Scientists are intensely interested in understanding the role neuroinflammation plays in neurodegenerative disease. In the April 16 Neuron, researchers led by Kim Green at the University of California, Irvine, provide a powerful new tool for doing so. Green and colleagues reported that a small molecule inhibitor eliminated virtually all microglia from the brains of wild-type mice, dousing ongoing inflammation. The mice remained healthy and active for at least two months, and even learned some cognitive tests faster than controls. Once the inhibitor was withdrawn, microglia rapidly repopulated the brain, returning to normal numbers within two weeks. Surprisingly, these microglia appeared to arise from progenitor cells scattered throughout the brain, rather than entering from the peripheral bloodstream as some previous studies had found. If confirmed, this would represent the first identification of a microglial progenitor in brain. “This has the potential to be the largest, most widespread stem cell pool in the brain,” Green told Alzforum.

Other scientists expressed enthusiasm for the work. “This is a milestone paper. The method they describe would allow you to test all kinds of questions that were not testable before,” said Tony Wyss-Coray at Stanford University, Palo Alto, California. Researchers could use it to study the role of microglia in neurodegenerative diseases, developmental disorders, brain injuries, and brain cancer, commentators suggested. Because the inhibitor is currently in clinical trials for peripheral cancers, it also holds potential for human therapy.

Previously, researchers used harsher methods to delete microglia. Mathias Jucker and colleagues wiped out the cells in transgenic mice using the antiviral ganciclovir, which inhibits DNA synthesis (see Varvel et al., 2012). Researchers led by Wenbiao Gan at New York University School of Medicine administered diphtheria toxin to transgenic mice to kill microglia (see Parkhurst et al., 2013). These approaches had drawbacks, as they required the use of transgenic animals and damaging toxins, noted Terrence Town at the University of Southern California, Los Angeles.

To develop a more versatile technique, Green and colleagues turned to colony-stimulating factor 1 receptor (CSF1R). In the brain, only microglia express this receptor. They require it during development for their proliferation and survival. CSF1R knockout mice lack microglia and die before adulthood (see Ginhoux et al., 2010Erblich et al., 2011). Green wanted to test whether CSF1R helped maintain microglia in the adult brain, as well, as the cells continue to express the receptor after birth.

Joint first authors Monica Elmore and Allison Najafi fed 290 mg/kg of the CSF1R inhibitor PLX3397 to year-old wild-type mice. The microglial numbers plummeted 50 percent within three days and by 90 percent after a week. The microglia expressed apoptotic proteins, indicating they were dying, and by two to three weeks of treatment they had vanished (see image below). Other cell types appeared unaffected. In behavioral tests of anxiety, motor skills, or fear conditioning, mice treated with inhibitor for up to two months had no changes; they even escaped from a maze faster than did controls. 

Vanishing microglia: Two weeks of inhibitor treatment removes nearly all microglia (bright green) from mouse brain. [Image courtesy of  Kim Green]

The authors then withdrew the inhibitor. To their surprise, the microglia rebounded within three days and replenished their former numbers after two weeks. Initially, these repopulating cells were larger than mature microglia, with shorter, thicker processes. They also expressed stem cell and proliferative markers, for example nestin, which is normally found in neuronal progenitors. By labeling proliferating cells with bromodeoxyuridine (BrdU), the authors confirmed that these nestin-expressing cells went on to mature into microglia, with typical microglial markers, size, and shape. The nestin-positive cells likely represent a previously unrecognized microglial progenitor, the authors concluded. 

“The existence of this stem cell has been postulated before, but this paper provides the best evidence so far that there is a microglial progenitor,” Wyss-Coray told Alzforum. Nonetheless, the idea may be controversial, commentators said. Several previous studies, including Jucker’s 2012 paper, reported that monocytes from the bloodstream flood into diseased or injured brains to replace lost microglia. However, monocytes cannot cross the blood-brain barrier in healthy brains, Town noted. Green and colleagues saw no evidence of peripheral monocytes entering the brain in their wild-type mice.

“The different modes of repopulation observed in different models of microglia depletion, involving infiltration of macrophages, proliferation of endogenous microglia, or recruitment of latent progenitors, highlight that there are redundant mechanisms to ensure that their functions are preserved,” wrote Ethan Hughes and Dwight Bergles at Johns Hopkins University School of Medicine, Baltimore, in an accompanying commentary in Neuron.

Carol Colton at Duke University, Durham, North Carolina, agreed that the evidence for a microglial progenitor appears strong. “This is one of the most detailed and complete studies I’ve seen. I’m very impressed with the body of work presented here,” she said. In particular, the authors’ analysis of gene expression changes in the repopulating microglia will provide valuable markers that will allow researchers to study microglial proliferation and differentiation in numerous disease and injury states, she noted.

In ongoing work, Green and colleagues are depleting microglia in mouse models of Alzheimer’s disease, traumatic brain injury, and stroke. “We now have a way not just to modulate neuroinflammation, but to eradicate it completely and controllably,” Green pointed out. Preliminary data suggests that getting rid of microglia benefits the AD brain, Green said. He noted that this fits with previous findings from Jucker’s group that the absence of microglia does not affect the size of amyloid plaques in AD mouse models at all (see Oct 2009 news story). Other studies have reported mixed results as to whether microglia help AD brains by clearing plaques, or harm them by stimulating neuroinflammation (see Jun 2013 news storyNov 2012 news story). 

Could CSF1R inhibitors such as PLX3397 dampen neuroinflammation in human AD? Made by the company Plexxikon in Berkeley, California, PLX3397 is currently in clinical trials for several types of cancer, including leukemia, lymphoma, solid tumors, and glioblastomas. CSF1R is a survival factor for macrophages, and the inhibitor helps suppress the growth of blood vessels in solid tumors. Many other pharma companies have CSF1R inhibitors as well. Nonetheless, several questions would have to be answered before trying this in people, including whether human microglia share the same dependence on CSF1R, Green said. Town suggested that the next logical step would be to test the drug in nonhuman primates to check for long-term toxicity and find the appropriate dosage. Some studies have identified a role for microglia in pruning synapses and remodeling neural circuits (see Dec 2013 conference storyDec 2013 news story), hinting that long-term depletion could have side effects. Despite these cautions, researchers were intrigued by the idea of being able to replace all the microglia in a diseased brain with a younger, presumably healthier set. Researchers have hypothesized that microglia in aged brains become sick from exposure to toxins, Town noted. “Would new microglia be better able to remove amyloid?” he asked. Scientists may soon have an answer.—Madolyn Bowman Rogers.


  1. This paper nicely demonstrates the dependence of postnatal microglia on CSF1R for survival, using an inhibitor from Plexxikon currently in clinical trials to treat non-Hodgkin lymphoma. While CSF1R-deficient mice fail to produce microglia and other myeloid lineage cells and eventually cause early postnatal death, it was unclear how sensitive microglia are to selective CSF1R inhibition. In addition, as also nicely demonstrated recently by Wenbiao Gan's group at New York University, and others, Green's group shows that microglia repopulate from cells within the CNS after ablation. However, this paper falls short in determining whether microglia repopulate in this case from progenitors (non-microglia cells with robust potential to become microglia) or rapidly proliferating microglia. The evidence for the first and against the second is largely correlative—and further studies are required to isolate and demonstrate the progenitor potential of putative microglia progenitor cells.

    View all comments by Mariko Bennett
  2. The new work from Kim Green’s lab is remarkable and provides several important insights into microglial biology. The most dramatic finding from this study is the discovery that virtually all adult microglia are entirely reliant upon CSF1R signaling for their survival. While the requirement for CSF1R activation was recognized developmentally, previous attempts to evaluate its role in the adult brain yielded less impressive results. The ready availability of a blood-brain-barrier-permeant drug to rapidly and selectively eliminate this cell type provides an important new tool. The most striking finding in this study is the observation that upon removal of the drug, the brain rapidly establishes its microglia population with normal tiling densities. The discovery of a proliferating, nestin-positive microglial progenitor that also expresses hematopoietic stem-cell markers in the adult brain and that appears between two and three days following drug withdrawal is striking and unexpected. The characterization of this progenitor and identification of the factors driving its genesis of microglia is of substantial importance and interest. It should be noted that Varvel et al. found that elimination of microglia using a CD11b-TK/ganciclovir approach resulted in repopulation of the brain by peripheral monocytes and not from cells endogenous to the brain at twice the normal tiling density. Recently, Parkhurst et al. depleted microglia in the adult brain by expressing the diphtheria toxin receptor selectively in these cells followed by diphtheria toxin administration. In this latter case no repopulation was observed. The basis for these widely varied experimental outcomes remains unexplained, but highlights the complex influences and dynamics involved in maintaining and occupying these myeloid niches in the mature brain.

    Green and colleagues reported that mice in which microglia had been depleted were behaviorally intact, as evaluated in a number of different tasks, even over a period of several months. This finding is quite different from that reported by Parkhurst et al., who found that learning a new motor skill was impaired in microglial-depleted mice, which also had impaired performance in a fear-conditioning assay and a novel-object-recognition task. Undoubtedly, resolution of this basis for these different experimental outcomes will be a high priority. Clearly, the extension of these approaches to evaluate microglial actions in disease models is an obvious next step and these experiments are likely to be underway. 

    Overall, this is an important paper from a dynamic young investigator who has made fundamental new discoveries. The biology of microglia is evolving at prodigious speed and in ways no one might have expected.  


    . Microglial repopulation model reveals a robust homeostatic process for replacing CNS myeloid cells. Proc Natl Acad Sci U S A. 2012 Oct 30;109(44):18150-5. PubMed.

    . Effects of anti-Ureaplasma urease antibody on homologous and heterologous urease activities. Microbios. 1987;49(198):47-54. PubMed.

    View all comments by Gary Landreth

Make a Comment

To make a comment you must login or register.


News Citations

  1. The Brain Minus Microglia—No Effect on Plaques
  2. Microglia Need Scara1 Receptor to Clear Soluble Amyloid
  3. Soothing Neuroinflammation Quells Plaques in Mice
  4. Glial Cells Refine Neural Circuits
  5. Beyond Neighborhood Watch—Microglia Nurture Synapses

Paper Citations

  1. . Microglial repopulation model reveals a robust homeostatic process for replacing CNS myeloid cells. Proc Natl Acad Sci U S A. 2012 Oct 30;109(44):18150-5. PubMed.
  2. . Microglia promote learning-dependent synapse formation through brain-derived neurotrophic factor. Cell. 2013 Dec 19;155(7):1596-609. PubMed.
  3. . Fate mapping analysis reveals that adult microglia derive from primitive macrophages. Science. 2010 Nov 5;330(6005):841-5. PubMed.
  4. . Absence of colony stimulation factor-1 receptor results in loss of microglia, disrupted brain development and olfactory deficits. PLoS One. 2011;6(10):e26317. Epub 2011 Oct 27 PubMed.

Further Reading

Primary Papers

  1. . Colony-stimulating factor 1 receptor signaling is necessary for microglia viability, unmasking a microglia progenitor cell in the adult brain. Neuron. 2014 Apr 16;82(2):380-97. PubMed.
  2. . Hidden progenitors replace microglia in the adult brain. Neuron. 2014 Apr 16;82(2):253-5. PubMed.