Contentions that activation of the cell cycle and apoptosis at least partly underlie the neurodegeneration seen in Alzheimer's disease (AD) have slowly gained experimental support (see ARF related news story, ARF live discussion, and ARF recent live discussion). Now, a report in the July 30 Journal of Neuroscience suggests that not only are both activated by familial AD mutations, but that the two pathways are intimately linked through PAK3, a kinase that is activated by the Rho family member, p21.

Rachel Neve and colleagues at McLean Hospital, Belmont, Massachusetts, together with coworkers in the private sector and at Oregon Health and Science University, Portland, made the connection after they found that mutants of amyloidβ precursor protein (AβPP) known to cause familial AD also lead to apoptosis and DNA synthesis in primary neurons cultured from rats. To understand the molecular basis for these responses, first author Donna McPhie and colleagues used a radiolabeled C-terminal fragment of AβPP to screen rat cDNA libraries for potential binding partners. After several rounds of screening they isolated only one protein that bound tightly to the fragment, PAK3.

To determine if the binding of AβPP mutants with PAK3 is physiologically relevant, McPhie expressed both proteins in primary neurons and then looked for evidence of their interaction. When the authors incubated cell extracts with anti-AβPP antibodies both proteins were precipitated, suggesting that they interact in the cell. Furthermore, using immunohistochemical analysis, McPhie found the two proteins co-localized in neurons, primarily in endosomes.

But what of the role of PAK3? The authors found that in the presence of a dominant though inactive variant of the protein, AβPP mutants failed to induce apoptosis or DNA synthesis, indicating that PAK3 plays a vital role in these responses. However, this response also did not occur in these cells when the authors expressed a dominant active form of PAK3 alone, indicating that the interaction between the kinase and the mutant precursor protein is critical.

PAK family members are known to interact with the G protein family of cell signaling molecules, as is AβPP. To test if G proteins may also be involved in re-entry into the cell cycle or the apoptotic pathway, the authors expressed mutated AβPP in cultured neurons in the presence or absence of pertussis toxin, a well known poisoner of G protein signaling. They found that the toxin does indeed block the action of the AβPP mutants indicating that PAK3, G proteins, and AβPP are all inextricably linked in the neurodegeneration caused by the amyloid precursor.—Tom Fagan


  1. McPhie et al. have done a beautiful job unraveling the signaling pathway linking FAD mutants of APP and neuronal apoptosis. Their evidence that reentry into the cell cycle is an obligatory step in this APP-induced apoptotic cascade is compelling. I'm intrigued by some similarities between the APP-driven mechanism described by McPhie and colleagues and what is seen in Niemann-Pick Type C disease:

    Mutations of both the APP gene and the NPC1 gene in humans lead to activation of similar cell cycle markers, cytoskeletal pathology with neurofibrillary tangle formation, and widespread neurodegeneration.

    Both proteins localize to transport vesicles and Rab-specific endosomal compartments. (Accumulation of b-amyloid in late endosomes has been described in NPC1 cells and in older NPC patients).

    Therefore, I wonder if there is a critical link between the function of the endosome or transport vesicles and cell cycle regulation in postmitotic neurons; in other words, if neurons exit the cell cycle to become highly specialized communicative cells, they might simply revert to cell cycling ("dedifferentiate") when this specialized function is compromised.

    Dysregulation of the neuronal trafficking process may be a key event leading to cytoskeletal pathology and neuronal death. The latter is certainly not a new idea (Sheetz et al., 1998; Kamal, 2000; Morfini et al., 2002; ARF related news story); and many more), but it may also explain a variety of disease outcomes arising from more specific or localized defects in trafficking. For instance, depending on which subset of vesicles, the type of cargo being transported, and which trafficking "arteriole" is affected, different sets of proteins/lipids may accumulate and aggregate forming different types of cytoskeletal lesions.


    . Mechanisms of trafficking in axons and dendrites: implications for development and neurodegeneration. Prog Neurobiol. 1998 Aug;55(6):577-94. PubMed.

    . Connecting vesicle transport to the cytoskeleton. Curr Opin Cell Biol. 2000 Aug;12(4):503-8. PubMed.

    . Fast axonal transport misregulation and Alzheimer's disease. Neuromolecular Med. 2002;2(2):89-99. PubMed.

  2. We have made an antibody to an orphan GPCR that stains very strongly human neurons with NFT and human platelets. Anybody interested in collaborating?
    Mariano Alvira, MD

Make a Comment

To make a comment you must login or register.


News Citations

  1. AD Cell Cycle Reentry—Early Rather Than Late

Other Citations

  1. ARF live discussion

Further Reading

No Available Further Reading

Primary Papers

  1. . DNA synthesis and neuronal apoptosis caused by familial Alzheimer disease mutants of the amyloid precursor protein are mediated by the p21 activated kinase PAK3. J Neurosci. 2003 Jul 30;23(17):6914-27. PubMed.