The amyloid precursor protein (APP) travels widely in neurons. After synthesis in the ER/Golgi, the protein makes its way to the cell surface and back, and out to axon terminals. Along the way, it meets up with the β- and γ-secretases and gets cleaved into a number of fragments, including Aβ and others with a plethora of proposed functions. Full-length APP has been suggested to be an adaptor protein that links vesicles to the kinesin motors that power microtubule-mediated transport out to axons (see ARF related news story), but a new study may be cause to revise that model. In a report out in the March 18 issue of Journal of Neuroscience, Zoia and Virgil Muresan at the University of Medicine and Dentistry of New Jersey in Newark, with Nicholas Varvel and Bruce Lamb at Case Western Reserve University in Cleveland, Ohio, look at the distribution of different APP epitopes in cultured neurons and in mouse brain in situ using specific antibodies. They conclude that cleavage of APP occurs for the most part before the protein reaches axons, and that the different fragments segregate into separate transport vesicles and are delivered to distinct destinations in the axon terminal. Only when APP is overexpressed in cultured neurons do many vesicles appear to contain much full-length APP. The results suggest that in normal conditions, APP cleavage occurs early and often, and that its fragments have their own functional roles in axons.

A second report on APP transit, from Huaxi Xu and colleagues at the Burnham Institute in La Jolla, California, looks at the role of APP in trafficking of the γ-secretase to the cell surface. The γ-secretase subunit presenilin-1 regulates APP trafficking, and the data, published online March 10 in the Journal of Biological Chemistry, suggest that the reciprocal is also true. Cells that lack APP show altered accumulation of presenilin-1 on the surface, which results in changes in the processing of another substrate, the Notch protein. Taken together, the two studies reveal more details of the complex interactions between APP’s movements, its processing, and the potential roles of the derived fragments.

To get an accurate picture of the location and timing of APP cleavage in cells, Virgil Muresan and coworkers immunostained endogenous APP in mouse neurons in culture and in situ with antibodies to three different epitopes. They compared staining patterns of the N-terminal antibody 22C11, the C-terminal antibody 2452, and 4G8, which recognizes the central (Aβ) domain. In cultured neurons, they found most of the APP C-terminal reactivity in the cell body, with lesser amounts in neurites and nerve terminals, while the N-terminal reactivity was evenly spread throughout the three regions. Aβ was mostly in the cell body and neurites, with little in terminals. Surprisingly, double labeling showed that the N- and C-terminal labels rarely appeared in the same place, suggesting that the cells contained relatively little full-length APP. They got the same results in primary neurons and under several conditions of cell fixation.

Taking a closer look, the investigators found that the three antibodies labeled distinct sets of transport vesicles along neurites. They also found many vesicles that carried a phospho-C-terminal APP epitope, but not the N-terminal epitope, suggesting that many carried phospho-CTFs (C-terminal fragments) or phospho-AICD (APP intracellular domain), rather than full-length phospho-APP. Separation of the different fragments was also seen in growth cones in the neurites of the cultured cells, with the pAPP antibody detected in the peripheral zones while the N-terminal antibody picked up vesicles in the central part of the structure. Only pAPP C-terminal fragments were detected in lamellipodium and fillipodium of growth cones and were concentrated in regions of advancement and turning, a finding which may indicate a role for the pCTFs in regulating growth cone mobility and neurite extension.

In brain tissue, the researchers found a similar segregated pattern of epitope staining when they used tissue from a transgenic mouse expressing low levels of human (Swedish mutant) APP. In cells expressing higher levels of APP, the epitopes were more often seen together, with 93 percent of vesicles appearing to carry full-length APP, compared to less than a quarter when only endogenously expressed APP was present. The authors conclude that the processing and transport of APP into neurites depends on the level of expression of APP.

“Overall, our results overturn the long-held view according to which transport of APP within neurites occurs, under normal conditions, mostly as full-length protein,” the authors write. All the same, Muresan pointed out in a e-mail to ARF that the CTFs should be capable of carrying out the same function of recruitment of kinesin-1 to vesicles as full-length APP, as they are still transmembrane proteins. “Moreover, even the AICD (i.e., C-γ, the product of γ-secretase cleavage) may not fall off the vesicle, and could thus still carry kinesin-1,” he wrote. “We (and others) have found that the AICD, when expressed in neurons, becomes associated with what appear to be vesicles. We have mentioned in the Discussion section of our paper that this association may involve the positively charged lysines in the AICD, which could easily interact with membrane acidic phospholipids. It may very well be that the release of the AICD from the vesicle is a regulated event, and does not just occur automatically after γ-secretase cleavage. Of course, this is only speculation at this time.”

How could the early cleavage of APP bear on the events that take place in Alzheimer disease? Muresan says, “From a conceptual point of view, our data suggest that the different APP fragments have a life of their own, independent of full-length APP, and that they are transported to different locations within the neuron, where they perform independent functions. If for some reason the processing of APP or the transport of the processed fragments is altered—which may happen in AD conditions—the normal function of these fragments may be perturbed. From this point of view, it is the alteration of the normal function of APP that may somehow trigger the neuronal pathology. More and more studies suggest that AD may start with a perturbation of the synapse function. Our paper suggests that some of the APP fragments (e.g., the phosphorylated CTFs) may indeed play crucial roles at the neurite terminals.”

The study also suggests that Aβ is present normally within neurons, where it could have deleterious effects on axonal transport (see ARF related news story). “Our results allow for the interpretation that the Aβ that is normally produced inside neurons may somehow form oligomers under still unknown conditions relevant to AD,” Muresan said.

The processing of APP falls to enzymes, the secretases, whose activity also depends on their location in cells. The γ-secretase controls APP trafficking, and the paper from Xu and colleagues shows that this relationship goes both ways. First authors Yun Liu and Yun-Wu Zhang find cells deficient in APP show a faster transit of presenilin-1 (PS1) from the trans-Golgi network to the cell surface, where it accumulates along with other components of the γ-secretase complex. This results in more active cleavage of the cell surface substrate, Notch. Restoring APP brings presenilin and other γ-secretase proteins back to normal levels. The researchers conclude that interaction between APP and PS1 may be required for the proper retention of APP/PS1 in the trans-Golgi network and subsequent delivery to the cell surface. The researchers also find that phospholipase D1 promotes cell surface accumulation of presenilin-1 independent of the effects of APP. Neither APP nor PLD1 affected trafficking of the β-secretase.—Pat McCaffrey


  1. It has been well suggested that APP is processed during its intracellular trafficking to generate APP CTFs, Aβ, and the APP intracellular domain (AICD). However, how these APP derivatives are transported intracellularly is much less known. In this paper by Zoia Muresan and colleagues, the authors utilized various antibodies against different APP domains for immunocytochemistry and found that full-length APP and APP derivatives are sorted into distinct vesicles and transported independently, with APP CTFs preferentially entering the lamellipodium and filopodia of growth cones and becoming concentrated in regions of growth cone turning and advancement.

    In some experiments, the authors used antibody 22C11 for detecting the extracellular fragment of APP and antibody 4G8 for Aβ. Since 22C11 cross-reacts with other APP family proteins (APLP1 and APLP2) while 4G8 only sees APP (and Aβ), the comparisons for the localizations of full-length APP and its derivatives (especially Aβ) may not be appropriate. Nevertheless, these results are very interesting and suggest that a large amount of APP can be cleaved before it is sorted into axonal transport vesicles, probably at the ER and the Golgi/TGN, as we have reported before (Xu et al., 1997 and Greenfield et al., 1999). Moreover, the results indicate that full-length APP and its derivatives may be transported to different locations and exert distinct functions. Several studies from Larry Goldstein and William Mobley’s labs have already suggested that APP plays an active role in axonal transport. Our very recent study also found that both full-length APP and APP bCTF (C99) can regulate intracellular trafficking of PS1/β-secretase components for their cell surface delivery (the results, Liu et al., 2009, are also commented on in this Research News). However, whether or not the intracellular trafficking of proteins other than PS1/γ-secretase components may be regulated by APP awaits further examination. Further, whether the distinct sorting paths or localizations of APP and its derivatives may differentially direct the transport of their respective “cargos,” especially at growth cones also deserves careful scrutiny.


    . Generation of Alzheimer beta-amyloid protein in the trans-Golgi network in the apparent absence of vesicle formation. Proc Natl Acad Sci U S A. 1997 Apr 15;94(8):3748-52. PubMed.

    . Endoplasmic reticulum and trans-Golgi network generate distinct populations of Alzheimer beta-amyloid peptides. Proc Natl Acad Sci U S A. 1999 Jan 19;96(2):742-7. PubMed.

    . Intracellular trafficking of presenilin 1 is regulated by beta-amyloid precursor protein and phospholipase D1. J Biol Chem. 2009 May 1;284(18):12145-52. PubMed.

  2. We read with great interest the paper by Liu et al. (1), and would like to comment on their exciting findings. The paper proposes a novel mechanism by which the amyloid-β precursor protein (APP) could regulate the intracellular transport of a select group of proteins with emphasis on those that form the γ-secretase complex.

    APP was previously proposed to regulate the intraneuronal transport by functioning as a receptor for the microtubule motor kinesin-1. However, it is still largely debated to what extent this model is relevant in vivo, and whether the interaction between APP and kinesin-1 is direct or mediated by bridging protein(s), such as JIP-1 (cJun NH2-terminal kinase-interacting protein-1). Related questions are now addressed by two papers: Liu et al. (1), and our paper, Muresan et al. (2), which are the subject of this Research News. Both articles show that the transport of APP, and the role of APP in regulating transport of other cargo proteins, are far more complex than previously anticipated.

    Liu et al. (1) propose that APP could modulate the delivery of the γ-secretase complex to the cell surface by regulating its exit from the trans-Golgi network (TGN). A possible mechanism is that APP interacts with the γ-secretase complex and prevents it from being recruited into cargo vesicles, at the TGN. A direct consequence of this mechanism is that APP regulates in this way the cleavage by γ-secretase of substrates localized to the cell surface, such as Notch. Implicitly, Notch signaling is regulated by APP in this way. Liu et al. (1) also show that the trafficking of γ-secretase to the cell surface is regulated not only by APP, but also by the activity of phospholipase D1 (PLD1), a lipid-modifying enzyme that converts phosphatidylcholine to phosphatidic acid. Intriguingly, PLD1 also regulates the transport of APP in a presenilin-1 (PS1, a component of the γ-secretase complex) independent manner. This is important, because several groups reported that PS1 regulates the axonal transport by mechanisms that are not fully understood. Although more work needs to be done to elucidate the complex regulation of the transport of APP and its processing machinery, Liu et al. (1) bring an exciting contribution into this picture.

    As perceived from the articles of Liu et al. (1) and of Muresan et al. (2), full-length APP is largely restricted to the intracellular compartments of the early secretory pathway. Biochemical data recently obtained with mouse brain (3) certainly support this scenario. In spite of this, a fraction of full-length APP does reach the plasma membrane, and full-length APP may also have functions at the plasma membrane. However, a significant fraction of it is cleaved prior to entering the cargo vesicles that transport APP (or, rather, its fragments) to various intracellular destinations, as shown by us (2).

    We would like to comment on the specificity of the antibody 22C11 (4), an issue raised by Drs. Xu and Zhang in their comment to our paper. This antibody, largely used to detect full-length APP and the soluble N-terminal fragments (sAPPs), was used in some of our immunolabeling experiments due to its high sensitivity. We are aware that this antibody may cross-react with the amyloid precursor-like proteins (APLP1 and APLP2), although with lower affinity. To circumvent this problem, along with 22C11, we employed a plethora of antibodies recognizing epitopes from various regions of APP polypeptide, including antibodies that do not cross-react with APLP1 or APLP2, such as Alz90 (Roche; recognizing residues 511-608 of APP, poorly conserved in APLPs).

    With regard to antibody 4G8, which also recognizes the full-length APP, in addition to cleaved fragments, in our study we included antibodies against the cleaved C-terminal ends of Aβ40 and Aβ42, which selectively detect the Aβ fragments. Although our study was not aimed at the precise identification of the APP-derived polypeptides that are transported within neuritis, our results indicate that the transport of APP within neurites occurs to a large extent as cleaved fragments generated by proteolytic processing of APP in the cell body. Thus, the most important idea that derives from our study is that each APP fragment has a life of its own that begins early in the secretory pathway, and may thus have functions that are largely independent from that of full-length APP. Thus, we think that APP is indeed an “All Purpose Protein” (a term that we first heard from Dr. Sangram Sisodia, during a seminar talk) which functions as full-length protein and also as cleaved fragments.


    . Intracellular trafficking of presenilin 1 is regulated by beta-amyloid precursor protein and phospholipase D1. J Biol Chem. 2009 May 1;284(18):12145-52. PubMed.

    . The cleavage products of amyloid-beta precursor protein are sorted to distinct carrier vesicles that are independently transported within neurites. J Neurosci. 2009 Mar 18;29(11):3565-78. PubMed.

    . The in vivo brain interactome of the amyloid precursor protein. Mol Cell Proteomics. 2008 Jan;7(1):15-34. PubMed.

    . Amyloid-like properties of peptides flanking the epitope of amyloid precursor protein-specific monoclonal antibody 22C11. J Biol Chem. 1993 Dec 15;268(35):26571-7. PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. The Skinny on FAT: APP’s Role in Fast Axonal Transport
  2. The Many Misdeeds of Aβ—Seizures and Axonal Transport Interference

Further Reading

Primary Papers

  1. . The cleavage products of amyloid-beta precursor protein are sorted to distinct carrier vesicles that are independently transported within neurites. J Neurosci. 2009 Mar 18;29(11):3565-78. PubMed.
  2. . Intracellular trafficking of presenilin 1 is regulated by beta-amyloid precursor protein and phospholipase D1. J Biol Chem. 2009 May 1;284(18):12145-52. PubMed.