Subscribe to eTOC

Microglia Deletion Protects Against Tau Pathology in Alzheimer's Mice

Article In Brief

Microglia are the driving force of neurodegeneration in a tauopathy mouse model, a new animal study found. The authors of the report suggest microglia may be an effective therapeutic target for preventing disease progression in the setting of tauopathy.


No caption available.

Removal of microglia prevents tau-induced neuronal damage in a model of Alzheimer's disease (AD), according to a new study published in the October 10 online issue of the Journal of Experimental Medicine.

“If this finding in the animal is also occurring in humans, it suggests that while the accumulation of abnormal forms of tau does lead to neurodegeneration, it requires microglia for the damage [to occur],” said the senior study author David M. Holtzman, MD, FAAN, professor and chair of neurology at Washington University School of Medicine in St. Louis.

The question of the interacting roles of amyloid, tau, and microglia in provoking neurodegeneration is a longstanding one in AD, Dr. Holtzman said.

“We know that amyloid-beta accumulates perhaps 20 years before symptom onset, and that maybe three to four years before the onset of memory decline, tau pathology begins to spread out of limited areas in the medial temporal lobe into the neocortex. When that happens, that generally correlates with damage to the brain and the cognitive symptoms and signs we see in patients.

“We also know that inflammatory cells in the brain, microglia in particular, begin to become activated as a consequence of these two pathologies, and it looks like this activation is linked to disease progression. So what we wanted to know is whether the microglial activation and resulting inflammation is causative in the damage to the brain.”

Previous studies have asked the same question, Dr. Holtzman said, but newer tools have emerged—in particular, inhibitors of a microglial surface receptor called CSF1—to make the experiments more directly informative.

“CSF1 is a key factor that enables microglial cells to survive, so the inhibitor eliminates most of the microglial cells from the brain.”

Independent experts said the findings, though not surprising, offer important insights into what is driving the neurodegeneration in AD, with downstream potential for identifying new therapeutic targets.

Study Design, Findings

To look more closely at the process underlying the neurodegeneration, the research team fed the CSF1 inhibitor, LX-3397, to mice bearing a mutation in a human tau gene that were either homozygous for human APOE4 or had no expression of their own version of the protein. They began the feeding at six months of age, “right at the time the mice begin to develop tau pathology,” Dr. Holtzman explained.

They used a dose that was higher than that used in other recent studies that had shown very limited depletion of microglial cells.

In male mice, treatment led to about 90 percent elimination of microglia within seven days, and almost complete elimination by 21 days. For unknown reasons, this effect was incomplete in female mice, and so further experiments were restricted to males. Treatment continued for three months, and mouse brains were examined at 9.5 months.


“Here, the authors have presented very compelling evidence that not only do microglia enhance tau pathology, but that they [also] actively participate in this process of neurodegeneration. This is really exciting, because up until this point, it has been unclear how exactly tau results in neuronal loss. This points to an important neuron-glial mechanism.”—DR. RACHEL E. BENNETT

As expected, mice with no APOE expression remained largely unaffected. Also, as expected, mice bearing both tau and APOE mutations that did not receive the CSF1 inhibitor developed severe brain atrophy, which was consistent with previously observed effects of these transgenes. But mice with the two mutations that received the inhibitor were completely protected from the effects of the mutations—with a complete blockade of neurodegeneration and retention of brain volume.

“By eliminating the microglia, we were able to completely prevent the loss of neurons that occurs with tau pathology,” Dr. Holtzman said.

The researchers had previously shown that removing APOE from the model is also highly protective, “almost as much as eliminating the microglia,” Dr. Holtzman said, “suggesting that the effect of APOE on enhancing degeneration is probably modulated through microglia.”

Overall, these findings place microglia at the forefront of neurodegeneration in AD. From the current study, Dr. Holtzman concluded that while tau may be necessary for damage, it is not sufficient: “As tau accumulates, it is somehow inciting the microglia to become activated, and ultimately that is mediating the damage,” he said.

A paper from another lab, published October 1 in Brain, used a different inhibitor of the same microglial receptor, and came to the same conclusion. The researchers showed that blocking microglial activation reduced tau-induced neurodegeneration. They also showed that the mice were able to perform better on cognitive tests.

The PLX-3397 inhibitor used by Dr. Holtzman, pexidartinib, was approved by the US Food and Drug Administration in August for treatment of symptomatic tenosynovial giant cell tumor, a group of rare connective tissue tumors.

Whether it can be effective in chronic treatment of AD is entirely unknown, but Dr. Holtzman noted that microglia have a number of different receptors that could be targeted, now that the potential for such treatment is becoming clear.

The question of how much inactivation is enough will also be crucial for assessing the viability of a microglial-targeted therapy.

“In some of our animals, there was incomplete inactivation, and there was some damage,” he said.

A 2018 paper in the Journal of Neuroinflammation by Rachel E. Bennett, PhD, and colleagues using the same inhibitor at a dose about three-quarters of that used by Dr. Holtzman showed little protection from tau-induced neurodegeneration.

“Microglia are probably important in prevention of infection in the brain,” Dr. Holtzman said. “How much do we need to inactivate to get a therapeutic effect, and are there ways to do that without completely removing them? We need to sort this out.”

Expert Commentary

“This study provides a very significant insight into what is driving neurodegeneration in this model,” commented Li Gan, PhD, professor of neuroscience and director of the Helen and Robert Appel Alzheimer's Disease Research Institute at Weill Cornell Medicine, where she studies microglia in AD.

“The authors showed a very striking effect of microglia depletion that led to almost complete protection against neurodegeneration.”

The conclusion that tau pathology by itself is not deleterious “is not actually surprising,” she said. “Several aspects of tau have been puzzling” in the context of neurodegeneration models, including that it is normally very highly expressed in neurons. In neuronal culture, even when overexpressed, “tau is rarely toxic, even with disease-causing mutations. But in vivo it is a completely different story,” and there one does observe neurodegeneration.

“One of the biggest differences is that in vivo you have glia, and there is a tremendous microgliosis in the tauopathy models,” Dr. Gan said.

Dr. Bennett, who is an instructor in neurology at Massachusetts General Hospital and Harvard Medical School, said: “We've known for some time that microglia secrete inflammatory factors that seem to exacerbate tau pathology, and we suspected that modifying this function could be beneficial.

“Here, the authors have presented very compelling evidence that not only do microglia enhance tau pathology, but that they [also] actively participate in this process of neurodegeneration. This is really exciting, because up until this point, it has been unclear how exactly tau results in neuronal loss. This points to an important neuron-glial mechanism.”

“We still have a way to go before we can turn these findings into a therapy for Alzheimer's disease,” she cautioned. Along with concerns about how completely and in what way microglia might need to be silenced, the question of when such a therapy could be effective looms large.

In her own work, not only was the dose smaller, but the animals were treated later in their disease course, and little benefit was seen.

Nonetheless, she said, “This work gives us a really clear line of research to ask how APOE and microglia are contributing to neurodegeneration.”


Dr. Holtzman has served on the scientific advisory board or consulted for Genentech, Eli Lilly, and AbbVie. Drs. Gan and Bennett reported no competing interests.

Link Up for More Information

• Shi Y, Mannis M, Long J, et al. Microglia drive APOE-dependent neurodegeneration in a tauopathy mouse model J Exp Med 2019; Epub 2019 Oct 10.
    • Mancuso R, Fryatt G, Cleal M, et al. CSF1R inhibitor JNJ-40346527 attenuates microglial proliferation and neurodegeneration in P301S mice Brain 2019;142(10):3243–3264.
      • Bennett RE, Bryant A, Hu M, et al. Partial reduction of microglia does not affect tau pathology in aged mice J Neuroinflammation 2018;15(1):311.