ApoE4 delivers a blow to neurons in mouse models of tauopathy but, according to a paper published October 10 in the Journal of Experimental Medicine, it only does so if mice have their microglia. Researchers led by David Holtzman at Washington University in St. Louis reported that ridding the brain of these immune cells pre-empted this neurodegeneration. ApoE knockouts were protected from tau toxicity and neurodegeneration, as well, regardless of whether they had microglia, the authors found. The work suggests that a toxic alliance between ApoE4 and microglia accelerates tau pathology.

  • In mice, ApoE4 drives tau pathology and neurodegeneration.
  • Depleting microglia prevented shrinkage of the brain.
  • It also halted the spread of tau pathology throughout the brain.

“Overall, the findings place ApoE and microglia as dominant contributors to neurodegeneration,” said Oleg Butovsky of Brigham and Women’s Hospital in Boston.

Holtzman and colleagues previously reported that all three isoforms of ApoE, but ApoE4 in particular, fueled neuroinflammation, tau pathology, and neurodegeneration in P301S mice. Deleting their endogenous ApoE, or the microglial receptor TREM2, for that matter, largely spared neurons (Apr 2017 conference news; Sep 2017 news; Oct 2017 news on Leyns et al., 2017). Around the same time, TREM2 and ApoE were placed at the heart of microglia’s so-called neurodegenerative phenotype (Sep 2017 news). This raised the question of whether ApoE’s effect on tau and neurons depends upon microglia.

To investigate, first author Yang Shi and colleagues wiped out microglia entirely using PLX3397, an inhibitor of the colony-stimulating factor 1 receptor. Microglia need CSF1R signaling to survive. Shi treated four different mouse lines with PLX3397: P301S-tau mice; P301S expressing human ApoE4; P301S mice with ApoE knocked out; and wild-type mice with ApoE knocked out. They started lacing the mice’s chow at 6 months of age, when tau pathology in tau/ApoE4 mice is still at an early stage and neurons are not degenerating yet, and settled on a protocol that eliminated all microglia in male mice within 21 days. Because they were unable to completely wipe out microglia in females, the researchers used males throughout the study.

Deadly Alliance. Tau/ApoE4 mice on control chow (left) had brain atrophy and an enlarged ventricle. Removing microglia prevented atrophy (middle). So did removing ApoE (right). [Courtesy of Shi et al., JEM, 2019.]

When the mice were 9 months old, the researchers observed a striking result: microglial depletion completely protected the tau/ApoE4 mice from neurodegeneration. Tau/ApoE4 mice on control chow had up to 50 percent smaller hippocampi and entorhinal/piriform cortices, respectively, and a massively enlarged brain ventricle, compared with their counterparts on PLX3397 (see image above). Tau/ApoE KO mice were likewise protected. Overall, the findings suggest that ApoE drives neurodegeneration in the tauopathy mice primarily via microglia.

In fact, removing microglia completely blocked the typical march of tau pathology through the brains of the tau/ApoE4 mice. Previously, the scientists used the AT8 antibody to describe four stages of worsening neurodegeneration. Stage 1 was marked by intense p-tau staining in hippocampal mossy fiber axons, with but diffuse staining in a smattering of neuronal cell bodies in the granule cell layer and CA1 regions. As disease progressed, p-tau accumulated further in these cell bodies, spread to other hippocampal neurons, and clumped up into tangles. The entire hippocampus was covered with p-tau by stage 4, which correlated with the most brain atrophy (Shi et al., 2017). Applying this scheme, the researchers found that 9-month-old tau/ApoE4 were predominantly stage 3 and 4, while both tau/ApoE knockouts and microglial knockouts were stage 1.

Might microglia harm neurons by raising expression of ApoE in other cells? Indeed, the authors found that depleting microglia caused astrocytes and, to a lesser extent, a subset of neurons in the hippocampus, to pump out more ApoE; however, since these mice were protected from neurodegeneration, the finding suggested that in the absence of microglia, ApoE was rendered toothless.

And how was ApoE affecting microglia? Shi found that it did not change their number, but did increase their expression of CD68, a lysosomal marker of microglial activation. In contrast, neither astrocyte number nor activation were affected by expression of ApoE or by the presence/absence of microglia. These findings suggested that ApoE4 likely kills neurons in the tauopathy mice by altering microglial function.

While Holtzman’s paper was in press, another study, led by Renzo Mancuso of the University of Southampton, U.K., reported similar data. These researchers used another CSF1R inhibitor, JNJ-527, to partially deplete microglia and prevent their proliferation (see Mancuso et al., 2019). In the spinal cords of P301S mice, JNJ-527 docked microglial numbers by more than half in the spinal cord. This assuaged neuroinflammation there, reduced phospho-tau, and spared motor neurons.

Kim Green of the University of California, Irvine, called the effects of removing microglia “stunning.” He said that the finding builds on the emerging theme that microglia play a critical role in neurodegeneration. Green, who collaborated with Plexxikon in Berkeley, California, to develop PLX3397, previously reported that microglial depletion prevented the development of Aβ plaques in the 5xFAD model of amyloidosis (Sep 2019 news), while depleting microglia after plaques had already formed slowed synapse loss around dystrophic neurites (Spangenburg et al., 2016). The Glabe lab at UCI reported similar results (Mar 2018 news). Together with the present work, these studies suggest microglia can exacerbate different types of proteinopathy. Exactly how ApoE mediates the damage microglia inflict in response to different types of pathology and at different stages of disease is now a critical line of research, Green said.

Holtzman agreed, pointing out that he has not nailed down the source of the ApoE that promotes microglia-mediated neurodegeneration. Butovsky believes it is microglial. Conditional microglial ApoE knockouts could confirm that. Butovsky added that non-pharmacological approaches, such as conditional knockout of CSF1R, are needed to deplete microglia without confounding effects on other cell populations. Indeed, Holtzman and colleagues found that PLX3397 altered blood cell counts, including the number of red blood cells, monocytes, and T cells. While the inhibitor has also been reported to bind CSF1R on neurons, the researchers do not believe this explains the protection in tau/ApoE4 mice, since PLX3397 had a detrimental effect on hippocampal neurons in culture, preventing neurite growth.

Susanne Krasemann of University Medical Center in Hamburg, Germany, said that the findings dovetail with previous work implicating ApoE and TREM2 as flippers of a microglial neurodegenerative switch (Krasemann et al., 2017). Because microglial depletion boosts astrocytic ApoE expression without harming neurons, preventing aberrant microglial responses, rather than lowering ApoE levels per se, could be a therapeutic strategy for neurodegenerative disease, she said.

Holtzman and Butovsky agree it will be important to consider disease stage when targeting microglia. For example, they might exercise important protective functions early on but deliver a toxic blow to neurons in later stages (May 2016 newsJan 2019 news). 

Another unknown is whether tau pathology harms neurons directly, or if activated microglia are responsible for all the damage, Holtzman said. It is clear from previous studies that accumulation of phospho-tau incites harmful microglial responses, including induction of the complement cascade (Jul 2018 conference news; Aug 2019 news). Holtzman speculated that low amounts of phospho-tau, as seen in early stages of disease, trigger microglia, and the inflammation they subsequently unleash drives the lion’s share of the ensuing neurodegeneration. Of course, activated microglia also drove p-tau accumulation itself. Holtzman proposed that this progression could reflect a secondary response to the neuronal damage inflicted by the activated microglia.—Jessica Shugart


  1. Both the Holtzman lab and our lab explored the role of microglia in mediating tau-induced neurodegenerative phenotypes in mice. Here, Shi et al. present exciting results of marked neuroprotection in a tau model after microglial ablation. In a different tau model, we did not observe strong changes in tau, astrocytic activation, or neurodegeneration markers following modest microglial reduction. In comparison, the Holtzman laboratory obtained a dramatic, near-total reduction of microglia with their treatment, likely due to differences in the PLX dose that was used, and possibly indicating a threshold phenomenon of microglia themselves that results in this protective effect.

    Congrats to the Holtzman lab on this intriguing observation. It will be fascinating to better understand how complex neuronal-glial interplay impacts neurodegeneration across multiple domains.

  2. Dissecting the contribution of p-tau itself versus the innate immune system to neuronal loss in a model of tau pathology, this new study from the lab of David Holtzman showed that neuronal pathology and loss is almost exclusively driven by microglia. Moreover, the effect of ApoE on neurodegeneration is dependent on microglia, while absolute amount of ApoE is not the only key to neurodegeneration. Rather, the source and location of ApoE, in addition to the quantity, matters for neurodegeneration. The authors speculate that the effect of ApoE may predominantly result from regulating microglial function. Interestingly, depletion of microglia led to increased expression of ApoE in astrocytes and neuronal subtypes without affecting neuronal survival! This is in accordance with our study where we showed that ApoE upregulation associated with a highly dysregulated microglia phenotype in disease and with decreased neuronal survival. We and others confirmed that upregulation of ApoE expression also plays a role in human microglia in disease.

    When investigating microglia functionality in the neurodegenerative brain, we already showed that loss of homeostatic signature seems to be more important for microglia dysfunction than aberrant activation. This concept was supported by this study of Shi et al.: They found that the activation status, rather than the absolute number of microglia, is functionally associated with neurodegeneration. Although ApoE deficiency is promoting proliferation of microglia, Shi et al. could show that ApoE-deficiency preserved the homeostatic microglia signature and neuronal survival in their model of tauopathy, a key finding that we already identified in our study using different disease models.

    In conclusion, microglia are the driving force of neuronal loss in tauopathy. Moreover, ApoE’s effect on neurodegeneration is mediated almost exclusively by microglia. Therefore, restoring microglial function rather than lowering general amounts of ApoE in the brain might be a suitable therapeutic strategy in the future.

  3. Microglia and neuroinflammation play a central role in the pathogenesis of Alzheimer’s disease, as highlighted by several genome-wide association studies (GWAS). While the role of microglia has been extensively studied within the context of multiple Aβ models, there are still some discrepancies about their exact contribution to pathology. On one hand, partial ablation of microglia reduced synaptic pathology and attenuated cognitive deficits (Olmos-Alonso et al., 2016; Dagher et al., 2015; Spangenberg et al., 2016) and even prevented Aβ plaque formation (Sosna et al., 2018; Spangenberg et al., 2019). On the other hand, loss of function of TREM2 stopped microglial activation and exacerbated plaque pathology, suggesting that switching microglia on may be a potential strategy to clear Aβ (Parhizkar et al., 2019). 

    However, the link between microglia and tau-induced neurodegeneration has remained elusive, until now. Both Holtzman's and our lab explored independently the role of microglia in tau pathology using two complementary approaches, resulting in either almost complete or only partial depletion (Mancuso et al., 2019) of microglia in mice overexpressing the human P301S mutated tau gene. Overall, our studies show that ablation of microglia using colony-stimulating factor-1 receptor (CSF1R) inhibitors reduced tau phosphorylation and aggregation, and preserved neurons. Interestingly, these effects were achieved not only by reducing microglial numbers but also by pushing cells into a homeostatic phenotype.

    This clearly indicates that microglial activation is a driving force of neuronal death in tauopathy. While it is encouraging to see that both our and the Holtzman labs’ results go in the same direction, they suggest a bit of a conundrum if we were to develop treatments targeting microglia for Alzheimer’s disease. Under the assumption that microglia are beneficial in the context of amyloid pathology (despite strong controversy around this aspect), boosting microglia to clear plaques will exacerbate tau pathology and neuronal death. 


    . Colony-stimulating factor 1 receptor inhibition prevents microglial plaque association and improves cognition in 3xTg-AD mice. J Neuroinflammation. 2015 Aug 1;12:139. PubMed.

    . CSF1R inhibitor JNJ-40346527 attenuates microglial proliferation and neurodegeneration in P301S mice. Brain. 2019 Oct 1;142(10):3243-3264. PubMed.

    . Pharmacological targeting of CSF1R inhibits microglial proliferation and prevents the progression of Alzheimer's-like pathology. Brain. 2016 Mar;139(Pt 3):891-907. Epub 2016 Jan 8 PubMed.

    . Loss of TREM2 function increases amyloid seeding but reduces plaque-associated ApoE. Nat Neurosci. 2019 Feb;22(2):191-204. Epub 2019 Jan 7 PubMed.

    . Sustained microglial depletion with CSF1R inhibitor impairs parenchymal plaque development in an Alzheimer's disease model. Nat Commun. 2019 Aug 21;10(1):3758. PubMed.

    . Eliminating microglia in Alzheimer's mice prevents neuronal loss without modulating amyloid-β pathology. Brain. 2016 Apr;139(Pt 4):1265-81. Epub 2016 Feb 26 PubMed.

    . Early long-term administration of the CSF1R inhibitor PLX3397 ablates microglia and reduces accumulation of intraneuronal amyloid, neuritic plaque deposition and pre-fibrillar oligomers in 5XFAD mouse model of Alzheimer's disease. Mol Neurodegener. 2018 Mar 1;13(1):11. PubMed.

Make a Comment

To make a comment you must login or register.


Research Models Citations

  1. hTau.P301S
  2. 5xFAD (B6SJL)

News Citations

  1. ApoE and Tau: Unholy Alliance Spawns Neurodegeneration
  2. ApoE4 Makes All Things Tau Worse, From Beginning to End
  3. Changing With the Times: Disease Stage Alters TREM2 Effect on Tau
  4. ApoE and Trem2 Flip a Microglial Switch in Neurodegenerative Disease
  5. Are Microglia Plaque Factories?
  6. Wiping Out Microglia Prevents Neuritic Plaques
  7. Barrier Function: TREM2 Helps Microglia to Compact Amyloid Plaques
  8. Without TREM2, Plaques Grow Fast in Mice, Have Less ApoE
  9. Synaptic Tau Clangs the Dinner Bell for Hungry Microglia
  10. Nixing Complement Protein Protects Neurons in Tauopathy Model

Paper Citations

  1. . TREM2 deficiency attenuates neuroinflammation and protects against neurodegeneration in a mouse model of tauopathy. Proc Natl Acad Sci U S A. 2017 Oct 24;114(43):11524-11529. Epub 2017 Oct 9 PubMed.
  2. . ApoE4 markedly exacerbates tau-mediated neurodegeneration in a mouse model of tauopathy. Nature. 2017 Sep 28;549(7673):523-527. Epub 2017 Sep 20 PubMed.
  3. . CSF1R inhibitor JNJ-40346527 attenuates microglial proliferation and neurodegeneration in P301S mice. Brain. 2019 Oct 1;142(10):3243-3264. PubMed.
  4. . Eliminating microglia in Alzheimer's mice prevents neuronal loss without modulating amyloid-β pathology. Brain. 2016 Apr;139(Pt 4):1265-81. Epub 2016 Feb 26 PubMed.
  5. . The TREM2-APOE Pathway Drives the Transcriptional Phenotype of Dysfunctional Microglia in Neurodegenerative Diseases. Immunity. 2017 Sep 19;47(3):566-581.e9. PubMed.

Further Reading

Primary Papers

  1. . Microglia drive APOE-dependent neurodegeneration in a tauopathy mouse model. J Exp Med. 2019 Nov 4;216(11):2546-2561. Epub 2019 Oct 10 PubMed.