In this interesting and important paper, Castellano and colleagues provide clear, quantitative evidence for the effect of the different human ApoE isoforms on Aβ clearance in the brains of humanized ApoE knock-in mice carrying also the PDAPP transgene. Using their signature microdialysis assay, the authors show convincingly that the half-life of Aβ in the interstitial fluid of ApoE4 mice is significantly greater than in the E3 or E2 knock-ins. Half-life of Aβ in the different strains also correlates well with Aβ deposition in aged mice. By contrast, Aβ production was identical and independent of ApoE isoforms, demonstrating that ApoE isoforms primarily affect turnover and deposition and not Aβ generation, at least not in the PDAPP mouse model in which APP is unphysiologically overexpressed.

It remains possible that this overexpression occludes a component of physiological Aβ production that might be regulated by synaptic activity (Kamenetz et al., 2003; Bero et al., 2011) and ApoE isoforms (Chen et al., 2010), and thus could be relevant for human disease. Nevertheless, the...  Read more

In this interesting and important paper, Castellano and colleagues provide clear, quantitative evidence for the effect of the different human ApoE isoforms on Aβ clearance in the brains of humanized ApoE knock-in mice carrying also the PDAPP transgene. Using their signature microdialysis assay, the authors show convincingly that the half-life of Aβ in the interstitial fluid of ApoE4 mice is significantly greater than in the E3 or E2 knock-ins. Half-life of Aβ in the different strains also correlates well with Aβ deposition in aged mice. By contrast, Aβ production was identical and independent of ApoE isoforms, demonstrating that ApoE isoforms primarily affect turnover and deposition and not Aβ generation, at least not in the PDAPP mouse model in which APP is unphysiologically overexpressed.

It remains possible that this overexpression occludes a component of physiological Aβ production that might be regulated by synaptic activity (Kamenetz et al., 2003; Bero et al., 2011) and ApoE isoforms (Chen et al., 2010), and thus could be relevant for human disease. Nevertheless, the quantitative data on Aβ turnover this study provides represents a milestone in our understanding of the molecular basis by which ApoE4 so powerfully affects the age of onset of Alzheimer’s disease.

References:
Bero AW, Yan P, Roh JH, Cirrito JR, Stewart FR, Raichle ME, Lee JM, Holtzman DM. Neuronal activity regulates the regional vulnerability to amyloid-β deposition. Nat Neurosci. 2011 Jun;14(6):750-6. Abstract

Chen Y, Durakoglugil MS, Xian X, Herz J. ApoE4 reduces glutamate receptor function and synaptic plasticity by selectively impairing ApoE receptor recycling. Proc Natl Acad Sci U S A. 2010 Jun 29;107(26):12011-6. Abstract

Kamenetz F, Tomita T, Hsieh H, Seabrook G, Borchelt D, Iwatsubo T, Sisodia S, Malinow R. APP processing and synaptic function. Neuron. 2003 Mar 27;37(6):925-37. Abstract

View all comments by Joachim Herz

This study further supports our finding, published last year, showing ApoE genotype and plasma protein concentrations are associated with brain amyloid burden. Previously using PIB-PET, we have demonstrated that ApoE4 carriers show widespread increase in Aβ deposition in cognitively normal individuals. Furthermore, we were able to show a significant relationship between plasma ApoE protein concentration and Aβ deposition in the brain. This indicates that there may be a peripheral signature.

Since then, Vuletic et al. (2008) showed a strong association between ApoE protein concentration in CSF and measures of both APP and tau metabolism in cognitively normal individuals. Together, these studies suggest intrinsic roles for ApoE gene and plasma ApoE protein in the brain amyloid cascade in older individuals at risk for AD.

References:
Thambisetty et al. Proteome-based plasma markers of brain amyloid-β deposition in non-demented older individuals. J Alzheimers Dis. 2010;22(4):1099-109. Abstract

Vuletic et al. Apolipoprotein E highly correlates with AbetaPP- and tau-related markers in human cerebrospinal fluid. J Alzheimers Dis. 2008 Nov;15(3):409-17. Abstract

View all comments by Abdul Hye
View all comments by Joanna Riddoch-Contreras

This paper provides really compelling data that one role of ApoE in AD pathogenesis is by promoting Aβ clearance. One strength of the work is that the authors use in-vivo systems to assess what is happening to the Aβ produced in the brain. The conclusion that ApoE4 mice are less competent at clearing Aβ from the CSF provides a clear model for how ApoE genotype affects risk of AD.

Like all good studies, these data also raise some interesting questions. Is the clearance of Aβ from the CSF to the blood, or into cells? Are there specific receptors that mediate this process? What other molecules besides Aβ are impaired in clearance from ApoE4 brains?

This study greatly helps our understanding of the ApoE genetic risk, and will help define the direction of future studies on other risk factors for AD.

View all comments by G. William Rebeck

The recent publication by Castellano et al. provides direct in-vivo evidence for the long-held hypothesis that ApoE mediates Aβ clearance in an isoform-specific manner.

Interestingly, the authors demonstrate that ApoE has no effect on Aβ production in PDAPP/TRE mice. However, Aβ transgenic mouse models overproduce Aβ, which could mask any ApoE-specific differences in Aβ production. Future studies could address whether ApoE isoforms affect Aβ production in non-demented and in late-onset AD patients, without FAD mutations.

A further interesting observation in the study is that the Aβ clearance rates are remarkably similar for young and old animals. These data are especially evident in PDAPP/ApoE4 mice, despite extensive Aβ deposition in older mice. This raises the questions: Would it be expected for Aβ clearance rates to change with age, as Aβ accumulates, and is ApoE-mediated Aβ clearance a major pathway for AD development?

There are some technical considerations that could be further clarified. Aβ clearance rates are calculated after treatment with a γ-secretase...  Read more

The recent publication by Castellano et al. provides direct in-vivo evidence for the long-held hypothesis that ApoE mediates Aβ clearance in an isoform-specific manner.

Interestingly, the authors demonstrate that ApoE has no effect on Aβ production in PDAPP/TRE mice. However, Aβ transgenic mouse models overproduce Aβ, which could mask any ApoE-specific differences in Aβ production. Future studies could address whether ApoE isoforms affect Aβ production in non-demented and in late-onset AD patients, without FAD mutations.

A further interesting observation in the study is that the Aβ clearance rates are remarkably similar for young and old animals. These data are especially evident in PDAPP/ApoE4 mice, despite extensive Aβ deposition in older mice. This raises the questions: Would it be expected for Aβ clearance rates to change with age, as Aβ accumulates, and is ApoE-mediated Aβ clearance a major pathway for AD development?

There are some technical considerations that could be further clarified. Aβ clearance rates are calculated after treatment with a γ-secretase inhibitor. However, the pharmacokinetics (PKs) of the compound will greatly impact on the calculated Aβ- t1/2, and it would be optimal to know if there are any ApoE-isoform differences in plasma, ISF, or CSF drug PKs.

View all comments by Leon Tai
View all comments by Chunjiang Yu

This important paper by the Holtzman group establishes the relationship between ApoE4 genotype and decreased clearance of the Alzheimer’s disease (AD) toxin amyloid-β peptide (Aβ) from brain, both in humans and animal models. The idea that ApoE4 diminishes elimination of Aβ from the brain is not necessarily novel, per se, and has been previously suggested by some experimental studies from the Holtzman and other groups. However, it has never been proven so convincingly and in a such complete way in humans and animal models of AD as Castellano et al. have done in their present study. The authors should be congratulated for their approach.

The translational impact of this paper to the field carries, in my opinion, true, well-needed quality to help us reveal the mystery of how ApoE4 accelerates the development of AD. Findings showing that individuals with ApoE4/ApoE4 genotype have substantially reduced levels of Aβ in the CSF compared to other ApoE genotypes are important and fascinating. Genetic modeling in mice expressing different human ApoE genotypes and human Aβ precursor...  Read more

This important paper by the Holtzman group establishes the relationship between ApoE4 genotype and decreased clearance of the Alzheimer’s disease (AD) toxin amyloid-β peptide (Aβ) from brain, both in humans and animal models. The idea that ApoE4 diminishes elimination of Aβ from the brain is not necessarily novel, per se, and has been previously suggested by some experimental studies from the Holtzman and other groups. However, it has never been proven so convincingly and in a such complete way in humans and animal models of AD as Castellano et al. have done in their present study. The authors should be congratulated for their approach.

The translational impact of this paper to the field carries, in my opinion, true, well-needed quality to help us reveal the mystery of how ApoE4 accelerates the development of AD. Findings showing that individuals with ApoE4/ApoE4 genotype have substantially reduced levels of Aβ in the CSF compared to other ApoE genotypes are important and fascinating. Genetic modeling in mice expressing different human ApoE genotypes and human Aβ precursor protein (APP) revealed substantially elevated levels of human Aβ in the mouse brain interstitial fluid only in the presence of ApoE4 compared to ApoE2 and ApoE3, convincingly demonstrating how the ApoE4 allele accelerates the development of Aβ/amyloid pathology by diminishing Aβ clearance but not affecting its production in brain.

However, the question still persists in the field as to whether ApoE4 affects, in any other major way, the nervous system and cerebrovascular function independently and/or in parallel with the demonstrated Aβ/amyloid accumulations. More research is needed to address this question.

View all comments by Berislav Zlokovic

The important early observation that PDAPP mice hemizygous for murine ApoE (PDAPP x mouse ApoE+/-) had less Aβ and amyloid burden than those with two copies (PDAPP x mouse ApoE+/+) (Bales et al., 1997; Bales et al., 1999) is unfortunately seldom mentioned in more recent publications. In our own studies, we also noticed that PDAPP x ApoE+/- mice tended to be intermediate to PDAPP x ApoE+/+ and PDAPP x ApoE-/- in terms of Aβ and amyloid deposition (Nilsson et al., 2004). Moreover, lower Aβ deposition was seen in an independent transgenic model (Tg2576; APP-Swe) when hemizygous for murine ApoE (Holtzman et al., 2000).

If ApoE mainly serves to clear Aβ, as suggested by Castellano and coauthors, wouldn't one then expect aged PDAPP x mouse ApoE+/- to have more extensive Aβ deposition? After all, the concentration of ApoE, which is supposed to help clear Aβ, was 50 percent lower in the brains of PDAPP x mouse ApoE+/- than those of PDAPP x mouse ApoE+/+. But the experimental findings were actually the opposite.

Here, in an impressive set of experiments, Castellano et al....  Read more

The important early observation that PDAPP mice hemizygous for murine ApoE (PDAPP x mouse ApoE+/-) had less Aβ and amyloid burden than those with two copies (PDAPP x mouse ApoE+/+) (Bales et al., 1997; Bales et al., 1999) is unfortunately seldom mentioned in more recent publications. In our own studies, we also noticed that PDAPP x ApoE+/- mice tended to be intermediate to PDAPP x ApoE+/+ and PDAPP x ApoE-/- in terms of Aβ and amyloid deposition (Nilsson et al., 2004). Moreover, lower Aβ deposition was seen in an independent transgenic model (Tg2576; APP-Swe) when hemizygous for murine ApoE (Holtzman et al., 2000).

If ApoE mainly serves to clear Aβ, as suggested by Castellano and coauthors, wouldn't one then expect aged PDAPP x mouse ApoE+/- to have more extensive Aβ deposition? After all, the concentration of ApoE, which is supposed to help clear Aβ, was 50 percent lower in the brains of PDAPP x mouse ApoE+/- than those of PDAPP x mouse ApoE+/+. But the experimental findings were actually the opposite.

Here, in an impressive set of experiments, Castellano et al. convincingly demonstrate that human ApoE affects steady-state levels and half-life of extracellular Aβ in an isoform-dependent manner (ApoE4 > ApoE3 > ApoE2) in young PDAPP mice.

How can the seemingly inconsistent observations with human ApoE and murine ApoE be understood? Does human ApoE only clear Aβ perhaps, while mouse ApoE only facilitates Aβ fibril formation? I doubt it.

First, it is worth reflecting upon the fact that human ApoE, which interacts with multiple receptors, is being transferred into the mouse genome environment in the experiments. The human and murine lipoprotein receptors (and also ApoE itself) do not have identical amino acid sequences and structures. Thus, human ApoE will likely not communicate perfectly in the murine genome environment, and this could trigger complex feedback mechanisms, leading to altered cholesterol metabolism and lipoprotein particle composition in human ApoE knock-in mouse brain (as compared to non-transgenic mouse brain expressing mouse ApoE). Indeed, in-vitro data (on astrocyte-based models) suggest that this warrants concern (Fagan et al., 1999).

One way to explain the inconsistent observations with human ApoE and murine ApoE would be that ApoE exerts two separate, but opposing, mechanisms, simultaneously facilitating Aβ clearance and Aβ fibril formation, but that the net lifetime effect of having ApoE4 versus ApoE3 will be increased Aβ accumulation and deposition. Two separate and opposing mechanisms would, for example, explain how human ApoE4 could initially decrease Aβ deposition as compared to ApoE-knockout in young mice (Holtzman et al., 1999), but then increase Aβ deposition in aged mice (Fagan et al., 2002). In young mice largely devoid of Aβ aggregates, the ApoE effect on extracellular Aβ clearance would have greater impact. In contrast, once Aβ aggregates are present in the extracellular space of older APP mice, the ability of ApoE to catalyze Aβ fibril formation by a direct molecular interaction would vastly overshadow its effect on Aβ clearance (Potter et al., 2001).

Multiple disparate effects of ApoE on Aβ metabolism unfortunately mean that there is no simple take-home message, such as “all presenilin and APP mutations increase Aβ42 synthesis.” Many excellent studies, such as the one by Castellano et al., in which robust effects on Aβ phenotypes are consistently seen when the ApoE gene is manipulated in transgenic models, are somewhat disregarded. Strangely, to this day, many papers have an introductory statement saying that “the mechanism of ApoE in Alzheimer’s disease is unknown” (which is hardly ever seen in papers on APP or presenilin).

References:
Bales KR, Verina T, Dodel RC, Du Y, Altstiel L, Bender M, Hyslop P, Johnstone EM, Little SP, Cummins DJ, Piccardo P, Ghetti B, Paul SM. Lack of apolipoprotein E dramatically reduces amyloid beta-peptide deposition. Nat Genet. 1997 Nov;17(3):263-4. Abstract

Bales KR, Verina T, Cummins DJ, Du Y, Dodel RC, Saura J, Fishman CE, DeLong CA, Piccardo P, Petegnief V, Ghetti B, Paul SM. Apolipoprotein E is essential for amyloid deposition in the APP(V717F) transgenic mouse model of Alzheimer's disease. Proc Natl Acad Sci U S A. 1999 Dec 21;96(26):15233-8. Abstract

Nilsson LN, Arendash GW, Leighty RE, Costa DA, Low MA, Garcia MF, Cracciolo JR, Rojiani A, Wu X, Bales KR, Paul SM, Potter H. Cognitive impairment in PDAPP mice depends on ApoE and ACT-catalyzed amyloid formation. Neurobiol Aging. 2004 Oct;25(9):1153-67. Abstract

Holtzman DM, Fagan AM, Mackey B, Tenkova T, Sartorius L, Paul SM, Bales K, Ashe KH, Irizarry MC, Hyman BT. Apolipoprotein E facilitates neuritic and cerebrovascular plaque formation in an Alzheimer's disease model. Ann Neurol. 2000 Jun;47(6):739-47. Abstract

Fagan AM, Holtzman DM, Munson G, Mathur T, Schneider D, Chang LK, Getz GS, Reardon CA, Lukens J, Shah JA, LaDu MJ. Unique lipoproteins secreted by primary astrocytes from wild type, apoE (-/-), and human apoE transgenic mice. J Biol Chem. 1999 Oct 15;274(42):30001-7. Abstract

Holtzman DM, Bales KR, Wu S, Bhat P, Parsadanian M, Fagan AM, Chang LK, Sun Y, Paul SM. Expression of human apolipoprotein E reduces amyloid-beta deposition in a mouse model of Alzheimer's disease. J Clin Invest. 1999 Mar;103(6):R15-R21. Abstract

Fagan AM, Watson M, Parsadanian M, Bales KR, Paul SM, Holtzman DM. Human and murine ApoE markedly alters A beta metabolism before and after plaque formation in a mouse model of Alzheimer's disease. Neurobiol Dis. 2002 Apr;9(3):305-18. Abstract

Potter H, Wefes IM, Nilsson LN. The inflammation-induced pathological chaperones ACT and apo-E are necessary catalysts of Alzheimer amyloid formation. Neurobiol Aging. 2001 Nov-Dec;22(6):923-30. Abstract

View all comments by Lars Nilsson

I want to thank Lars Nilsson for his comments on our recent paper. I think it is important to point out that in the paper by Castellano et al., we demonstrate that human ApoE isoforms differentially influence endogenous Aβ clearance. However, we do not show whether human ApoE isoforms increase or decrease Aβ clearance relative to the absence of ApoE. I believe it is possible that human ApoE isoforms decrease Aβ clearance (E4>E3>E2) relative to no ApoE. If true, one would predict that two copies of human ApoE isoforms would result in greater amyloid deposition and fibril formation than one copy. This experiment has not yet been tested adequately in animal models. As Lars points out, two copies of mouse ApoE results is significantly greater amyloid deposition than one copy of mouse ApoE or than no copies. It will be important to see the effects of two versus one versus zero copies of human ApoE isoforms on amyloid deposition in mouse models.

View all comments by David Holtzman

I would like to echo the comments of Lars Nilsson: simply that mice are not humans, which has been my experience in transplantation. However, that does not mean I do not believe the data; rather, this is great evidence that the E4 subtype can increase the risk through this mechanism.

As for the "cause" of AD, I would like to point out that a majority of LOAD subjects are E3/E3, not E4/E4, although the severity and time to onset is worse in the latter; E3 genotype risk for AD is still quite high (certainly at unacceptable levels). This, in my mind, weakens the argument for a clearance mechanism being the "cause" of AD. Rather, this is a modifier of disease risk and not the cause, but represents a big breakthrough if drugs/conditions can be found to reverse the effects and speed up clearance. Such drugs would be extremely attractive to develop using Dr. Holtzman's assay, given the chronic need for therapies in the patient population. Unquestionably, an effective treatment for AD is needed as soon as possible, and this may certainly be a way forward. Very good work by the...  Read more

I would like to echo the comments of Lars Nilsson: simply that mice are not humans, which has been my experience in transplantation. However, that does not mean I do not believe the data; rather, this is great evidence that the E4 subtype can increase the risk through this mechanism.

As for the "cause" of AD, I would like to point out that a majority of LOAD subjects are E3/E3, not E4/E4, although the severity and time to onset is worse in the latter; E3 genotype risk for AD is still quite high (certainly at unacceptable levels). This, in my mind, weakens the argument for a clearance mechanism being the "cause" of AD. Rather, this is a modifier of disease risk and not the cause, but represents a big breakthrough if drugs/conditions can be found to reverse the effects and speed up clearance. Such drugs would be extremely attractive to develop using Dr. Holtzman's assay, given the chronic need for therapies in the patient population. Unquestionably, an effective treatment for AD is needed as soon as possible, and this may certainly be a way forward. Very good work by the Holtzman group.

View all comments by John Neu

ApoE doesn't just clear amyloid or carry cholesterol. A paper from Weintraub et al., 1987, showed that it is involved in dietary fat clearance, with very interesting differences between isoforms in the carriage of retinyl palmitate, a vitamin A precursor. This brings in retinoids and retinoic acid receptors, which play an important role in AD. Retinoic acid inhibits herpes simplex replication, as well as Chlamydia infection and growth. Vitamin A also stunts the growth of Helicobacter pylori. All of these pathogens have been implicated in Alzheimer's disease.

Vitamin A deficiency is a feature of aging and possibly AD.

References:
Weintraub MS, Eisenberg S, Breslow JL. Dietary fat clearance in normal subjects is regulated by genetic variation in apolipoprotein E. J Clin Invest. 1987 Dec;80(6):1571-7. Abstract

Little CS, Hammond CJ, MacIntyre A, Balin BJ, Appelt DM. Chlamydia pneumoniae induces Alzheimer-like amyloid plaques in brains of BALB/c mice. Neurobiol Aging. 2004 Apr;25(4):419-29. Abstract

Kountouras J, Boziki M, Gavalas E, Zavos C, Grigoriadis N, Deretzi G, Tzilves D, Katsinelos P, Tsolaki M, Chatzopoulos D, Venizelos I. Eradication of Helicobacter pylori may be beneficial in the management of Alzheimer's disease. J Neurol. 2009 May;256(5):758-67. Abstract

Itzhaki RF, Lin WR, Shang D, Wilcock GK, Faragher B, Jamieson GA. Herpes simplex virus type 1 in brain and risk of Alzheimer's disease. Lancet. 1997 Jan 25;349(9047):241-4. Abstract

Carter CJ. The Fox and the Rabbits—Environmental Variables and Population Genetics (1) Replication Problems in Association Studies and the Untapped Power of GWAS (2) Vitamin A Deficiency, Herpes Simplex Reactivation and Other Causes of Alzheimer's Disease ISRN Neurology, 2011.

View all comments by Chris Carter