Both amyloid-β precursor protein (APP) and apolipoprotein E (ApoE) are key players in the pathophysiology of Alzheimer disease (AD). But are the two directly connected? In tomorrow’s Neuron, researchers led by Guojun Bu at Washington University School of Medicine, St. Louis, report that APP plays a key role in suppressing lipoprotein receptor 1 (LRP1) expression. LRP1 is one of two major brain lipoprotein receptors; it contributes to ApoE uptake and intracellular degradation. “The finding extends the function of APP to brain lipoprotein metabolism and it also links the two major genetic determinants for early- and late-onset AD,” said Bu in an interview with ARF.

There are already numerous lines of evidence linking Aβ with ApoE and cholesterol (see ARF related news story). “The interesting thing about this new APP/ApoE connection is that it is completely independent of Aβ,” said Bu. His team discovered that it is the APP intracellular domain, or AICD, that suppresses LRP1 expression. As such, the research not only links APP and ApoE, but also proposes an answer to the contested question of what physiological role, if any, AICD fulfills.

Bu and colleagues made the AICD/LRP1 connection while trying to determine if APP processing is connected with ApoE/cholesterol metabolic pathways. The scientists found that in APP knockout and APP/APLP2 double knockout mice, brain lysates contained less ApoE and more cholesterol than in wild-type. The same was true in mouse embryonic fibroblasts derived from the same knockout strains, suggesting that APP and/or its homolog APLP2 play a role in lipid and lipoprotein regulation.

The researchers wondered if these changes might be due to faster intracellular degradation of ApoE. Since LRP1 and the low-density lipoprotein receptor (LDLR) are both involved in ApoE uptake into cells, the researchers next looked at whether knocking out APP or APLP2 altered lipoprotein receptor levels, as well. They found that while LDLR expression was unchanged, LRP1 levels were significantly higher in APP or APP/APLP2 double knockout (KO) mouse embryonic fibroblasts. There also seemed to be a dose response; deleting APP increased LRP1 about eightfold, deleting both genes about 16-fold.

How might APP suppress LRP1 expression? Given the various links between Aβ and ApoE/cholesterol metabolism, the amyloid peptide might seem a likely nexus, yet the researchers found that LRP1 levels stayed stable in BACE1 knockouts. This made the Aβ connection less likely, because BACE1 is the major catalyst for APP β-cleavage. Moreover, the researchers found that mouse embryonic fibroblast cells expressing APP with the Swedish mutation also had normal LRP1 levels despite increased Aβ production.

With Aβ out of contention, the researchers next looked to see if LRP1 suppression required γ-secretase activity. Indeed, presenilin 1/presenilin 2 double KO mouse embryonic fibroblasts had a ninefold increase in LRP1. This increase appears to be directly related to loss of γ-secretase activity, since it also showed up in nicastrin knockouts and in cells treated with a γ-secretase inhibitor.

AICD is one product of γ-secretase activity, and there is some evidence that the intracellular domain might regulate transcription upon binding to other factors such as Fe65 and Tip60 (see ARF related news story and ARF news story). To test if AICD might regulate LRP1 expression, the researchers transiently expressed AICD in U87 cells, where it reduced LRP1 by about 30 percent but did not affect LDLR. Fe65 alone also slightly reduced LRP1, and AICD and Fe65 expression combined cut LRP1 levels in half.

Together, this and further experiments suggest that APP processing, particularly production of AICD, influences lipoprotein and cholesterol metabolism. “I would not go as far as to say that AICD is a transcription factor, because our experiments were all cellular with many components present. But what this work does show is that the presence or absence of AICD makes a big difference to lipid metabolism,” said Bu.

How this relates to AD pathology is not clear, but links between APP and lipid metabolism seem to be growing by leaps and bounds. Researchers led by Tobias Hartmann, now at the University of the Saarland, Homburg, Germany, have also reported that presenilin knockout leads to increased cholesterol in mouse embryonic fibroblasts, though they attributed this finding to loss of Aβ’s proposed inhibitory effect on cholesterol synthesis (see ARF related news story). “It is possible that Aβ affects cholesterol synthesis while AICD affects its metabolism,” said Bu. That suggests that both the head and tail of APP are involved in cholesterol metabolism.

Whether AICD’s suppression of LRP might be a cause or effect of AD pathology is uncertain, as well. Given that mature neurons cannot make their own cholesterol—a necessary component of synapses—and typically import it via LRP1, the authors speculate that loss of LRP1 may contribute to synaptic dysfunction. There is some independent evidence that LRP1 is reduced in AD (see Kang et al., 2000). Bu hopes that his latest work will generate interest in APP beyond Aβ and spur research into the role of synaptic cholesterol in the aging brain and AD.

On the broader lipid research front, European researchers recently concluded that the ApoE4 genotype, besides being the strongest genetic risk factor for late-onset AD, also puts people at risk for coronary heart disease, albeit less strongly. In the September 19 Lancet, John Danesh, University of Cambridge, England, and colleagues describe their meta-analysis of 82 lipid level studies and 121 coronary outcome studies. It found that people carrying two ApoE4 alleles face a slightly higher risk of coronary disease (odds ratio 1.06), while E2/E2 carriers have about a 20 percent reduced risk compared to the E3/E3 genotype. Previous reviews of ApoE and coronary disease are susceptible to bias because they were dominated by small studies. To overcome this weakness, first author Anna Bennet and colleagues focused on lipid studies that had at least 1,000 participants and coronary studies that had at least 500 participants. The analysis further predicts a linear relationship between ApoE genotype and serum low density lipoprotein cholesterol (LDL-C) —the bad kind. People with E2/E2 genotype had about 30 percent less serum LDL-C than those homozygous for ApoE4. The difference is about as much as statin medication achieves, write the authors.—Tom Fagan


  1. The study by Liu et al. shines a new light on the connection between cholesterol metabolism and APP processing. However, at this time the story is reversed—now it appears that APP proteolytic fragments are regulating cholesterol levels in brain and inside the cells. In a series of elegant experiments, the authors describe a mechanism by which the C-terminal fragment of APP, named AICD, modulates brain ApoE and cholesterol metabolism by directly regulating the expression and function of the lipoprotein receptor LRP1. Knocking out APP/APLP2 or components of the γ-secretase complex significantly affected the expression of LRP1, as well as ApoE and intracellular cholesterol levels. These alterations were partially restored by forced expression of AICD. Finally, Liu et al. provide evidence that deletion of LRP1 in forebrain neurons of adult mice significantly increased ApoE levels, while cholesterol levels were conversely decreased.

    What makes this study powerful is that, to confirm their in-vitro findings, the authors have used numerous mutated cell lines, including APP/APPL2-DKO and PS1/PS2-DKO primary cells, as well as several lines of knockout mice. As the authors suggest, a possible overall explanation is that the altered expression of LRP significantly affects the catabolism of ApoE, eventually affecting intracellular cholesterol concentration. It is well known that the neuronal uptake of ApoE particles, and possibly ApoE uptake by other cells in brain, is mediated by lipoprotein receptors of the LDLR family (1). Interestingly, in other studies, brain cholesterol levels were not affected in mice with targeted disruption of genes directly related to cholesterol transport and metabolism such as Abca1, LDRL, and, most importantly, ApoE (2-5). In this connection, it would be worth knowing what effect AICD has on the level and lipidation of ApoE-containing lipoproteins in CSF and lipoproteins secreted in the conditioned media of astrocytes isolated from APP/APPL2-DKO or LRP1 KO mice.

    Some questions require further investigation. Because LRP1 and ApoE affect Aβ clearance (6-8), it is important to reveal how the lack of LRP1 in LRP1 knockout mice or the suppression of LRP1 expression by γ-secretase inhibitors would affect this process. Most importantly, and relevant to AD, it is interesting to see how the familial mutations of presenilin that increase AICD would affect LRP1 expression and ApoE/cholesterol metabolism.


    . Lipoprotein receptors in the nervous system. Annu Rev Biochem. 2002;71:405-34. PubMed.

    . Novel role for apolipoprotein E in the central nervous system. Modulation of sulfatide content. J Biol Chem. 2003 Mar 7;278(10):8043-51. PubMed.

    . ABCA1 is required for normal central nervous system ApoE levels and for lipidation of astrocyte-secreted apoE. J Biol Chem. 2004 Sep 24;279(39):40987-93. PubMed.

    . Elevated plasma cholesterol does not affect brain Abeta in mice lacking the low-density lipoprotein receptor. J Neurochem. 2007 Aug;102(4):1220-31. PubMed.

    . The low density lipoprotein receptor regulates the level of central nervous system human and murine apolipoprotein E but does not modify amyloid plaque pathology in PDAPP mice. J Biol Chem. 2005 Jul 8;280(27):25754-9. PubMed.

    . Apolipoprotein E promotes astrocyte colocalization and degradation of deposited amyloid-beta peptides. Nat Med. 2004 Jul;10(7):719-26. PubMed.

    . Clearance of amyloid-beta by circulating lipoprotein receptors. Nat Med. 2007 Sep;13(9):1029-31. PubMed.

    . Clearance of Alzheimer's amyloid-ss(1-40) peptide from brain by LDL receptor-related protein-1 at the blood-brain barrier. J Clin Invest. 2000 Dec;106(12):1489-99. PubMed.

    View all comments by Radosveta Koldamova
  2. AICD Rules!
    Coming on the heels of the report that AICD represses transcription of EGF-receptor (see Zhang et al., 2007), this paper now demonstrates that AICD also represses transcription of LRP1. The role of AICD as a transcriptional regulator had its share of skeptics (Hébert et al., 2006), but together the two studies convincingly support the role of AICD as it was conceived originally (Cao and Sudhof, 2001; Gao and Pimplikar, 2001). In the present studies, Liu et al. used multiple approaches to show that AICD indeed modulates LRP1 expression at the transcriptional level.

    The significance of these observations in terms of AD pathogenesis is unclear and needs future studies. Nonetheless, it is becoming clear that AICD, long neglected in favor of its Aβ cousin, is finally commanding attention, and respect, in its own right. The amyloid hypothesis tries to explain, not always satisfactorily, the origin of the AD pathology in terms of Aβ. While it is premature to claim that AICD plays a role in AD pathogenesis, the observations of Liu et al. present a linkage between APP processing and ApoE/cholesterol metabolism that does not go through Aβ but through AICD. Indeed, AICD alone, without Aβ, affects other parameters that feature prominently in AD. AICD activates GSK3β without involving Aβ (Ryan and Pimplikar, 2005), and a recent study from Karen Ashe's group (Ma et al., 2007) implicated AICD in enhanced memory and synaptic plasticity. Finally, the D664A mouse created by Dale Bredesen and colleagues shows that APP-CTF is pivotal in AD pathogenesis (Galvan et al., 2006; Saganich et al., 2006). Although these data were interpreted in terms of inhibition of cleavage of APP-CTF by caspases, it is quite likely that D664A-AICD is simply “dysfunctional.” In any case, it is no longer prudent—or possible—to ignore AICD when talking about APP function or AD pathology.

    Although Liu et al. show a fairly strong connection between APP and LRP1 expression through AICD, the link between APP-LRP1 and altered cholesterol metabolism may be a red herring. Liu et al. find that AICD does not affect LDLR1 expression, so the effects on cholesterol uptake and catabolism by neurons are likely to be less pronounced. LRP1 is thought to mediate intracellular signaling kinases through ApoE binding or APP interaction (Rebeck et al., 2006). Moreover, several studies have shown LRP1 to modulate APP levels and APP-processing. Thus, it is not a given that APP-LRP1 interplay involves cholesterol homeostasis.


    . Presenilin/gamma-secretase-dependent processing of beta-amyloid precursor protein regulates EGF receptor expression. Proc Natl Acad Sci U S A. 2007 Jun 19;104(25):10613-8. PubMed.

    . Regulated intramembrane proteolysis of amyloid precursor protein and regulation of expression of putative target genes. EMBO Rep. 2006 Jul;7(7):739-45. 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.

    . Activation of GSK-3 and phosphorylation of CRMP2 in transgenic mice expressing APP intracellular domain. J Cell Biol. 2005 Oct 24;171(2):327-35. PubMed.

    . Involvement of beta-site APP cleaving enzyme 1 (BACE1) in amyloid precursor protein-mediated enhancement of memory and activity-dependent synaptic plasticity. Proc Natl Acad Sci U S A. 2007 May 8;104(19):8167-72. PubMed.

    . Reversal of Alzheimer's-like pathology and behavior in human APP transgenic mice by mutation of Asp664. Proc Natl Acad Sci U S A. 2006 May 2;103(18):7130-5. PubMed.

    . Deficits in synaptic transmission and learning in amyloid precursor protein (APP) transgenic mice require C-terminal cleavage of APP. J Neurosci. 2006 Dec 27;26(52):13428-36. PubMed.

    . The generation and function of soluble apoE receptors in the CNS. Mol Neurodegener. 2006;1:15. PubMed.

    View all comments by Sanjay Pimplikar
  3. The work by Bu et al. significantly—and elegantly—extends our understanding of the role of APP metabolism in lipid regulation. Their experiments clearly show that γ-secretase-mediated processing of APP regulates LRP1 transcription and hence one essential aspect of cellular cholesterol (and general lipid) homeostasis. AICD expression in APP or PS knockouts nicely restored LRP1 regulation and partly rescued the damaged cellular lipid homeostasis, thus identifying a role for AICD in LRP1 transcription control. Point mutations in AICD further support this interpretation.

    It is tempting to speculate that Aβ40 (as downregulator of the cholesterol synthesis enzyme HMGR) together with AICD (as a regulator of cholesterol uptake via LRP1) might be able to rescue cellular cholesterol levels fully. However, lipid homeostasis is a fairly complex game, and more players might be involved. Interestingly, Bu and colleagues report that PS1, but not PS2, is involved in LRP1 regulation. For sphingomyelin metabolism (a function of Aβ42), this is different. Here PS2 is able to functionally replace PS1.

    It should be kept in mind that several type 1 membrane receptors contain actual or putative γ-secretase cleavage domains. This complicates the potential mechanism as much as it causes difficulties in the interpretation of the data. LRP is but one such example. Bu et al. also tested different AICD constructs and found no functional differences for these constructs. This, too, is remarkable, because for Aβ peptides the C-terminus is essential to determine whether the peptide’s activity is directed toward cholesterol or sphingomyelin.

    For those familiar with lipid homeostasis, all of this is déjà vu. It very closely resembles another system for cholesterol homeostasis—the SREBPs. SREBP processing and signaling is very complex and still far from being completely understood. It therefore does not require the art of prophecy to predict that it will take many more publications until the role of APP processing and lipid homeostasis is fully understood. One can only hope that these will be of the same high quality as the work presented here by Bu and colleagues.

    View all comments by Tobias Hartmann

Make a Comment

To make a comment you must login or register.


News Citations

  1. ApoE—Breaking Ties to Aβ Offers Potential Therapy
  2. Long-elusive Function for APP Cleavage Product Comes into View: It's Gene Transcription
  3. AICD—Salute to Salutary Effects
  4. A Better GRIP on the Aβ-Lipid Connection

Paper Citations

  1. . Modulation of amyloid beta-protein clearance and Alzheimer's disease susceptibility by the LDL receptor-related protein pathway. J Clin Invest. 2000 Nov;106(9):1159-66. PubMed.

Further Reading

Primary Papers

  1. . Association of apolipoprotein E genotypes with lipid levels and coronary risk. JAMA. 2007 Sep 19;298(11):1300-11. PubMed.
  2. . Amyloid precursor protein regulates brain apolipoprotein E and cholesterol metabolism through lipoprotein receptor LRP1. Neuron. 2007 Oct 4;56(1):66-78. PubMed.