In the January 31 online PNAS, Meir Scheinfeld, Shuji Matsuda, and Luciano D'Adamio of the Albert Einstein College of Medicine in Bronx, New York, show how the carboxyl terminus of AβPP could play a significant role in gene transcription, with implications for neurodegeneration in Alzheimer's disease.

The authors shift our attention to the "other" peptide-the AβPP intracellular domain (AICD)-which slips quietly away after γ-secretase cleaves AβPP, leaving Aβ in the harsh glare of the searchlight. The AICD peptide is less stable and more difficult to detect than its sibling, but recent investigations (beginning with Cao and Sudhof's work, ARF news story) have suggested it could play a role in transcriptional regulation (see also ARF news story; ARF news story; and ARF live discussion). There is even the suggestion from D'Adamio's group that AICD plays a role in apoptosis (Passer et al, 2000; see also related ARF discussion). In the present study, Scheinfeld and colleagues tracked the AICD peptide in its interactions with other proteins, especially the Janus kinase interacting protein (JIP-1), a scaffold protein that binds various elements of the Janus kinase (JNK) cascade. They also explore differences in the way AβPP and its relatives, AβPP-like protein-1 (APLP1) and APLP2, interact with other proteins to help regulate gene activity.

The researchers found that JIP-1 activates transcription in cultured cells transfected with the carboxyl terminus of AβPP. By contrast, other proteins (ShcA, ShcC, NCl) that bind AICD in the same region (the phosphotyrosine binding YENPTY motif) do not activate transcription with this interaction. In addition, the authors showed that individual domains of the protein were insufficient for this activity, and that full-length JIP-1 is required to activate transcription. Similarly, they used point mutations in AβPP to confirm that the gene activation can be traced to JIP-1 binding to AβPP. Scheinfeld and colleagues also confirm that AβPP must be cleaved by the PS-1/γ-secretase complex before this JIP-1-mediated transcription can occur. The activity is blocked by γ-secretase inhibitors and in cells where PS-1 activity is blocked.

Similar to JIP-1, the protein Fe65 interacts with AβPP to regulate transcription. It does this by moving into the nucleus together with AICD, and this process is regulated by the protein Tip60 (see also Rosenfeld et al., 2002). Importantly, Scheinfeld and colleagues report, JIP-1 is not translocated to the nucleus in its AICD-related transcriptional activities, nor is it dependent on Tip60.

Finally, the researchers determined that JIP-1-related transcriptional activation is something AβPP's cousins APLP-1 and APLP-2 (via their AβPP-like intracellular domains, or ALIDs) are not capable of. These data, write the authors, may help to explain why AβPP and the APLPs are apparently interchangeable for some functions (such as normal development), but not for others. "Considering that the AICD fragment is produced by processing, and its ability to induce gene activation with JIP-1 is not shared with the ALIDs, it is possible that the functional consequences of the AICD/JIP-1 interaction, including its transcription-modulating properties, may also be important in the pathology of AD," the authors conclude.—Hakon Heimer


Make a Comment

To make a comment you must login or register.

Comments on this content

  1. In this paper, Scheinfeld and colleagues from the D’Adamio laboratory extended their work on the interaction between JIP-1 and APP. JIP is JNK-interacting protein-1, which several groups, including the D’Adamio lab, last year showed to bind to the cytoplasmic tail of APP. Those labs showed that JIP-1 interacted with APP and that overexpression of JIP-1 altered APP processing and metabolism (principally dealing with APP phosphorylation and secretion and Aβ release).

    An area of APP biology that has taken center stage recently is the potential role in nuclear signal transduction. This idea has been too inviting by analogy to Notch signaling ever since γ-secretase was shown to cleave both APP and Notch, the latter to generate the nuclear signaling-competent NICD fragment. Evidence has been building in the last two years that the APP cognate of NICD, coined AID or AICD, indeed has signaling properties. First shown in a reporter system by Cao and Sudhof, this observation was confirmed by the finding of APP translocation into the nucleus, stabilizing of AID/AICD by Fe65, and recently, by showing the APP complex assembling on the KAI1 promoter using chromatin immunoprecipitation.

    The present study used a heterologous reporter system to show that the interaction of JIP-1 and APP activates gene expression. This signaling system is different from the APP/Fe65/Tip60 pathway for several reasons: Only JIP-1 and APP are required, not Fe65 or Tip60; the APLPs do not share this property; and there is no obvious translocation of JIP-1 into the nucleus. These new observations are interesting and intriguing. The fact that JIP-1 does not translocate into the nucleus is not too bothersome, as discussed at some length by the authors. Curiously, however, Tip60 was not only not required for the APP/Fe65 signaling, but it repressed activation of the reporter gene. This is in contrast to the Cao and Sudhof and the recent Baek and colleagues reports. Perhaps this is, as pointed out by the authors, only a technical difference. For example, Scheinfeld and colleagues previously showed that APLPs, like APP, activated reporter gene expression with Fe65, but this was apparently also independent of Tip60. Nonetheless, it is important to remember that all the studies to date required overexpression in cultured cells. Therefore, it is possible that these studies may have introduced artificial elements that may not be physiological.

    In sum, this interesting paper brings in another mechanism whereby APP may participate in signal transduction. The most interesting and pressing questions to address now are: Does APP signaling occur endogenously in vivo (whether constitutive or only upon activation)? If so, what are these downstream genes? How do the two pathways mediated by Fe65/Tip60 and JIP-1 differ, and when are they activated in vivo? Finally, do perturbations of these signaling pathways contribute to AD pathogenesis?


    . [Increased activity of stress-regulating systems in Alzheimer disease]. Tijdschr Gerontol Geriatr. 2001 Feb;32(1):17-23. PubMed.

  2. The observation that AICD and Fe65 are nuclear proteins (Minopoli et al., 2001; Kimberly et al., 2001; Cao & Sudhof, 2001; Gao & Pimplikar, 2001) and the similar proteolytic processing of APP and Notch have suggested the hypothesis that APP and its intracellular partners have some role in gene regulation. The paper of Scheinfeld et al. reports elegant data supporting this hypothesis and introducing new possible players in the scenario, namely JIP-1 and JNK. However, the exclusion of JIP-1 from the nucleus renders very unlikely the possibility that JIP-1 regulates transcription by interacting with gene promoters.

    I don't think that the extensive use of the experimental approach based on Gal4-dependent transcription of a reporter gene can give enlightening results. In fact, it became evident that a huge amount of protein sequences, often completely unrelated to gene transcription, are able to activate transcription when fused to the Gal4 DNA-binding domain. Therefore, this approach should be used to study transcription factors and not to demonstrate that a protein is a transcription factor.

    Schenfield and coworkers stress that only APP, and not APLP1 and APLP2, is involved in the JIP-1-based transcription regulation. I agree with the authors’ suggestion that this observation could be of interest in the pathogenesis of AD. However, considering that APP knockout mice are actually normal, it is expected that AID-JIP-mediated gene regulation, if any, concerns minor functions.

    In my opinion, the most important issue at this moment is to try to identify genes that are directly regulated by AICD, Fe65, JIP, and so on. There is one very solid result demonstrating that AICD, Fe65, and Tip60 are associated in vitro with the KAI-1 gene promoter (Baek et al., 2002). Another possible target of nuclear Fe65 is the thymidylate synthase gene (Bruni et al., 2002), although evidence of a direct interaction of Fe65 and/or of Fe65-AID complex with this promoter is still lacking.


    . The beta-amyloid precursor protein functions as a cytosolic anchoring site that prevents Fe65 nuclear translocation. J Biol Chem. 2001 Mar 2;276(9):6545-50. PubMed.

    . The intracellular domain of the beta-amyloid precursor protein is stabilized by Fe65 and translocates to the nucleus in a notch-like manner. J Biol Chem. 2001 Oct 26;276(43):40288-92. PubMed.

    . A transcriptionally [correction of transcriptively] active complex of APP with Fe65 and histone acetyltransferase Tip60. Science. 2001 Jul 6;293(5527):115-20. PubMed.

    . The gamma -secretase-cleaved C-terminal fragment of amyloid precursor protein mediates signaling to the nucleus. Proc Natl Acad Sci U S A. 2001 Dec 18;98(26):14979-84. PubMed.

    . Exchange of N-CoR corepressor and Tip60 coactivator complexes links gene expression by NF-kappaB and beta-amyloid precursor protein. Cell. 2002 Jul 12;110(1):55-67. PubMed.

    . Fe65, a ligand of the Alzheimer's beta-amyloid precursor protein, blocks cell cycle progression by down-regulating thymidylate synthase expression. J Biol Chem. 2002 Sep 20;277(38):35481-8. Epub 2002 Jun 27 PubMed.

  3. In a detailed and carefully performed study, Inomata and colleagues provide some intriguing new observations about what JIP-1b may be doing when it interacts with AβPP. Last year, several labs reported that AβPP interacts with JIP-1b, the JNK interacting scaffold protein (see Tare et al., 2002; Matsuda et al., 2002). It was never clear what this interaction meant. Now, the authors provide new data to indicate that this interaction facilitates JNK phosphorylation of AβPP, although, parenthetically, JNK does not require JIP-1b to phosphorylate AβPP. More interesting is the second observation in the study: The authors suggest that JIP-1b may be a required intermediate to bring AβPP together with kinesin light chain (KLC1). Goldstein and colleagues reported two years ago that AβPP directly interacts with kinesin light chain 1 (KLC1), and that AβPP may be the molecule that links certain axonal cargo vesicles to the kinesin machinery (see ARF related news story). This was a very attractive idea, as AβPP was known to undergo fast axonal transport. The problem is that other investigators have not replicated this finding. Therefore, Inomata and colleagues may have provided the critical piece of evidence showing that KLC1 and AβPP do come together, but only with JIP-1b as a scaffold. Finally, it has been shown that AβPP phosphorylation at Thr668 may be involved in neurite outgrowth. Thus, JIP-1b may play dual roles, not only bringing AβPP down the axon, but also facilitating its phosphorylation by JNK. Whether the latter is only in response to JNK signaling or related to what AβPP might normally do in axonal terminals is unclear.


News Citations

  1. Long-elusive Function for APP Cleavage Product Comes into View: It's Gene Transcription
  2. Stockholm: Educated Guessing Game: Which Genes Does “Other” APP Snippet Regulate?
  3. Stockholm: The Presenilin Signaling Hub—A RIP-off or the Real Deal?

Paper Citations

  1. . Generation of an apoptotic intracellular peptide by gamma-secretase cleavage of Alzheimer's amyloid beta protein precursor. J Alzheimers Dis. 2000 Nov;2(3-4):289-301. PubMed.
  2. . Exchange of N-CoR corepressor and Tip60 coactivator complexes links gene expression by NF-kappaB and beta-amyloid precursor protein. Cell. 2002 Jul 12;110(1):55-67. PubMed.

Other Citations

  1. ARF live discussion

Further Reading

No Available Further Reading

Primary Papers

  1. . JNK-interacting protein-1 promotes transcription of A beta protein precursor but not A beta precursor-like proteins, mechanistically different than Fe65. Proc Natl Acad Sci U S A. 2003 Feb 18;100(4):1729-34. PubMed.