Besides patrolling the brain for tissue damage and other emergencies, microglia also help with routine maintenance, according to new research in the December 19 Cell. In adult mice engineered to lose microglia on demand, Wenbiao Gan, New York University School of Medicine, New York, and colleagues showed that the brain-resident immune cells facilitate synaptic plasticity. Mice lacking microglia performed poorly on several learning tasks. What’s more, ridding microglia of a single molecule—brain-derived neurotrophic factor (BDNF)—largely recapitulated the effects of depleting the brain phagocytes altogether.

“This paper reports a groundbreaking piece of work,” noted Richard Ransohoff of Cleveland Clinic, Ohio, who was not involved in the research. “One can no longer think of ‘resting’ microglia because they are never at rest, any more than one speaks of ‘resting’ and ‘activated’ neurons,” he wrote in an email to Alzforum.

Using two-photon imaging to examine microglia in the brains of adult mice, Gan and colleagues found the phagocytes constantly on the go (Davalos et al., 2005). “They responded quickly to injury,” Gan said. “What if they don’t have to wait for bad things to happen?” Gan wondered if microglia also help in normal everyday circumstances, such as during learning. Other studies show microglia interacting with synapses and suggest the phagocytes play a role in activity-dependent plasticity (see Nov 2010 news storySchafer et al., 2012). However, tools for manipulating microglia seemed too invasive or unspecific, often affecting both peripheral and brain myeloid cell populations, Gan said.

To overcome these challenges, first author Christopher Parkhurst and colleagues created mice that undergo inducible Cre-mediated recombination only in brain-resident microglia. These animals were bred with mice engineered to express the diphtheria toxin receptor (DTR) in cells where Cre is induced. Administering diphtheria toxin then wiped out brain microglia in the crosses. The researchers confirmed that myeloid cells developed and matured normally in the Cre-DTR mice, and that the recombinase activity was strong and specific to brain microglia.

With this working system in hand, the NYU researchers determined that removing microglia did no obvious harm to surrounding brain tissue. They checked cytokine production, blood-brain-barrier permeability, and neuronal and synaptic densities—all looked normal in microglia-depleted mice (see image below). Researchers had previously reported that short-term ablation of microglia in the brain had no effect on amyloid plaques in a mouse model of Alzheimer's disease (see Oct 2009 news story).

Image courtesy of Wen-Biao Gan and Cell, December 19, 2013

However, loss of the immune cells seemed to compromise synaptic plasticity during motor learning. The NYU team looked at this by mating Cre-DTR mice with Thy1-YFP-H transgenic mice that express yellow fluorescent protein in motor and sensory neurons. When microglia were depleted in 19- and 30-day-old crosses, the mice formed and eliminated spines less readily than control mice. Microglia-depleted mice also fell short on behavioral tests of motor learning, fear conditioning, and novel object recognition.

What causes these structural and behavioral abnormalities? Collaborating with co-author John Yates at Scripps Research Institute in La Jolla, California, Parkhurst and colleagues used a liquid chromatography/mass spectrometry approach to screen for protein expression changes induced by microglia removal. Of more than 6,500 proteins quantified, 61 changed in microglia-depleted mice—21 of them with known roles in synaptic function. The list includes the postsynaptic glutamate receptor subunit epsilon-2 (GluN2B) and the presynaptic vesicular glutamate transporter 1 (VGlut1). Electrophysiology data from patch-clamp experiments supported the proteomic screen, showing altered synaptic transmission in microglia-depleted brains.

Furthermore, the scientists recapitulated many of these abnormalities by removing BDNF from microglia. They looked at this molecule because of its involvement in synaptic development and plasticity (Chao, 2003), and because other work showed that microglial BDNF regulates neuronal plasticity in a mouse model of neuropathic pain (Coull et al., 2005).

All told, the present study suggests that “microglia help facilitate learning and memory in normal adult brain,” Gan said. In a similar vein, research reported last month at the Society for Neuroscience annual meeting in San Diego identified glial cells as critical players in maintaining neural circuitry (see Dec 2013 conference story). Earlier work had implicated the immune cells in synapse pruning during development (see Paolicelli et al., 2011), but until recently very few studies had demonstrated a role for brain microglia in adulthood, Gan noted. 

Beyond the new biological insight, scientists praised the current study for contributing a valuable tool for manipulating microglia in genetic models. Other researchers used a similar strategy to remove another gene (TGFb-activated kinase 1) from microglia (Goldmann et al., 2013).

Nick Varvel at Emory University School of Medicine, Atlanta, wonders if microglial BDNF levels drop in neurodegenerative disease and whether enhancing release of this molecule could help. Gan told Alzforum he would like to examine AD mice to see if they have reduced levels of microglial-derived BDNF. His team is continuing to delve into mechanisms for how glial BDNF affects neurons. BDNF expression in the hippocampus drops in AD (Murray et al., 1994), and neuroimaging analyses correlate the decline with hippocampal shrinkage and memory impairment (see Apr 2010 news story). Studies with rats suggest that exercise drives up hippocampal BDNF expression and promotes neurogenesis (see May 2002 news story).—Esther Landhuis


  1. This paper reports a groundbreaking piece of work. The major new conceptual insight is that microglia play critical roles (as do all brain cells) in the development and maintenance of the CNS. One can no longer think of “resting” microglia because they’re never at rest, any more than one speaks of “resting” and “activated” neurons. From this point forward, one must consider that altered microglial physiology (as during systemic infection, neurodegeneration, trauma, stroke, immune-mediated inflammation) may entail loss of physiological function (such as BDNF production, as in this paper, or IGF1 [Ueno et al., 2013]) possibly in addition to gain of toxic function, culminating in CNS dysfunction. One only has to think of the recent publication from Frank Heppner's lab (Krabbe et al., 2013) describing microglial impairment in a murine AD model to begin to imagine the many research questions emerging from the present publication, in context with other reports from this year alone. On a technical level the mice described here, in combination with a different model recently published (Goldmann et al., 2013), will enable unprecedented manipulation of microglia in genetic models. Finally there is a plethora of fascinating questions coming from this paper. How does the specific requirement for microglial BDNF, for example, relate to microglial-neuronal communication? Why does loss of microglial BDNF affect some but not other learning tasks?


    . Layer V cortical neurons require microglial support for survival during postnatal development. Nat Neurosci. 2013 May;16(5):543-51. Epub 2013 Mar 24 PubMed.

    . Functional impairment of microglia coincides with Beta-amyloid deposition in mice with Alzheimer-like pathology. PLoS One. 2013;8(4):e60921. PubMed.

    . A new type of microglia gene targeting shows TAK1 to be pivotal in CNS autoimmune inflammation. Nat Neurosci. 2013 Nov;16(11):1618-26. Epub 2013 Sep 29 PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. No Rest for Microglia: These Immune Cells Manage Healthy Synapses
  2. The Brain Minus Microglia—No Effect on Plaques
  3. Glial Cells Refine Neural Circuits
  4. Research Brief: BDNF Data Speak Volumes, Offer Therapeutic Target
  5. Run For Your Brain: Exercise Boosts Hippocampal Gene Expression, Neurogenesis

Paper Citations

  1. . ATP mediates rapid microglial response to local brain injury in vivo. Nat Neurosci. 2005 Jun;8(6):752-8. PubMed.
  2. . Microglia sculpt postnatal neural circuits in an activity and complement-dependent manner. Neuron. 2012 May 24;74(4):691-705. PubMed.
  3. . Neurotrophins and their receptors: a convergence point for many signalling pathways. Nat Rev Neurosci. 2003 Apr;4(4):299-309. PubMed.
  4. . BDNF from microglia causes the shift in neuronal anion gradient underlying neuropathic pain. Nature. 2005 Dec 15;438(7070):1017-21. PubMed.
  5. . Synaptic pruning by microglia is necessary for normal brain development. Science. 2011 Sep 9;333(6048):1456-8. PubMed.
  6. . A new type of microglia gene targeting shows TAK1 to be pivotal in CNS autoimmune inflammation. Nat Neurosci. 2013 Nov;16(11):1618-26. Epub 2013 Sep 29 PubMed.
  7. . Differential regulation of brain-derived neurotrophic factor and type II calcium/calmodulin-dependent protein kinase messenger RNA expression in Alzheimer's disease. Neuroscience. 1994 May;60(1):37-48. PubMed.

External Citations

  1. Thy1-YFP-H

Further Reading


  1. . Microglia sculpt postnatal neural circuits in an activity and complement-dependent manner. Neuron. 2012 May 24;74(4):691-705. PubMed.
  2. . Synaptic pruning by microglia is necessary for normal brain development. Science. 2011 Sep 9;333(6048):1456-8. PubMed.

Primary Papers

  1. . Microglia promote learning-dependent synapse formation through brain-derived neurotrophic factor. Cell. 2013 Dec 19;155(7):1596-609. PubMed.