A Match Made in Microglia? ApoE and Aβ Click in Lysosomes, Seeding Plaque

Scientists broadly agree that both apolipoprotein E and microglia are necessary ingredients for amyloidosis. In mice devoid of ApoE, or of their microglia, scant plaques form. Now, a sweeping study published October 9 in Immunity unveils a mechanistic link between the two. Scientists led by Mikael Simons of the German Center for Neurodegenerative Diseases in Munich report that, within the bowels of microglial lysosomes, ApoE forms fibrillar aggregates alongside Aβ. There, the two festering fibrils team up, forming seeds with the potential to grow into Aβ plaques once microglia either spew them out, or die.

  • Some Aβ plaques consist mostly of fibrillar aggregates of ApoE.
  • Both ApoE and Aβ form aggregates within microglial lysosomes.
  • ApoE4 does this more readily than ApoE3.
  • ApoE lipidation promotes microglial uptake; delipidation in lysosomes may goad aggregation.

Gobs of these intra-lysosomal deposits formed when lysosomes were hobbled with inhibitors, or when interferon signaling was dialed up. The findings implicate a nexus of ApoE, microglia, and endolysosomal dysfunction as the driving force behind amyloidosis in AD.

“This is one of the better papers I’ve read in the past few years,” said David Holtzman of Washington University in St. Louis. He thinks it offers a convincing explanation for why both microglia and ApoE seem to be needed to seed plaques.

The study also underscores the primary role of lysosomal dysfunction in the etiology of AD, commented Ralph Nixon of New York University. Nixon’s research has primarily focused on how this pathology plays out in neurons, and this new work demonstrates the importance of parallel lysosomal crises unfolding in microglia. “This paper pulls together a lot of seemingly disparate observations in the field,” he said.

Jessica Young of the University of Washington Medical School, Seattle, shares this sentiment. “The endo-lysosomal, immune, and lipid trafficking pathways are highly interrelated as lipids and immune mediators utilize vesicular trafficking,” she wrote. “This study highlights how events that converge at the lysosome can initiate some of the earliest events in AD pathogenesis.” 

Ground Zero for Amyloidosis? Microglia internalize lipidated ApoE and Aβ into the lysosome. There, ApoE is delipidated, promoting its aggregation and supporting Aβ aggregation. The resulting co-aggregates may later seed amyloid plaques. [Courtesy of Kaji et al., Immunity, 2024.]

More papers than a writer could list here have shown that ApoE—the strongest genetic risk factor for AD—plays some part in amyloid plaque formation, and is itself integrated into Aβ plaques (e.g., Kim et al., 2012; Dec 2017 news). While microglia have long been credited with clean-up and containment of plaques, recent studies have pointed to a potential role as plaque builders, as well (Sep 2019 news; Apr 2021 news; Apr 2024 conference news).

Endolysosomal problems have also taken some blame as fueling AD pathogenesis. Some scientists favor the idea that endosomal traffic jams spark amyloidosis inside of neurons (Cataldo et al., 2000; Jul 2015 news; Jan 2010 news). Many AD risk genes are implicated in immune, lipid, and endolysosomal pathways.  

How do all these different puzzle pieces fit together? Co-first authors Seiji Kaji, Stefan Berghoff, and Lena Spieth and colleagues initially had their sights set on humbler questions relating to ApoE’s role of in remyelination. To answer them, they generated ApoE-Halo mice, which express a tagged version of ApoE that can be visualized with high resolution within cells.

Alas, when they crossed ApoE-Halo to 5xFAD amyloidosis mice, they were in for a surprise. As expected, they saw both Aβ and Halo-tagged ApoE co-mingling in most plaques in 3.5-month-old mice, as well as some plaques that contained mostly Aβ and little ApoE. However, the scientists also spotted an unexpected breed of plaque: one that contained mostly ApoE and hardly any Aβ. These “ApoE-intense” aggregates comprised 14 percent of all plaques. Like their more evenly mixed counterparts, they stained positive with methoxy-04, thiazine red, and thioflavin, suggesting their contents were fibrillar. In human AD brain samples, the scientists saw a similar repertoire of plaques, with varying compositions of Aβ and ApoE.

Intense Surprise. This confocal image of a 5xFAD/Halo-ApoE mouse brain shows two of three plaque types. An “Aβ intense” plaque in the top left corner contains mostly Aβ (turquoise) and no ApoE (yellow). The “ApoE-intense” plaque in the middle contains mostly ApoE and scant Aβ (MX04, white; Iba1, magenta). [Courtesy of Kaji et al., Immunity, 2024.]

Deploying biochemical and microscopic experiments to analyze these ApoE aggregates, the scientists found that, in sarkosyl-insoluble fractions isolated from ApoE-Halo/5xFAD mice, purified ApoE-Halo existed as multimers devoid of Aβ, suggesting they were able to form independently. Under the electron microscope, these ApoE multimers were spied twisting into wavy, snake-like fibrils. Similar structures appeared when the researchers isolated ApoE aggregates from postmortem brain samples of ApoE3/E3 and ApoE4/E4 carriers with AD. 

Serpentine ApoE. Electron micrographs of sarkosyl-insoluble ApoE isolated from 5xFAD/ApoE-Halo mouse brain (left) or from human AD brain samples (right panels). [Courtesy of Kaji et al., Immunity, 2024.]

In seeding experiments, ApoE aggregates revved up amyloidosis in the mouse brain. For example, 2-month-young 5xFAD mice injected with ApoE aggregates extracted from other 5xFAD mice formed twice as many plaques over the following two months as controls. The ApoE aggregates even managed to seed plaques in ApoE knockout 5xFAD mice, which typically have few plaques. Human ApoE had the same plaque-seeding effect in human ApoE knock-in 5xFAD mice. When injected with ApoE aggregates extracted from human AD brain samples, plaque numbers doubled in ApoE3-KI and tripled in ApoE4-KI mice. This suggested that both isoforms of aggregated ApoE stoke amyloidosis, with ApoE4 doing so with more gusto.

How and where does ApoE aggregate in the first place? To investigate, Kaji and colleagues hunted them down in the ApoE-Halo/5xFAD mouse brain with confocal microscopy and three-dimensional imaging. In addition to the usual large extracellular plaques, ApoE aggregates also turned up in tiny punctate structures within Iba1+ microglia, where they overlapped with MX04+ aggregates (image below).

Toxic Merger. A three-dimensional rendering of a microglial cell reveals co-localization between ApoE (yellow), CD68 (turquoise), and methoxy-04 fibrillar aggregates (gray). Far-left image includes Iba1+ staining (magenta) to show the whole cell; the rest show the same cell without Iba1 for clarity. [Courtesy of Kaji et al., Immunity, 2024.]

This microglial hide-out for ApoE aggregates appeared crucial for the formation of extracellular Aβ plaques. Consider what happened when the mice had no microglia thanks to the CSF-1R inhibitor, PLX5622. Without microglia around, injected ApoE aggregates had no plaque-seeding power in 5xFAD mice. Notably, when the researchers injected purified ApoE-Halo into ApoE knockout mice, and 5xFAD mice with or without microglia, they found that plaques only formed when both ApoE and microglia were present.

The scientists next explored how lysosomal function influenced the development of ApoE aggregates within microglia. Both in a cultured microglial cell line and in human iPSC-derived microglia, the scientists sparked intracellular aggregates by treating the cells with recombinant ApoE and fluorescently labeled Aβ42 peptides. Hobbling lysosomal function with a suite of inhibitors ramped up this aggregation. So did knockdown of the microglial receptor TREM2, which promotes lysosomal function. These same results were borne out in 5xFAD mice that received daily injections of the lysosomal acidification blocker chloroquine.

In contrast, shutting down the interferon response with the JAK-STAT inhibitor baricitinib had the opposite effect. It quelled production of ApoE aggregates within microglia. So far, the findings suggested that subpar lysosomal function, or overactive interferon signaling, could accelerate the process of ApoE/Aβ aggregation within microglia.

“The findings of this paper add to the accumulating evidence that dense-core, highly-aggregated Aβ plaques do not form spontaneously within the extracellular spaces of the AD brain, but rather are assembled intracellularly—within the acidic environment of microglial lysosomes,” wrote Greg Lemke of the Salk Institute in La Jolla, California.

As an apolipoprotein, ApoE’s main job is to hook up with lipids and deliver them to cells. To investigate how this process might relate to its aggregation within microglial lysosomes, the researchers added Aβ42 along with lipid-free or lipidated ApoE to microglial cultures. They found that microglia more readily internalized lipidated ApoE than its nonfat version, and therefore led to more aggregates within the lysosome. Further studies revealed that once inside the cells, lipidated ApoE proteins were rapidly stripped of their lipids.

Notably, in experiments with human ApoE, ApoE4 was more readily taken up by microglia, and it more efficiently lost its lipids, relative to ApoE3. Simons thinks that while associated lipids promote receptor-mediated uptake of ApoE, subsequent delipidation—which occurs within the acidic environment of the lysosome—renders ApoE more prone to aggregate. Aβ has also been shown to aggregate more readily in acidic compartments.

Lipids Open the Door. Microglial cells treated with unlipidated (top) or lipidated (bottom) ApoE (yellow) along with fluorescently tagged Aβ (green). Lipidated ApoE was more readily internalized and aggregated in lysosomal compartments (magenta) along with Aβ. [Courtesy of Kaji et al., Immunity, 2024.]

Holtzman told Alzforum that this idea is consistent with findings from his lab using an antibody specific for delipidated ApoE. It not only revealed that ApoE is largely delipidated within plaques, but that targeting this lipid-free ApoE with antibodies removed plaques (Apr 2018 news). Perhaps the acidic environment of the microglial lysosome is where this critical lipid stripping step occurs, Holtzman suggested. It also happens to be where the newly disrobed ApoE rendezvous with Aβ.

When the scientists disabled cholesterol biosynthesis in cell cultures, or in 5xFAD mice by conditionally knocking out a key sterol synthesis enzyme in microglia, the cells cranked up expression of lipoprotein receptors, and both intracellular lysosomal aggregates and extracellular plaques soared, according to methoxy-04 staining. This suggested that in a lipid-deprived state, microglia take up more ApoE, exacerbating intracellular aggregate formation and ultimately, Aβ plaques.

Jason Ulrich of Washington University in St. Louis wondered whether reduced receptor-mediated uptake of ApoE2 and of ApoE3-Christchurch isoforms by microglia might explain their protective effects. Simons has yet to test these variants.

Simons proposes a chain of events whereby microglia consume both Aβ and ApoE, which are typically degraded within the lysosome. When lysosomes become overwhelmed, as can happen due to any number of age-related and genetic risk factors, these two proteins might linger in the lysosomes long enough to aggregate.

“While the demonstration of ApoE and Aβ aggregates in microglial lysosomes is certainly exciting, it remains to be determined how the intracellular aggregates seed the extracellular plaques, and to what degree that contributes to the overall Aβ pathology,” noted Hui Zheng of Baylor College of Medicine in Houston.

The authors have yet to answer these questions. Simons proposed that the intracellular aggregates may be spewed out by microglia, or exposed once microglia die, possibly due to lysosomal damage inflicted by the aggregates. Recent studies suggest that microglia can form “lysosomal synapses” with the plasma membrane to release their contents (Apr 2023 conference news). Furthermore, when Simons recently presented this work at the Eibsee conference in Germany, John Hardy of University College London commented that these new findings harken back to “amyloid-related cells” spotted in AD brain samples more than three decades ago (Roher et al., 1988). These ARCs were likely microglia, and they contained amyloid filaments within intracellular vesicles. At the time, the authors proposed that these vesicles fused with the plasma membrane, releasing their acidic contents and fueling amyloidosis.

Nixon favors the idea that microglial cell death, caused by ruptured lysosomes, could give rise to extracellular plaques. He has reported such a phenomenon within neurons, which produce intracellular Aβ themselves (Jun 2022 news). Neurons also produce and take up ApoE when stressed, he noted (Feb 2023 news). Nixon suspects microglia may take up aggregated Aβ from dying neurons, and that this could contribute to its subsequent aggregation along with ApoE in microglial lysosomes. Simons told Alzforum that his lab only spotted ApoE/Aβ aggregates in microglia.

To Holtzman’s mind, the study provides compelling evidence that ApoE aggregation within microglial lysosomes is important in the early phase of amyloidosis. But would it contribute later on in disease, during the phase of amyloid-induced tauopathy? Holtzman and others have pegged ApoE, ApoE4 in particular, as exacerbating tauopathy-induced neurodegeneration (Sep 2017 news). In tauopathy models, microglia lysosomes are chock-full of lipids, and this accumulation requires ApoE. “This suggests that ApoE influences lysosomal function,” he said, implying that perhaps ApoE-induced lysosomal problems are a common feature of both Aβ and tau pathologies.—Jessica Shugart

Comments

  1. Kaji and colleagues use a variety of experimental approaches to demonstrate that, in the AD brain, lipid-associated ApoE is phagocytically internalized into microglia together with lower-order polymers of Aβ, and that ApoE facilitates the dense fibrillar aggregation of Aβ within microglial lysosomes. This co-internalized and co-aggregated ApoE/Aβ is apparently eventually deposited into dense-core plaques, since it is stained by the β-pleated-sheet-binding dyes (e.g., Congo Red) that identify these plaques. The findings of this paper add to the accumulating evidence that dense-core, highly-aggregated Aβ plaques do not form spontaneously within the extracellular spaces of the AD brain, but rather are assembled intracellularly—within the acidic environment of microglial lysosomes.

    View all comments by Greg Lemke

  2. This is a fantastic study from the Simons lab that brings together a lot of key observations into an attractive theoretical framework for how APOE and microglia conspire to form nascent amyloid fibrils to seed pathology. It begs the question of whether decreased receptor-mediated microglial uptake of APOE variants such as APOE2 or APOE3ch influences their ability to seed amyloid pathology.

    Another question is, what is it about microglial lysosomes as opposed to other cell types that foments a fibrillogenic interaction between APOE and Aβ? Aβ and APOE can also be taken up and trafficked to lysosomes in neurons or astrocytes. Yet depletion of microglia prior to pathology has such a profound effect on amyloid burden.

    Translationally, these findings further suggest that decreasing APOE expression might be a way to slow amyloid pathology, although this would need to be done very early. Potentially it could also suggest that diverting APOE uptake into other cell types could also be protective. This may explain why LDLR overexpression using a Prion promoter decreased amyloid pathology (Kim et al., Neuron, 2009), while upregulation of APOE receptors in microglia in this paper augmented aggregation.

    View all comments by Jason Ulrich

  3. Genetics studies point to endo-lysosomal, immune, and lipid pathways as dysfunctional in AD, and many of these genes are highly expressed in glia. In this study, Kaji et al., perform an impressive amount of experiments to show microglial cells are necessary for Aβ seeding, exacerbated by APOE, and that this event is mediated from the microglial lysosome. The endo-lysosomal, immune, and lipid trafficking pathways are highly interrelated as lipids and immune mediators utilize vesicular trafficking.

    This study highlights how events that converge at the lysosome can initiate some of the earliest events in AD pathogenesis. This lends weight to the idea that targeting this pathway therapeutically may be beneficial for mitigating some of the drivers of AD pathology. It also shows that the lysosome is more than just an organelle that degrades unwanted material, suggesting that it may also function as a signaling hub that can respond to stimuli such as interferon signaling. How this is mediated is still an open and intriguing question.

    View all comments by Jessica Young
    • User Profile Image Hui Zheng
      Baylor College of Medicine
    • Posted:

    The current work is built on the creation of a knock-in mouse model expressing a HaloTag-tagged APOE, thus allowing its in situ visualization and biochemical purification. Using this elegant system, coupled with various imaging modality and biochemical methods, the authors reveal the formation of APOE aggregates in microglia within the endo-lysosomal compartment. The aggregates serve as a co-factor to facilitate Aβ aggregation and also impinge on microglial inflammation and lipid metabolism. Overall, the finding adds further evidence supporting a critical role of the immune, lipid, and endo-lysosomal pathways in AD.

    While the demonstration of APOE and Aβ aggregates in microglial lysosome is certainly exciting, it remains to be determined how the intracellular aggregates seed the extracellular plaques and to what degree it contributes to the overall Aβ pathology. In addition, given the prominent role of APOE genotypes in microglia lipid homeostasis and tau-induced neurodegeneration, as demonstrated by the multiple publications from the Holtzman group, it would be interesting to assess whether similar APOE aggregates also form in the microglial lysosome of tauopathy conditions.

    View all comments by Hui Zheng

  5. Kaji et al. propose that fibrillar ApoE—likely forming in microglia lysosomes after internalization and delipidation of the lipid carriers—promotes aggregation of Aβ. The authors preaggregated HFIP Aβ by incubating a solution in PBS for 48 hours at 37°C, then fed this to microglial cell cultures. This Aβ preparation most likely included a mixture of soluble and aggregated Aβ species. Microglial cells internalized the Aβ, promoting aggregation in the presence of ApoE fibrils, and this effect was consistent across BV2, primary, and iPSC-derived microglia, reinforcing the reliability of the findings beyond iPSC models. Large aggregates could have resulted from soluble Aβ binding to pre-existing fibrils during cellular incubation in the endolysosomal compartment and/or extracellularly.

    In their experiments, lipidated ApoE was internalized by microglial cells and incorporated into lysosomes at a higher rate than its unlipidated counterpart, which enhanced Aβ aggregation. It is, however, worth reflecting on the fact that apoE4 is poorly lipidated relative to their ApoE2 and ApoE3 counterparts (Hanson et al., 2013), which would imply a lower rate of internalization -and fibrillation- into acidic compartments.

    The authors ablated the JAK/STAT pathway in microglia using baricitinib, a JAK/STAT1 inhibitor, and they observed a reduction in Aβ brain deposition. Baricitinib has also been shown to ablate the PI3K-AKT pathway in vivo (Hindam et al., 2024). We have found that primary microglial cells establish extracellular, sealed compartments onto large synthetic Aβ aggregates into which lysosomal contents are secreted. This mechanism, known as digestive exophagy, is mediated by the PI3K-AKT pathway (Apr 2023 conference news). It is tempting to speculate that the decreased Aβ aggregation in 5xFAD mice treated with baricitinib could be in part due to reduced lysosomal exocytosis of previously internalized fibrillar material by microglia towards extracellular Aβ deposits.

    References:

    Hanson AJ, Bayer-Carter JL, Green PS, Montine TJ, Wilkinson CW, Baker LD, Watson GS, Bonner LM, Callaghan M, Leverenz JB, Tsai E, Postupna N, Zhang J, Lampe J, Craft S. Effect of apolipoprotein e genotype and diet on apolipoprotein e lipidation and amyloid peptides: randomized clinical trial. JAMA Neurol. 2013 Aug 1;70(8):972-80. PubMed.

    Hindam MO, Ahmed LA, El Sayed NS, Khattab M, Sallam NA. Repositioning of baricitinib for management of memory impairment in ovariectomized/D-galactose treated rats: A potential role of JAK2/STAT3-PI3K/AKT/mTOR signaling pathway. Life Sci. 2024 Aug 15;351:122838. Epub 2024 Jun 17 PubMed.

    View all comments by Santiago Sole-Domenech

Make a Comment

To make a comment you must login or register.

References

News Citations

  1. ApoE4 Promotes Amyloidosis, But Only in Plaque-Free Mice
  2. Are Microglia Plaque Factories?
  3. Microglia Build Plaques to Protect the Brain
  4. Over the Span of AD, Roles of Astrocytes and Microglia Change
  5. Partners in Crime: APP Fragment and Endosomal Protein Impair Endocytosis
  6. Inside Out—Plaques May Have Intracellular Origin
  7. Human ApoE Antibody Nips Mouse Amyloid in the Bud
  8. From Phagocytosis to Exophagy: Microglia's Digestive Tract Dissected
  9. Behold PANTHOS, a Toxic Wreath of Perinuclear Aβ That Kills Neurons
  10. Secreted by Neurons, ApoE4 Makes Tangles and Degeneration Worse
  11. ApoE4 Makes All Things Tau Worse, From Beginning to End

Mutations Citations

  1. APOE C130R (ApoE4)

Therapeutics Citations

  1. Baricitinib

Paper Citations

  1. Kim J, Eltorai AE, Jiang H, Liao F, Verghese PB, Stewart FR, Basak JM, Holtzman DM. Anti-apoE immunotherapy inhibits amyloid accumulation in a transgenic mouse model of Aβ amyloidosis. J Exp Med. 2012 Nov 19;209(12):2149-56. PubMed.
  2. Cataldo AM, Peterhoff CM, Troncoso JC, Gomez-Isla T, Hyman BT, Nixon RA. Endocytic pathway abnormalities precede amyloid beta deposition in sporadic Alzheimer's disease and Down syndrome: differential effects of APOE genotype and presenilin mutations. Am J Pathol. 2000 Jul;157(1):277-86. PubMed.
  3. Roher A, Gray EG, Paula-Barbosa M. Alzheimer's disease: coated vesicles, coated pits and the amyloid-related cell. Proc R Soc Lond B Biol Sci. 1988 Jan 22;232(1269):367-73. PubMed.

Further Reading

Primary Papers

  1. Kaji S, Berghoff SA, Spieth L, Schlaphoff L, Sasmita AO, Vitale S, Büschgens L, Kedia S, Zirngibl M, Nazarenko T, Damkou A, Hosang L, Depp C, Kamp F, Scholz P, Ewers D, Giera M, Ischebeck T, Wurst W, Wefers B, Schifferer M, Willem M, Nave KA, Haass C, Arzberger T, Jäkel S, Wirths O, Saher G, Simons M. Apolipoprotein E aggregation in microglia initiates Alzheimer's disease pathology by seeding β-amyloidosis. Immunity. 2024 Nov 12;57(11):2651-2668.e12. Epub 2024 Oct 16 PubMed. Correction.

Annotate

To make an annotation you must Login or Register.