APP and clarify its role in AD pathogenesis.

View all comments by John Trojanowski

The study of the biology of APP and its proteolytic products, although pioneered in the early 1990s by Eddie Koo, Joseph Buxbaum, Sam Sisodia, and others, has nevertheless remained mostly out of the limelight until the last few years. The present study from Elaine Bearer’s laboratory now illuminates part of a picture that has been taking shape in the last few years suggesting that APP is likely involved in the modulation of synaptic activity in adults (Priller et al., 2006; Yang et al., 2005; Seabrook et al., 1999), in synapse formation and function (Wang et al., 2005), and in neuronal migration and adhesion during development (Herms et al., 2004).

APP is a synaptic protein that is anterogradely transported to terminals. A few years ago Kamal et al. suggested that the C-terminus of APP could serve as a receptor for kinesin (  Read more

The study of the biology of APP and its proteolytic products, although pioneered in the early 1990s by Eddie Koo, Joseph Buxbaum, Sam Sisodia, and others, has nevertheless remained mostly out of the limelight until the last few years. The present study from Elaine Bearer’s laboratory now illuminates part of a picture that has been taking shape in the last few years suggesting that APP is likely involved in the modulation of synaptic activity in adults (Priller et al., 2006; Yang et al., 2005; Seabrook et al., 1999), in synapse formation and function (Wang et al., 2005), and in neuronal migration and adhesion during development (Herms et al., 2004).

APP is a synaptic protein that is anterogradely transported to terminals. A few years ago Kamal et al. suggested that the C-terminus of APP could serve as a receptor for kinesin (Kamal et al., 2000), but this observation was subsequently questioned by Lazarov et al. (Lazarov et al., 2005). The present study by Satpute-Krishnan et al. provides strong evidence that the C-terminus of APP may indeed contain sequences sufficient for its association with axonal transport components. The careful experiments addressed this question using a fairly well-defined system, the squid giant axon, and the investigators’ observations indicate that the C-terminal domain of APP, either through a direct interaction with kinesin or indirectly via scaffolding proteins such as JIPs, participates in fast anterograde axonal transport. Quoting their discussion, “The robust motility of C99 beads in the intact axon argues for a physiological role of APP in recruitment of anterograde transport machinery inside cells.” It certainly does, and it comes as no surprise. Although the study by Satpute-Krishnan et al. does not answer the question of whether the interaction of APP with kinesin is or is not direct, it significantly adds to the rapidly growing evidence suggesting a crucial role of the C-terminus of APP (and possibly its family members APLP1 and 2) in neuronal biology, possibly at synaptic sites.

The remarkable conservation of the C-terminal sequences of APP across phyla suggests conservation of function. Supporting this idea, the phenotypes of APP/APLP2 double and APP/APLP1/APLP2 triple knockouts and those of two prominent APP-interacting proteins (X11 and the Fe65 family) involve alterations in neuronal function, synaptic formation, function, and regulation (Wang et al., 2005; Ho et al., 2003; Yang et al., 2005; Priller et al., 2006), and in the case of the Fe65/FE65L double and APP/APLP1/APLP2 triple knockouts, result in cortical dysplasias and heterotopias (Herms et al., 2004, Guenette et al., 2006). Interestingly, it was recently shown that transgenic expression of AICD in combination with Fe65 causes alterations in signaling (Ryan and Pimplikar, 2005) and activation of proteins involved in growth cone collapse and axonal guidance.

Why is this important? Most of all, because a significant component of amyloid-β toxicity requires multimerization of APP and cleavage of its C-terminus at Asp664 (Lu et al., 2003; Lu et al., 2003; Shaked et al., 2006). This cleavage not only releases a toxic peptide, but also removes the sequences required for the formation of a multiplicity of protein complexes at APP’s cytoplasmic domain, and as Satpute-Krishnan now suggest, for fast axonal transport. Consistent with what may be an important role of the extreme C-terminal sequences of APP in transducing amyloid-β toxicity, we recently showed that stabilization of APP’s cytoplasmic tail by mutation of the Asp664 cleavage site had a dramatic effect in the development of AD-like deficits in transgenic mice (Galvan et al., 2006)—even in the presence of abundant amyloid-β. With this in mind, the question arises as to whether cleavage at Asp664 while in transit towards synaptic sites would, as expected, prevent delivery of the molecule to its destination—and if the hypothesis of Kamal et al. is correct, whether it would affect the delivery of any subset of associated axonal transport vesicles. Thus, a population of Asp664-intact (transport-competent) and Asp664-cleaved (transport-incompetent) APP molecules may exist. Satpute-Krishnan et al. estimate that 3,000 copies of APP may be associated with each motile bead in their system; although in this study they don’t address the question of what is the minimal number of APP molecules required for transport, it is conceivable that transport-incompetent (Asp664-cleaved) APP molecules may be “carried along” in vesicles containing a sufficient number of transport-competent (Asp664-intact) APP. Cleavage of APP at Asp664 would thus affect not only the transport-competence (and thus the rate of delivery) of APP to neuronal terminals, but since the motifs required for the interaction of APP with a variety of cellular functions reside downstream of Asp664, it would also affect the overall signaling ability of populations of APP molecules at their destination at synaptic sites.

View all comments by Veronica Galvan