APOE immunotherapy mitigates cerebral Aβ accumulation and Aβ-associated tau seeding and spreading in a mouse model of amyloidosis

Abstract

APOE is the strongest genetic factor for late-onset Alzheimer’s disease (AD). A specific conformation of the ApoE protein is present in amyloid-β (Aβ) containing plaques. Immunotherapy targeting ApoE in plaques reduces brain Aβ deposits in mice. Here, we evaluated the effects of the anti-human APOE antibody HAE-4 on amyloid plaques, Aβ-mediated tau seeding and spreading, and neuritic dystrophy in the 5XFAD amyloid mice expressing human ApoE4. HAE-4 reduced Aβ plaques as well as Aβ-driven tau seeding/spreading and neuritic dystrophy. These results demonstrate that HAE-4 may provide therapeutic effects on amyloid removal and Aβ driven downstream consequences such as tauopathy.

Introduction

Alzheimer’s disease (AD) is a neurodegenerative disorder in which synaptic and neuronal loss is associated with cognitive decline. AD is characterized by amyloid-β (Aβ) plaques as well as neurofibrillary tangles (NFTs) and neuritic plaques (NP-tau) formed by pathological aggregates of tau proteins^1,2. Genetic predisposition is considerable in AD, with heritability estimates of 60–80%. The Apolipoprotein E (APOE) gene is the strongest genetic risk factor for the late-onset form of AD. One copy of the ε4 allele of APOE increases AD risk approximately 3–4 fold, while two ε4 copies increases AD risk by ~12-fold compared to APOE3 carriers³. Among several mechanisms by which ApoE4 appears to play a role in AD pathogenesis, an important factor is that ApoE4 leads to alterations in Aβ clearance, aggregation, and metabolism leading to promotion of amyloid-β (Aβ) aggregation. In addition, ApoE4 promotes tau pathology and tau-mediated brain atrophy, suggesting a gain of toxic function of ApoE4 on tau-mediated neurodegeneration (for review⁴). Importantly, during the long-time course over which AD pathology develops, there is strong evidence that Aβ deposition builds up throughout the neocortex over ~20 years prior to symptom development and subsequently promotes the spreading of tau pathology and other downstream damage. Both how Aβ promotes seeding and spreading of tau pathology and whether therapies targeting Aβ can block these downstream changes are not clear⁵.

Previously, our group has shown that passive immunotherapy using HAE-4, an antibody targeting a non-lipidated form of human ApoE3 and ApoE4 that is present in amyloid plaques and cerebral amyloid angiopathy (CAA), reduces parenchymal Aβ pathology in mice expressing human ApoE4⁶. Moreover, we recently demonstrated that HAE-4 treatment reduces not only amyloid plaques but also CAA⁷ and improved vessel function without leading to microhemorrhages seen with many anti-amyloid antibodies. HAE-4 also dampened Aβ-associated reactive microglial, astrocytic, and proinflammatory-associated genes in amyloid mice⁷. However, to date, this antibody has not been tested to determine whether it can mitigate amyloid-induced tau seeding and spreading in model systems⁸. Addressing this question is important as tau, but not Aβ pathology, is strongly correlated with cognitive decline and neurodegeneration⁸. Herein, we tested whether HAE-4 treatment affected Aβ-induced NP-tau seeding and spreading in 5XFAD amyloid mice expressing human ApoE4 injected with tau aggregates isolated from human AD brain tissue (AD-tau)⁹. We demonstrated that HAE-4 not only reduced Aβ burden but also strongly decreased Aβ-driven tau seeding and neuritic dystrophy.

Methods

Mice

5XFAD amyloid mice¹⁰ were purchased from The Jackson Laboratory (034840-JAX). Human ApoE knock-in mice apoE4^flox/flox were generated as previously described¹¹. 5XFAD and hApoE4 mice were crossed for several generations to generate 5XFAD-hApoE4 (5XE4) mice. All mice were housed on a 12-hour light/dark cycle with ad libitum access to food. All animal studies were approved by the Animals Studies Committee at Washington University School of Medicine in St. Louis.

Treatment

HAE-4 and control antibodies were generated as described⁶. Five-month-old 5XE4 mice were injected intraperitoneally (i.p.) with a weekly dose of 100 mg/kg body weight with either control IgG or HAE −4 antibody for 13 weeks. At the age of 8 months, the mice were perfused with PBS containing 0.3% heparin and whole brains were carefully extracted and immersion-fixed in 4% paraformaldehyde for 24 h before being transferred to 30% sucrose¹².

Stereotactic intracerebral injections of AD-tau.

AD-tau was isolated as previously described^12,13 from a human AD brain. 5-month-old 5XE4 mice were anesthetized with isofluorane, immobilized in a stereotactic frame, and unilaterally injected with 2 μg AD-tau in the dentate gyrus and overlying cortex (1 μg at each injection site) as previously described^12,14. Mice were allowed to recover on a heating pad and monitored for 48 h after surgery.

Immunohistochemistry (IHC) and immunofluorescence (IF)

IHC and IF were performed as previously described^12,14. For IHC staining of NP-tau and Aβ, primary antibodies AT8 (Thermo Fisher Scientific, MN1020B, 1:500) and HJ3.4 (2 μg/ml in-house) respectively were used.

For IF staining, co-stains were performed for X34, BACE1, and AT8. Fibrillar Aβ was stained with X34 dye (SML-1954) and antibodies to AT8 and BACE1 (abcam, ab108394, 1:500) were used to evaluate peri-plaque pathologies.

Image acquisition and analysis

Image acquisition for IHC and IF was done as previously described^12,14. Briefly, for IHC stains, slides were scanned on the NanoZoomer 2.0-HT system. Images were processed using NDP viewing and Fiji software. For IF images, six z-stacks (20 μm) per section were acquired on a Leica Stellaris 5 confocal microscope. Quantification of confocal images for AT8 and BACE1 around X34⁺ plaques was performed on a semi-automated platform using MATLAB and Imaris 9.5 software to create surfaces of each stain based on a threshold applied to all images, dilatation of X34 surfaces to 15 μm, and co-localization of various immunostained surfaces and dilated X34 surfaces. All staining experiments were imaged and quantified by a blinded investigator.

Statistics

All data are presented as mean ± SEM. GraphPad Prism 8.0 was used to perform statistical analyses. Gaussian distribution was evaluated using the D’Agostino–Pearson normality test. Statistical analysis was performed using unpaired t tests. If samples deviated from normal distribution, statistical analysis was performed using Mann-Whitney test. Statistical significance was set as *, p < 0.05; **, p < 0.01; and ***, p < 0.001 vs. CTL group.

Results

In this study, 5XE4 mice were treated with a weekly i.p. injection of control or HAE-4 antibodies starting at 5 months of age. One week after the beginning of treatment, mice were unilaterally injected in the dentate gyrus and the overlying cortex with AD-tau and brains were analyzed 3 months post AD-tau injection.

Anti-human ApoE HAE-4 antibody significantly decreased Aβ plaques in 5XE4 mice.

We previously described that HAE-4 binds to ApoE in the core of plaques and recruits microglial mediated Aβ phagocytosis leading to lower Aβ burden⁶. To validate the efficacy of HAE-4 on Aβ pathology, we first stained brains using an anti-Aβ antibody (HJ3.4) and a stain for fibrillar plaques (X34) (Fig.1). HAE-4 significantly reduced Aβ staining in the hippocampus (HC) and the cortex (CTX) of both male (green) and female (orange) 5XE4 mice compared to controls (Fig.1A, B). Similarly, reduced fibrillar plaques were observed in all regions assessed except for the ipsilateral HC in males and ipsilateral CTX in females in HAE-4 treated mice compared to control-treated mice (Fig.1 C, D). These data confirm our previous results demonstrating that targeting ApoE with therapeutic antibodies can reduce Aβ pathology^6,7.

Figure 1: HAE-4 significantly decreased parenchymal Aβ plaques in 8-month-old AD-tau injected 5XE4 mice.

(A) Representative images of Aβ-stained plaques in 5XE4 mice. Scale bars, 1 mm. (B) Quantification of the area covered by Aβ staining in the ipsi- and contralateral hippocampi and cortices. (C) Representative images of X34⁺ fibrillar plaques. Scale bars, 1 mm. (D) Quantification of the area covered by X34 staining in the ipsi- and contralateral hippocampi and cortices of 5XE4 mice. Data are presented as mean ± SEM. *, P < 0.05; **, P < 0.01; ***, P < 0.001. n = 9–11/group. Arrows indicate the ipsilateral side of the brain (AD-tau injection side). In the graphs, males are represented in green, and females are represented in orange.

HAE-4 significantly reduced Aβ-driven tau seeding and spreading in 5XE4 mice.

Since amyloid plaques are able to facilitate seeding and spreading of phosphorylated, aggregated tau associated with amyloid plaques, we next investigated whether HAE-4 treatment could dampen Aβ-mediated tau seeding and spreading. Using AT8 staining which recognizes tau phosphorylated at both serine 202 and threonine 205, we observed widespread seeded NP-tau throughout the ipsilateral HC and CTX of 5XE4 mice as well as NP-tau that spread to the contralateral CTX and HC. Interestingly, despite that fact that the control-treated female 5XFAD mice had greater hippocampal Aβ load than the control treated 5XFAD males, they had the similar levels of Aβ-related tau seeding ipsilateral and less spread contralateral to the injection. In the future, it will be important to assess tau seeding and spreading in younger female 5XE4 mice with similar amyloid load comparable to males to determine whether this difference is a sex or Aβ load-dependent effect. HAE-4 strongly reduced NP-tau seeding in all groups compared to control mice (Fig. 2A, B). The protective effect of HAE-4 on Aβ-driven tau spreading to the contralateral side of the brain was also significantly reduced but only in males. As HAE-4 decreased Aβ pathology (Fig.1), we next evaluated whether the observed decrease of NP-tau aggregation is seen locally around amyloid deposits. We performed confocal analysis and quantified the amount of NP-tau surrounding individual X34⁺ Aβ plaques (Fig. 2C, D). In females, we did not detect a change in NP-tau on a per plaque basis between treatment groups, possibly as a result of the increased amyloid burden in female 5XFAD mice. However, in males, we observed a significant decrease of seeded NP-tau pathology per plaque in HAE-4-treated mice, suggesting that HAE-4 can affect tau pathology that is independent from Aβ plaque reduction.

Figure 2: HAE-4 significantly reduced Aβ-driven tau seeding and spreading in 8-month-old AD-tau injected 5XE4 mice.

(A) Representative images of AT8⁺ NP-tau in 5XE4 mice. Scale bars, 1 mm. (B) Quantification of the area covered by AT8⁺ staining in the ipsi- and contralateral hippocampi and cortices of AD-tau–injected 5XE4 mice. (C) Confocal images of AT8⁺ NP-tau (green) around X34⁺ plaques (blue) in the ipsilateral subiculum of the hippocampus. Scale bars, 20 μm. (D) Quantification of percentage AT8⁺ volume within 15 μm of plaques in the ipsilateral subiculum of the hippocampus. Data are presented as mean ± SEM. *, P < 0.05; **, P < 0.01. n = 9–11/group. Arrows indicate the ipsilateral side of the brain (AD-tau injection side). In the graphs, males are represented in green, and females are represented in orange.

HAE-4 treatment decreased global BACE1⁺ neuritic dystrophy but not BACE1⁺ neuritic dystrophy per plaque in 5XE4 mice.

Because of the HAE-4 protective effect on Aβ-mediated tau seeding, we further evaluated Aβ-dependent neuritic dystrophy burden in these mice (Fig.3). Dystrophic neurites are abnormal neuronal processes characterized by aberrant swelling with accumulation of various cellular organelles and proteins that occur around amyloid plaques. We immunostained for BACE1, which accumulates in dystrophic neurites¹⁵ and observed a strong decrease in the global amount of BACE1⁺ processes in the HAE-4 group compared with CTL in both ipsi- and contralateral CTX and HC (Fig. 3A, B). However, when we evaluated BACE1⁺ neuritic dystrophy per Aβ plaque, we did not observe any changes in males or females (Fig. 3C, D), suggesting that the protective effect from HAE-4 on neuritic dystrophy is dependent on HAE-4’s ability to reduce total Aβ burden.

Figure 3: HAE-4 treatment decreased the total BACE1⁺ neuritic dystrophy but not BACE1⁺ neuritic dystrophy per plaque in 8-month-old AD-tau injected 5XE4 mice.

(A) Representative images of BACE1⁺ neuritic dystrophy in 5XE4 mice. Scale bars, 1 mm. (B) Quantification of the area covered by BACE1⁺ staining in the ipsi- and contralateral hippocampi and cortices of AD-tau–injected 5XE4 mice. BACE1 staining in hippocampal mossy fibers is present even without amyloid plaques and was not quantified. (C) Confocal images of BACE1⁺ neuritic dystrophy (red) around X34⁺ plaques (blue) in the ipsilateral subiculum of the hippocampus. Scale bars, 20 μm. (D) Quantification of percentage BACE1⁺ volume within 15 μm of plaques in the ipsilateral subiculum of the hippocampus. Data are presented as mean ± SEM. *, P < 0.05; **, P < 0.01. n = 9–11/group. Arrows indicate the ipsilateral side of the brain (AD-tau injection side). In the graphs, males are represented in green, and females are represented in orange.

Discussion

AD is a significant public health crisis with around 50 million people affected worldwide, highlighting the urgent need to develop strategies to slow/prevent AD. In this regard, it is estimated that a treatment that delays the onset of cognitive decline in AD by 5 years would reduce the number of cases by 50%. To counteract AD, several therapies have been developed to decrease Aβ pathology. Although some anti–Aβ antibodies have showed promise in decreasing amyloid in the brain with preliminary evidence for slowing cognitive decline¹⁶ in very mild AD, side effects include both asymptomatic and symptomatic amyloid-related imaging abnormalities^17,18. In the past, we demonstrated that the anti-ApoE antibody HAE-4 can decrease Aβ burden in a similar fashion to anti–Aβ antibodies in mice, while not inducing microhemorrhages associated with Aβ immunotherapy⁷. There is limited data on whether anti-Aβ antibodies can impact Aβ-driven pathologies such as tau seeding and spreading^19,20 and no data on the effects of this or other anti-ApoE antibodies. In TauPS2APP mice, a mouse model that develop both amyloid and tau pathology, chronic exposure to anti-amyloid-β antibodies decreases amyloid plaque burden and prevents phospho-tau pathology in the hippocampus²⁰. Another study demonstrated that anti-Aβ immunization increases plaque compaction, reduces the spread of tau in the hippocampal formation and prevented the formation of tau-positive dystrophic neurites in a mouse model of amyloidosis overexpressing human tau through AAV injection¹⁹. However, the treatment did not significantly reduce tau-induced neurodegeneration in the dentate gyrus¹⁹. In this study, we report that HAE-4 has a protective effect on reducing Aβ plaques while attenuating Aβ-mediated tau seeding/spreading and neuritic dystrophy. This strengthens the therapeutic potential of HAE-4 treatment since tau pathology strongly correlates with neuronal loss and cognitive decline in AD⁸. Although more than 75% of participants who have qualified to date for anti-Aβ immunotherapy trials are ApoE4+^16,21, it will be important to assess HAE-4 antibody efficiency in models expressing other ApoE isoforms as well as determining its safety, tolerability, and therapeutic potential in humans after humanization of the antibody. Of note, the HAE-4 antibody only binds ApoE in plaques. It does not lower overall ApoE levels. In addition to the effects of HAE-4 on amyloid-induced tau seeding and spreading, another approach that has potential to decrease tau-mediated neurodegeneration is by lowering brain ApoE levels with approaches such as anti-sense oligonucleotides, recently described (ref).

There are some limitations in interpreting the translational implications of our study. First, although the 5XE4 mouse model that we used recapitulates amyloid deposits and tau seeding as seen in humans, this model does not develop tau-mediated neurodegeneration and cognitive deficits. Thus, our findings suggest that HAE-4 might be highly effective in the late preclinical and possibly very early clinical phase of AD when Aβ appears to drive tau-seeding and spreading but when there is not much neuronal and synaptic loss or cognitive decline. As the model we utilized does not display tau-mediated behavioral deficits, how this potential therapy would translate into human AD is not clear. Altogether, this study strongly strengthens the therapeutic interest for assessing an anti-ApoE antibody with properties of HAE-4 for diseases involving Aβ deposition such as AD and CAA.

Summary for Social Media:

1. If you and/or a co-author has a Twitter handle that you would like to be tagged, please enter it here. (format: @AUTHORSHANDLE).

   @holtzman4

2. What is the current knowledge on the topic? (one to two sentences)

   APOE is the strongest genetic factor for late-onset Alzheimer’s disease (AD). A specific conformation of the ApoE protein is present in amyloid-β (Aβ) containing plaques. Immunotherapy targeting ApoE in plaques reduces brain Aβ deposits in mice.

3. What question did this study address? (one to two sentences)

   Here, we evaluated the effects of the anti-human APOE antibody HAE-4 on amyloid plaques, Aβ-mediated tau seeding and spreading, and neuritic dystrophy in 5XFAD amyloid depositing mice expressing human ApoE4.

4. What does this study add to our knowledge? (one to two sentences)

   HAE-4 reduced Aβ plaques as well as Aβ-driven tau seeding/spreading and neuritic dystrophy.

5. How might this potentially impact on the practice of neurology? (one to two sentences)

   These results demonstrate that HAE-4 may provide therapeutic effects on amyloid removal and Aβ driven downstream consequences such as tauopathy.

Abbreviations:

ApoE

apolipoprotein E

AD

Alzheimer’s disease

Aβ

amyloid-β

NFTs

neurofibrillary tangles

CNS

central nervous system

CAA

cerebral amyloid angiopathy

NP-tau

phosphorylated forms of tau in neuritic plaque

CTX

cortex

HC

hippocampus