In brain disease, microglia can be villains or heroes. How to encourage the latter? In the June 10 Nature Neuroscience, researchers led by Veronique Miron at the University of Edinburgh tried by killing off the troublemakers. In a mouse model of demyelination, the authors found that lesions healed only after inflammatory microglia had died by way of necroptosis. When the researchers blocked this type of programmed cell death, remyelination lagged. “Necroptosis of pro-inflammatory microglia is required to generate pro-regenerative microglia,” Miron told Alzforum. She is investigating whether a similar dynamic plays out in neurodegenerative diseases such as Alzheimer’s, where demyelination also occurs.

  • After demyelination, inflammatory microglia die by necroptosis.
  • Healthy cells replenish the dead as axons heal.
  • Blocking necroptosis holds up remyelination.

“With this work, the researchers solve the long-standing mystery of the fate of microglia that become acutely activated during demyelination,” Constanze Depp and Stefan Berghoff at the Max Planck Institute for Experimental Medicine in Göttingen, Germany, wrote to Alzforum (full comment below). “This work certainly provokes thought about how this natural form of microglial depletion can be boosted experimentally to further stimulate remyelination.”

Miron and colleagues study multiple sclerosis (MS), a disease marked by cycles of demyelination and remyelination. They previously reported that microglia around myelin lesions in mouse brain switch from a pro-inflammatory to an anti-inflammatory state as remyelination begins (Miron et al., 2013). To explore how this happens, first author Amy Lloyd compared microglial gene expression in mouse corpus callosum three and 10 days after inducing demyelination by injecting this region with the toxin lysophosphatidyl choline (LPC). She found differences in 1,020 genes. At three days, when demyelination peaked, microglia around the lesion expressed many cell death and pro-inflammatory genes. At 10 days, when remyelination had revved up, they expressed more regeneration and anti-inflammatory genes.

Death Before Remyelination. Two weeks after demyelination, new myelin (red) covers many more axons (green) in cerebellar explants of control mice (left) than mice treated with a necroptosis inhibitor (right). [Courtesy of Lloyd et al., Nature Neuroscience.]

Did the microglia expressing cell-death genes actually die? Corpus callosum microglial numbers fell by half seven days after the researchers injected the LPC. To confirm that cells were dying, the authors applied LPC to cerebellar explants and used live imaging to watch what happened. Hours after the toxin triggered demyelination, microglia rounded up and burst (see video below). Importantly, LPC had no effect on primary microglia cultures, demonstrating it does not directly induce cell death.

Dying microglia expressed markers of necroptosis, but not apoptosis. Necroptosis is a form of cell suicide that occurs under inflammatory conditions. It triggers different pathways than apoptosis, with receptor-interacting protein kinase 1 (RIPK1) being a key mediator (Jun 2005 news). The authors examined two additional mouse models of demyelination, one induced by cuprizone and the other by an autoimmune reaction to myelin oligodendrocyte glycoprotein. In both cases, RIPK1 levels in microglia peaked around the time remyelination began, suggesting that necroptotic death is a consistent feature of myelin repair.

Annihilated. In the 12 hours after LPC was added to mouse cerebellar explants, microglia expressing green fluorescent protein retract processes and rupture. [Courtesy of Lloyd et al., Nature Neuroscience.]

But is this necessary for myelin to repair itself? Lloyd and colleagues added the RIPK1 inhibitor necrostatin-1 to cerebellar explants at the same time as LPC. A day later, these explants contained twice as many pro-inflammatory microglia as did those treated with LPC alone. Remyelination faltered, with the explants producing only half as much new myelin as control explants one and two weeks later (see image above).

The same thing happened in mice; when microglia around a lesion took up necrostatin-1, that region produced half as much new myelin as normal by day 10. This was true even though microglia containing necrostatin-1 mopped up myelin debris as well as did control microglia. Together, the findings suggested that activated microglia can remove debris, but then need to die off to allow new myelin to form. Why that is, is not clear.

The remyelination 10 days after LPS injection was accompanied by a rebound in the number of microglia. Where do these new cells come from? In remyelinating explants, less than 30 percent of repopulating microglia present during remyelination expressed the precursor cell marker nestin. What about the rest? The researchers traced their lineage by inducing expression of red fluorescent protein in Iba1-positive cells before demyelination. During remyelination, 70 to 80 percent of repopulating microglia expressed this label, indicating that most of these cells arose through expansion of the existing pool, rather than from de novo differentiation. Interferon signaling appeared crucial for this proliferation. Its levels rose during remyelination, and an antibody against the interferon receptor suppressed microglial numbers and delayed myelination.

To see if microglial necroptosis aids remyelination in people, the authors examined lesions in postmortem brain tissue from seven people who had died with relapsing-remitting MS. In active, remyelinating lesions, they found higher levels of necroptotic and proliferating microglia than in inactive lesions that do not heal.

The finding implies that stimulating necroptotic death in inactive lesions might nudge them toward remyelination, the authors claim. They are now trying to identify the signals that trigger necroptosis in hopes of finding a therapeutic target. They will also examine whether subpopulations of microglia are particularly prone or resistant to necroptosis. Human microglia are diverse, with studies identifying up to seven distinct types (Dec 2018 newsFeb 2019 news). 

Do the successive death and renewal of microglia play a role in other disorders? Myelin breaks down in Alzheimer’s disease as well, and myelination-related genes are consistently perturbed (May 2019 news). Markers of necroptosis have been found in microglia in the AD brain. In the 5XFAD mouse, they correlate with brain atrophy and cognitive decline, raising the possibility that necroptosis could aid remyelination in AD (Jul 2017 news). 

Could microglial replacement encourage healthier, more phagocytic microglia to surround plaques? The answer is unknown, but in mouse models of amyloidosis, depleting microglia slows neurodegeneration, and has been reported to block plaque formation. Intriguingly, remaining microglia still seem to surround plaques (Apr 2014 news; Spangenberg et al., 2016; Mar 2018 news). Miron and colleagues are collaborating with Josef Priller and Tara Spires-Jones at the Edinburgh Dementia Research Institute to investigate whether microglial necroptosis and repopulation can affect AD pathology. Others are studying RIPK1 inhibitors as treatments for inflammatory and neurodegenerative disorders (Harris et al., 2017). Denali Therapeutics, South San Francisco, has begun Phase1 trials of its RIPK1 inhibitor DNL747 in AD (see and amyotrophic lateral sclerosis ( and is targeting MS. GlaxoSmithKline completed Phase 1/2 trials of GSK2982772 for rheumatoid arthritis ( and psoriasis ( and is testing in a Phase 2 trial for ulcerative colitis (—Madolyn Bowman Rogers





  1. What we’ve learned
    In this elegant study, Lloyd et al. show that a large number of pro-inflammatory microglia during acute demyelination undergo cell death in the form of necroptosis to give way to repopulating residual microglia with pro-remyelinating properties. Conversely, inhibiting microglial necroptosis and thereby clamping microglia in a pro-inflammatory state hindered effective remyelination. With this work, the team of Veronique Miron solves the long-standing mystery regarding the fate of the microglia population that becomes acutely activated during demyelination.

    While experimental microglia depletion has been shown to be beneficial in a variety of brain diseases, this work identifies a naturally occurring form of beneficial microglia, via depletion, which seems to be a very efficient mode of silencing inflammatory microglia gene signatures in vivo. It is tempting to speculate that this is especially true for the highly active microglia digesting myelin that probably have exceeded their lifetime capacity for phagocytosis. This work certainly provokes thought about how this natural form of microglial depletion can be boosted experimentally to further stimulate remyelination. In this context, it would have been interesting to show if the process of natural depletion is dysregulated in chronic active MS lesions. Likewise, it must be clear that the therapeutic window for such an intervention is small: The coordination of depleting pro-inflammatory and increasing pro-regenerative microglia needs to be tightly controlled in order to prevent premature remyelination in a pro-inflammatory environment—a remyelination that is doomed to fail.

    What this means for AD-related demyelination
    In AD, demyelination has been shown to occur focally at Aβ plaques (Behrendt et al., 2013). White-matter damage in general, however, can be detected earlier than overt plaque pathology (e.g Agosta et al., 2011). Recently, two high-profile single-cell/nuclei transcriptomic studies in patients consistently identified myelination-related processes to be perturbed in AD (Mathys et al., 2019; Del-Aquila et al., 2019). With this, oligodendrocytes (which are yet the "forgotten players" in the context of AD) might quickly move into focus—and with them remyelinating therapies. Microglia depletion has been already suggested as potential therapeutic intervention in AD as it was shown to combat cognitive decline and neurodegeneration in aged and AD mice (Spangenberg et al., 2016; Elmore et al., 2018) and if applied at pre-plaque stages even Aβ plaque deposition (Sosna et al., 2018). 

    In this context, the study by Lloyd et al. contributes to our understanding of how microglial responses are tuned by depletion to enable efficient remyelination. When finding the correct time window (a caveat mentioned above) to eliminate pro-inflammatory microglia cells, experimental microglia depletion might be all the more attractive as a therapeutic intervention in AD, benefiting both remyelination and neuroprotection. Intriguingly, comparison of the post-demyelination microglial gene signature with the disease-associated microglia (DAM) gene signature (Karen-Shaul et al., 2017) revealed that DAMs are more similar to pro-regenerative than to demyelinating pro-inflammatory microglia. This finding awaits single-cell transcriptomic proof. Nevertheless, this would suggest that DAMs partly show a myelination-boosting gene signature and it is tempting to speculate that the switch to this pro-remyelinating microglia comes too early in AD and remyelination is therefore somewhat inefficient.

    Altogether, we foresee extensive research on the role of demyelination in AD and the evaluation of pro-remyelinating agents to combat AD-related demyelination in the near future. With the prominent and well-studied role of microglia in AD, this study by Lloyd et al. outlines a potential direction of this research very worth pursuing, namely the interplay of microglia states and remyelination. Again, we congratulate the authors for this excellent paper. 


    . Dynamic changes in myelin aberrations and oligodendrocyte generation in chronic amyloidosis in mice and men. Glia. 2013 Feb;61(2):273-86. PubMed.

    . White matter damage in Alzheimer disease and its relationship to gray matter atrophy. Radiology. 2011 Mar;258(3):853-63. PubMed.

    . Single-cell transcriptomic analysis of Alzheimer's disease. Nature. 2019 Jun;570(7761):332-337. Epub 2019 May 1 PubMed.

    . A single-nuclei RNA sequencing study of Mendelian and sporadic AD in the human brain. Alzheimers Res Ther. 2019 Aug 9;11(1):71. 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.

    . Replacement of microglia in the aged brain reverses cognitive, synaptic, and neuronal deficits in mice. Aging Cell. 2018 Dec;17(6):e12832. Epub 2018 Oct 2 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.

    . A Unique Microglia Type Associated with Restricting Development of Alzheimer's Disease. Cell. 2017 Jun 15;169(7):1276-1290.e17. Epub 2017 Jun 8 PubMed.

  2. A mechanism of microglia cell-state transition at different stages of injury

    Microglia change their state, including physical or biochemical changes, to adapt to specific brain environments. Until recently these changes have been difficult to characterize, and were mostly grouped together using the term “activation.” It is becoming increasingly apparent that microglia assume many diverse states in the healthy (Hammond et al., 2019; Li et al., 2019) and diseased brain (Keren-Shaul et al., 2017; O’Koren et al., 2019; Mathys et al., 2017; Hammond et al., 2019), raising the important question of whether and how microglia transition between different cell states and the roles of each state/subpopulation in the brain.

    Dr. Miron’s previous work showed that over the course of demyelinating injury, microglia transition from an inflammatory to regenerative state, a change that is required for normal tissue repair (Miron et al., 2013). In this study, Lloyd and colleagues use genomics, ex vivo explant assays, and in vivo manipulation of different microglia states to understand how this transition is regulated. Surprisingly, the authors show that the microglia do not transition between the two states, but rather the inflammatory microglia die by necroptosis to clear the way for the pro-regenerative subpopulation.

    These findings are important for several reasons: 1.) The ability of microglia to shift between inflammatory and regenerative activation states in disease could be limited. 2.) Once microglia become activated, it might be difficult for them to transition back to a resting state without dying. In AD, it is still unclear whether microglia occupy the different states identified in this study, particularly the inflammatory subpopulation that dies by necroptosis. The most well-known state, disease-associated microglia (DAM) (Keren-Shaul et al., 2017) associate with amyloid plaques and have a distinct profile that aligns partly with the activated microglia profiles found in the demyelination model used in this study but largely lacks the interferon signature found in the pro-regenerative microglia (see comparison in Hammond et al., 2019). Direct comparisons between several models of neurodegeneration will be needed to tease apart these differences.

    In AD it has been difficult to determine how and when microglia become activated and whether they occupy unique states at different stages of disease, as found in this study. Most AD mouse models are slow progressing and pathology accumulates at varying rates in different brain regions, making it difficult to address this question. However, an inducible model of neurodegeneration showed that following neuron death microglia occupy several distinct states similar to those found in the demyelination model (Mathys et al., 2017). Therefore, it is possible that microglia necroptosis also plays a role in AD, but this will need to be examined in greater detail.

    The development of new tools to track and manipulate resident microglia in the brain is needed to understand how and when microglia become “activated,” change state, and tailor their responses to specific disease conditions. This study provides an important leap forward in our understanding of microglia biology and functional states that could be broadly relevant to Alzheimer’s and other diseases.


    . Temporal Tracking of Microglia Activation in Neurodegeneration at Single-Cell Resolution. Cell Rep. 2017 Oct 10;21(2):366-380. PubMed.

    . A Unique Microglia Type Associated with Restricting Development of Alzheimer's Disease. Cell. 2017 Jun 15;169(7):1276-1290.e17. Epub 2017 Jun 8 PubMed.

    . Single-Cell RNA Sequencing of Microglia throughout the Mouse Lifespan and in the Injured Brain Reveals Complex Cell-State Changes. Immunity. 2019 Jan 15;50(1):253-271.e6. Epub 2018 Nov 21 PubMed.

    . Developmental Heterogeneity of Microglia and Brain Myeloid Cells Revealed by Deep Single-Cell RNA Sequencing. Neuron. 2019 Jan 16;101(2):207-223.e10. Epub 2018 Dec 31 PubMed.

    . M2 microglia and macrophages drive oligodendrocyte differentiation during CNS remyelination. Nat Neurosci. 2013 Sep;16(9):1211-1218. Epub 2013 Jul 21 PubMed.

    . Microglial Function Is Distinct in Different Anatomical Locations during Retinal Homeostasis and Degeneration. Immunity. 2019 Mar 19;50(3):723-737.e7. Epub 2019 Mar 5 PubMed.

    View all comments by Beth Stevens

Make a Comment

To make a comment you must login or register.


News Citations

  1. A New Program for Cell Death: Necroptosis Premiering in a Neuron Near You
  2. Local Flavor: At Protein Level, Too, Human Microglia Are Diverse
  3. Single-Cell Profiling Maps Human Microglial Diversity, Flexibility
  4. When It Comes to Alzheimer’s Disease, Do Human Microglia Even Give a DAM?
  5. Necroptosis Rampant in the Alzheimer’s Brain?
  6. Microglial Magic: Drug Wipes Them Out, New Set Appears
  7. Wiping Out Microglia Prevents Neuritic Plaques

Paper Citations

  1. . M2 microglia and macrophages drive oligodendrocyte differentiation during CNS remyelination. Nat Neurosci. 2013 Sep;16(9):1211-1218. Epub 2013 Jul 21 PubMed.
  2. . 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.
  3. . Discovery of a First-in-Class Receptor Interacting Protein 1 (RIP1) Kinase Specific Clinical Candidate (GSK2982772) for the Treatment of Inflammatory Diseases. J Med Chem. 2017 Feb 23;60(4):1247-1261. Epub 2017 Feb 10 PubMed.

External Citations


Further Reading

Primary Papers

  1. . Central nervous system regeneration is driven by microglia necroptosis and repopulation. Nat Neurosci. 2019 Jul;22(7):1046-1052. Epub 2019 Jun 10 PubMed.