Microglia, the brain’s janitorial crew, boast a variety of subtypes, taking on different activation states to handle specialized tasks. In the Feb 11 Neuron, researchers led by Mikael Simons at Technical University Munich and Ozgun Gokce at the University Hospital of Munich introduced a new one. In old mice, whose myelin breaks down, some microglia near fraying white matter engulfed and digested myelin debris. To do so, these cells, which the authors dubbed white-matter-associated microglia (WAM), revved up expression of phagocytosis and lysosomal genes. Overall, their gene-expression profile resembled that of disease-associated microglia (DAMs) found around amyloid plaques. Nonetheless, there were differences; for example, unlike DAMs, induction of WAMs did not require APOE. WAMs also had a greater effect on lipid metabolism than did DAMs.

  • Microglia in white matter of aging mice assume a quasi-DAM phenotype.
  • Dubbed WAMs, these cells mop up crumbling myelin.
  • WAMs also appear in mouse models of demyelination and amyloidosis.

The same microglial subtype appeared in young mice that had a demyelinating condition, implying that WAMs arise as a specific response to myelin breakdown. The cells also cropped up, along with DAMs, in three different mouse models of amyloidosis. Curiously, in these mouse models, WAMs did depend on APOE, bringing them even closer to DAMs.

“We think WAMs represent a partial DAM response,” Simons told Alzforum. Because the cells activate during normal aging in mice, he believes they might contribute to age-related diseases such as Alzheimer’s. It is unclear, however, if the same subtype occurs in human brains.

Janitor on the Job. Microglia (red) in the white matter of aging mice sweep up myelin debris (green, arrows; close-up on right). Nuclei are blue. [Courtesy of Safaiyan et al., Neuron.]

“This is an important paper,” said Oleg Butovsky at Brigham and Women’s Hospital, Boston. He agreed that WAMs are a subset of DAMs, and was particularly intrigued that they were able to activate without APOE in aging brain. “This gives us new avenues to explore in terms of contributions to health and disease,” he told Alzforum.

Several previous studies have catalogued microglia in mouse brain, finding numerous subtypes (Jul 2018 conference news; Dec 2018 news). For Alzheimer’s researchers, the most famous subtypes are DAMs and the very similar MGnD, which associate with neurodegeneration (Jun 2017 news; Sep 2017 news). Researchers do not yet know what most of the other subtypes do.

Simons focused on white-matter microglia because he studies myelin. This insulating material degrades with age, forming small lesions visible by MRI. Simons wondered how this degeneration would affect microglia in those areas. Joint first authors Shima Safaiyan at Technical University Munich and Simon Besson-Girard at University Hospital of Munich isolated microglia from 18- to 20-month-old wild-type mice, comparing cells from frontal cortex gray matter with those from corpus callosum white matter. The authors analyzed gene expression by single-cell RNA-Seq, finding four distinct profiles. Two of the microglial subtypes expressed mainly homeostatic genes and were the dominant subtypes in both gray and white matter. The other two were specific to white matter. Of these, one, which the authors called activated microglia, ramped up metabolic genes such as those involved in ribosomal and mitochondrial function. The other subtype was dubbed WAM. They boosted expression of a suite of DAM genes, including APOE, CST7, CLEC7A, and CTSB. WAMs were absent from the brains of 4-month-old mice, confirming they are induced with age.

The authors repeated the findings in a larger sample of 10,599 microglia isolated from 17 2-year-old mice, again identifying the same four subtypes in white matter. In addition, the authors analyzed older scRNA-Seq mouse microglial datasets from Bart De Strooper’s group at the KU Leuven, Belgium, and Beth Stevens’ group at Boston Children’s Hospital, Massachusetts (Apr 2019 conference news; Dec 2018 news). Simons and colleagues identified the WAM signature in these datasets, and found these cells to become more numerous the older the mice were.

“It was surprising there was a very specific microglial response to aging in the white matter,” Simons noted. He believes WAMs are likely beneficial, protecting the brain by eliminating garbage, but this remains to be proven.

Histopathology supported the idea that these cells were related to myelin clearance. In the corpora callosa of 2-year-old mice, about 20 percent of microglia were WAMs, with large cell bodies and thick processes. They tended to cluster in small clumps of three to five cells, which the authors termed nodules (see image above). Electron microscopy revealed the microglia to be stuffed with myelin debris, and the myelin sheaths around the nodules were breaking down.

If WAMs are a response to degenerating white matter, not a cause of it, then demyelination might induce the cells in younger mice. The authors examined a mouse model of the demyelinating disorder Pelizaeus-Merzbacher Disease (PMD; Readhead et al., 1994). Sure enough, by 2 months of age, when the myelin sheaths of PMD mice began to crumble, microglial nodules appeared in the corpora callosa. These cells stained for numerous WAM markers, such as CLEC7A, AXL, and LGALS3.

Other disease models show this, too. Take amyloidosis. The authors found WAMs in 6-month-old 5XFAD mice. In the RNA-Seq data from de Strooper’s lab, the authors identified WAMs in APP/PS1 and APP NL-G-F mice. All of these amyloidosis models contained DAMs as well; intriguingly, though, WAMs appeared first, popping up at about 3 months of age in the knock-ins, for example.

Garbage Piles. Electron microscopy reveals myelin debris (arrows) in white matter of aging TREM2 knockouts (right) but not age-matched controls (left). [Courtesy of Safaiyan et al., Neuron.]

What controls the formation of WAMs? Because TREM2 is required for DAMs, the authors analyzed microglia from 18-month-old TREM2 knockout mice. They found no WAMs or microglia nodules in their corpora callosa. White matter contained more extracellular myelin clumps than in age-matched controls, and nearby microglia were more engorged with debris, hinting at a failure to digest the scraps (image at left). Supporting this, cultured primary microglia from TREM2 knockouts were able to engulf myelin pieces, but not degrade them. The data suggest that TREM2 is essential for WAM induction and function, as it is in DAMs.

ApoE was a more complicated matter. While this apolipoprotein is essential for the DAM signature, aging APOE knockouts had normal numbers of WAMs. Curiously, however, when APP/PS1 mice were crossed with APOE knockouts, their offspring generated few WAMs. The authors do not know why APOE has different effects in these varied contexts. Possibly, ApoE acts indirectly in amyloidosis models, through amyloid plaques or some other aspect of pathology, Simons suggested. Butovsky noted that it would be worth examining conditional knockouts that lose ApoE expression only in microglia, to find out whether the effects are cell-autonomous or not.

Butovsky questioned whether WAMs arise from endogenous brain microglia or from infiltrating peripheral macrophages, noting that the corpus callosum lies just over the lateral ventricles, which act as a gateway to the brain for immune cells.

The big question is whether the findings apply to people. This is hard to answer, because most human studies use postmortem tissue, which is often of low quality, Simons said. One scRNA-Seq study of living microglia taken from human brain biopsies did find differences between those in white and gray matter (Nov 2019 news). But such tissue is rare. Simons wants to explore the issue using chimeric mouse models that contain transplanted human microglia, as created by De Strooper’s and Matthew Blurton-Jones’ labs at University of California, Irvine (Apr 2019 conference news; Aug 2019 news). 

Nicola Fattorelli, KU Leuven, and Renzo Mancuso, now at VIB-UAntwerp Center for Molecular Neurology, agreed this approach could be helpful. “Corroborating these findings in stem-cell-derived microglia in vivo with xenotransplantation models would also be an important step forward in deciphering the complexity of human microglia biology,” they wrote to Alzforum.

Richard Ransohoff of Third Rock Ventures, Boston, noted that mice are poor models for human disease. In particular, mouse white matter and the aging process are different than in people. “Mice exhibit known major shortfalls in regard to the biology under consideration here,” he wrote to Alzforum. “As a drug developer, I’d much prefer to see such scientific talent applied in a way that is more likely to generate therapeutic progress.”

If WAMs exist in human brain, how would they affect therapies that target microglia? Because both WAMs and DAMs depend on TREM2, strategies that activate this receptor might promote beneficial microglial activity in both gray and white matter, Simons believes. For example, he and others have found that stimulating TREM2 promotes myelin clearance in mouse models of demyelination (Takahashi et al., 2007Cignarella et al., 2020Bosch-Queralt et al., 2021).—Madolyn Bowman Rogers

Comments

  1. The efforts of several groups that have recently applied single-cell transcriptomics to characterize microglial populations in physiology and disease have converged to the description of microglial reactive phenotypes with overlapping signatures across development, aging, and various disease-related scenarios (Keren-Shaul et al., 2017; Friedman et al., 2018; Hammond et al., 2019; and others). In Alzheimer’s disease mouse models, homeostatic microglia undergo a progressive phenotypic transition toward a phagocytic-prone disease-associated state that was named DAM/MGnD/ARM, and that depends on TREM2/APOE signaling (Keren-Shaul et al., 2017; Krasemann et al., 2017; Sala Frigerio et al., 2019). 

    In this interesting work, Safaiyan et al. describe an aging-associated reactive microglial phenotype that is identified in the white matter of the aged mouse brain, referred to by the authors as white-matter-associated microglia (WAMs). WAMs share a gene-expression signature with DAMs, and like DAMs require TREM2 signaling for their phenotypic transition. The authors reanalyzed previously published scRNA-Seq datasets of mouse microglia in aging and disease (Sala Frigerio et al., 2019; Hammond et al., 2019), finding again the WAM signature. Surprisingly, WAMs do not depend on APOE in normal aging but cannot form in the absence of APOE in a mouse model of amyloidosis.

    The finding of a DAM-like phenotype in white matter in aging and disease is important for understanding the full spectrum of microglial physiology. While accumulation of DAM-like cells was already shown in the aged mouse brain by our and other labs (Sala Frigerio et al., 2019; Hammond et al., 2019), this study shows their presence and specific peculiarities in white matter, where these cells operate their known surveillance activity by forming nodule structures to clear degenerated myelin.

    Based on several lines of evidence, it is conceivable that WAMs are a specialized subtype of DAMs, e.g., their substantial overlap in signature, their increased phagocytic function, the dependency on TREM2 signalling, and the lack of unique WAM markers when compared to DAM. A similar transcriptomic signature the authors found in our dataset, which originated from dissected gray matter areas, also indicates strong similarity (Sala Frigerio et al., 2019). Additionally, the bioinformatic analysis of Safaiyan et al. suggests that WAMs may be an intermediate state in the trajectory toward full DAM activation, but further investigation is needed to characterize this conversion and how it may occur spatially. On the other hand, the authors also report that WAM forms normally in the absence of APOE in the aged brain, while ApoE is necessary in AD mice similarly to DAMs, suggesting these are two separate entities. This remains an interesting open question to be elucidated in the future.

    Finally, lipid-droplet-accumulating microglia (LDAMs) recently have been described to accumulate with age in mouse and human brains (Marschallinger et al., 2020). This phagocytosis-deficient phenotype was described to be proinflammatory and to produce high levels of reactive oxygen species. It would be of interest to test whether LDAMs, which were reported to express a peculiar signature differing from DAMs, share part of their gene profile with WAMs or independently co-exist in the aging brain.

    Recent large-scale, single-nuclei RNA-Seq showed human microglia to have a more complex heterogeneity than their murine counterparts (Gerrits et al., 2021), and we and others found strong species-specific differences with unique expression of genes related to neurodegeneration in human microglia (Geirsdottir et al., 2019; Mancuso et al., 2019). Thus, corroborating these findings in stem-cell derived microglia in vivo with xenotransplantation models would also be an important step forward in deciphering the complexity of human microglia biology (Mancuso et al., 2019Fattorelli et al., 2021). 

    References:

    . 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.

    . Diverse Brain Myeloid Expression Profiles Reveal Distinct Microglial Activation States and Aspects of Alzheimer's Disease Not Evident in Mouse Models. Cell Rep. 2018 Jan 16;22(3):832-847. 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.

    . The TREM2-APOE Pathway Drives the Transcriptional Phenotype of Dysfunctional Microglia in Neurodegenerative Diseases. Immunity. 2017 Sep 19;47(3):566-581.e9. PubMed.

    . The Major Risk Factors for Alzheimer's Disease: Age, Sex, and Genes Modulate the Microglia Response to Aβ Plaques. Cell Rep. 2019 Apr 23;27(4):1293-1306.e6. PubMed.

    . Lipid-droplet-accumulating microglia represent a dysfunctional and proinflammatory state in the aging brain. Nat Neurosci. 2020 Feb;23(2):194-208. Epub 2020 Jan 20 PubMed. Correction.

    . Distinct amyloid-β and tau-associated microglia profiles in Alzheimer's disease. Acta Neuropathol. 2021 May;141(5):681-696. Epub 2021 Feb 20 PubMed.

    . Cross-Species Single-Cell Analysis Reveals Divergence of the Primate Microglia Program. Cell. 2019 Dec 12;179(7):1609-1622.e16. PubMed.

    . Stem-cell-derived human microglia transplanted in mouse brain to study human disease. Nat Neurosci. 2019 Dec;22(12):2111-2116. Epub 2019 Oct 28 PubMed.

    . Stem-cell-derived human microglia transplanted into mouse brain to study human disease. Nat Protoc. 2021 Feb;16(2):1013-1033. Epub 2021 Jan 11 PubMed.

  2. Safiyan, Besson-Girard, Gokce, Simons et al. identified white-matter-associated microglia named WAMs, which are an aging-specific microglial population. WAMs handle phagocytosis of demyelinating cellular debris, contributing to efferocytosis of apoptotic neurons (as shown with phosphatidylserine (PS)-receptor dependent engulfment) found in white matter of wild-type mice (Márquez-Ropero et al., 2020).

    The team beautifully dissected subpopulations of microglia in transcriptomic and spatial manners and resolved a spectrum of microglia present in normal aging and diseases. In normal aging WT mice, 10 to 15 percent of microglia are WAMs, which share signatures and functions (lipid metabolism, phagocytosis, MHC class II gene activation) with previously identified disease-associated microglia (DAMs). WAMs are Trem2-dependent like DAMs, but ApoE-independent unlike DAMs. In aged AD mice, both WAMs and DAMs (20 percent and 40 percent, respectively, at 12 months) co-exist, depending on Trem2 and Apoe.

    Interestingly, the authors showed that there are no differences in myelin binding and uptake between WT and Trem2 knockout microglia, likely because multiple PS receptors can engulf myelin debris. However, the degradation is reduced in Trem2 KOs due to reduced lysosomal proteins, e.g. Cathepsin L, indicating that Trem2 is critical for digestion of cellular debris.

    From a therapeutic perspective, since WAMs appear before DAMs, targeting WAMs at an earlier Alzheimer’s disease stage could kill two birds with one stone, tackling effects of both aging and disease. Specifically improving Trem2 activity, on par with ongoing efforts in the field (Price et al., 2020; Schlepckow et al., 2020), or enhancing microglial digestion, are other strategies to enhance WAMs. Although the APOE dependency in AD and replications of the findings in humans have to be further explored, the authors’ finding of an aging-specific signature in microglia would impact broadly many age-associated diseases and open doors for targeted therapeutics.

    References:

    . Microglial Corpse Clearance: Lessons From Macrophages. Front Immunol. 2020;11:506. Epub 2020 Mar 27 PubMed.

    . Therapeutic Trem2 activation ameliorates amyloid-beta deposition and improves cognition in the 5XFAD model of amyloid deposition. J Neuroinflammation. 2020 Aug 14;17(1):238. PubMed.

    . Enhancing protective microglial activities with a dual function TREM2 antibody to the stalk region. EMBO Mol Med. 2020 Apr 7;12(4):e11227. Epub 2020 Mar 10 PubMed.

Make a Comment

To make a comment you must login or register.

References

News Citations

  1. A Delicate Frontier: Human Microglia Focus of Attention at Keystone
  2. Microglia Reveal Formidable Complexity, Deep Culpability in AD
  3. Hot DAM: Specific Microglia Engulf Plaques
  4. ApoE and Trem2 Flip a Microglial Switch in Neurodegenerative Disease
  5. Parsing How Alzheimer’s Genetic Risk Works Through Microglia
  6. The Human Brain Hosts a Menagerie of Microglia
  7. Chimeric Mice: Can They Model Human Microglial Responses?
  8. Human Microglia Make Themselves at Home in Mouse Brain

Research Models Citations

  1. 5xFAD (B6SJL)
  2. APPSwe/PSEN1(A246E)
  3. APP NL-G-F Knock-in
  4. Trem2 KO (Colonna)

Paper Citations

  1. . Premature arrest of myelin formation in transgenic mice with increased proteolipid protein gene dosage. Neuron. 1994 Mar;12(3):583-95. PubMed.
  2. . TREM2-transduced myeloid precursors mediate nervous tissue debris clearance and facilitate recovery in an animal model of multiple sclerosis. PLoS Med. 2007 Apr;4(4):e124. PubMed.
  3. . TREM2 activation on microglia promotes myelin debris clearance and remyelination in a model of multiple sclerosis. Acta Neuropathol. 2020 Oct;140(4):513-534. Epub 2020 Aug 9 PubMed.

External Citations

  1. APOE knockouts

Further Reading

Primary Papers

  1. . White matter aging drives microglial diversity. Neuron. 2021 Apr 7;109(7):1100-1117.e10. Epub 2021 Feb 18 PubMed.