The task of restraining microglia’s neurotoxic tendencies may fall to a small GTPase called RhoA, according to a study published June 23 in Cell Reports. Scientists led by João Relvas, University of Porto, Portugal, found that deleting RhoA from microglia unleashed a pro-inflammatory cascade that addled neuronal synapses and sapped memory—all in mice. Mice missing RhoA in their microglia produced more Aβ; on the flip side, Aβ aggregates squelched RhoA, tripping off a vicious cycle that hobbled microglial function and exacerbated neuronal damage. The researchers pieced together the signaling pathways involved, and used pharmacological inhibitors to stem the damage. Their findings cast RhoA as a stopgap against inflammation and amyloidosis in the brain.

  • Deleting RhoA in mouse microglia causes neuroinflammation, synaptic loss, memory deficits.
  • Without RhoA, mouse microglia stoke amyloidosis.
  • Aβ accumulation inhibits RhoA, dampens plaque clearance.

RhoA belongs to the family of Rho GTPases, which toggle between active, GTP-bound and inactive, GDP-bound states. When activated, Rho GTPases influence many cellular functions. They include the dynamic rearrangement of the cytoskeleton that happens during cell division and—importantly for the current study—during microglial activation. In particular, RhoA and its downstream effectors have been implicated in stroke, Alzheimer’s, Parkinson’s, and amyotrophic lateral sclerosis (Droppelmann et al., 2014). RhoA reportedly plays a role in the handling and toxicity of Aβ (Lee et al., 2019; Sycheva et al., 2019Zhang et al., 2019; Lee et al., 2019). Alas, RhoA and other small GTPases are active in multiple cell types, making it difficult to pin down exactly how RhoA influences neurodegeneration.

RhoA Says “WhoA!” to Inflammation. In healthy conditions (left), RhoA restrains pro-inflammatory signaling in microglia. Without RhoA (right), Src trips off a neurotoxic, amyloidogenic cascade. [Courtesy of Socodato et al., Cell Reports, 2020.]

To home in on the role of RhoA in microglial function, co-first authors Renato Socodato and Camila Portugal and colleagues generated conditional knockout mice, in which deletion of RhoA can be induced in CX3CR1-expressing cells. In the brain, that means microglia, in addition to infiltrating myeloid cells. Forty days after the researchers ablated RhoA from microglia, they observed fewer microglia in the hippocampi, cortices, and striata of the conditional knockouts compared with control mice. The remaining microglia expressed markers of necrosis, projected fewer processes, and pumped out more of the pro-inflammatory cytokine TNF-α. This suggested that RhoA supports microglial survival and homeostasis.

The researchers report that cultured primary mouse microglia lacking RhoA secreted TNF-α, which spurred the autocrine release of glutamate. In neurons cultured with conditioned media from these RhoA-deficient microglia, neurites beaded up, a telltale sign of excitotoxicity. In the mice missing RhoA in their microglia, the hippocampus lost synapses, and long-term potentiation weakened. Hippocampal neurons took a hit, dropping in number by nearly a quarter compared with controls. Mice lacking microglial RhoA also had faltered in a novel object recognition test.

Sapping Synapses? Compared with untreated control mice (left), mice missing RhoA in their microglia (middle) had fewer functional synapses (yellow). Treatment with the Src blocker AZD (right) protected synapses. Quantitation of synaptic puncta (right) represents three images of each of four hippocampal sections per mouse. [Courtesy of Socodato et al., Cell Reports, 2020.]

Using pharmacological inhibitors, knockdown approaches, and dominant-negative mutants, the researchers pieced together the signaling pathways responsible for the neurotoxic consequences of RhoA deficiency in microglia. Under healthy conditions, RhoA bolsters expression of Csk, a repressor of Src family kinases. Absent RhoA, Csk’s repression of Src slips, which unleashes TNF-α secretion, glutamate release, and excitotoxicity. The researchers found that injecting the Src blocker AZD0530 into the abdomens of RhoA-deficient mice counteracted the effects of RhoA deficiency.

How about Aβ production or deposition? The researchers report elevated levels of Aβ40 and Aβ42 peptides, and their amyloidogenic precursor, β-CTF, in the brains of mice lacking microglial RhoA. They also detected an excess of fibrillar Aβ wound into plaque-like structures. The researchers hypothesized that the excess glutamate released from RhoA-deficient microglia may have incited activity-dependent Aβ production by nearby neurons (Lesné et al., 2005; Dec 2005 news). 

If RhoA deficiency in microglia sparks Aβ production in neurons, might Aβ accumulation also dampen RhoA activity in microglia? In microglia isolated from 4-month-old APP/PS1 mice, the researchers observed less activated RhoA-GTP, lower expression of Csk, and more phosphorylated Src than in microglia from non-transgenic mice. Treating a microglial cell line with synthetic Aβ42 oligomers dampened RhoA activity and triggered TNF-α and glutamate release. The researchers believe Aβ oligomers dampen RhoA activity in microglia, fueling a loop in which TNF-α triggers glutamate release, which overexcites neurons and generates more Aβ. Alzforum was unable to obtain comment from the authors of this study.

Even before plaques formed, APP/PS1 mice lost synapses in the hippocampus, and their microglia became less ramified, a sign that the cells were rousing from homeostasis. Treating the mice with weekly doses of the Src inhibitor AZD0530 for one month substantially preserved synapses and microglial ramification, the researchers reported.

Oleg Butovsky of Brigham and Women’s Hospital in Boston called the work a technical tour de force exploring the role of RhoA signaling in microglial regulation. He found the induction of amyloidosis in mice lacking microglial RhoA particularly striking.

At the same time, Butovsky said, the role of RhoA in neurodegenerative processes remains unclear (for review, see Aguilar et al., 2017). For example, in gene-expression studies from his group, microglia encircling Aβ plaques expressed more RhoA than did microglia farther away from plaques (Krasemann et al., 2017). Some AD mouse models have elevated RhoA activity (Pozueta et al., 2013; Park et al., 2017). 

How RhoA tracks with neurodegeneration in humans needs to be explored. Some studies have spotted the protein mingling with tau (Huesa et al., 2009). The GTPase is expressed throughout the body, and in multiple cell types, making RhoA or its downstream effectors tricky therapeutic targets, Butovsky added.

To Richard Ransohoff of Third Rock Ventures, Boston, the findings seem at odds with a well-documented example of RhoA inhibition in humans, i.e., statin use. “If RhoA were functioning in human brain microglia in vivo as the authors suggest for mouse cells, then it would be anticipated that statin treatment, by inhibiting Rho family member prenylation and signaling, should worsen neuroinflammation,” he wrote. “However, this result has not been observed and, in fact, results opposite in direction have been reported. It would be useful for the authors to address this disconnect.” Along those lines, a recent mouse study found that suppressing RhoA signaling does not affect the inflammatory status of peripheral immune cells (Akula et al., 2019). 

Recent studies have shown that human and mouse microglia differ in many regards, including in their response to amyloidosis (Jul 2017 news; May 2019 newsJan 2020 news).—Jessica Shugart


  1. We very much appreciate the attention given by Alzforum to our recent paper. Concerning Dr. Ransohoff's comments on statins and RhoGTPases, we would like to draw attention to the following:

    1. Besides RhoA, statins impact several other signaling pathways, including those regulated by Rac1 and cdc42 (and most likely other RhoGTPases), and by members of the Rab and Ras GTPase subfamilies. Therefore, any effect of statins in microglia is ultimately the net result of changing the localization and activities of many small GTPase proteins simultaneously. In this context, it is nearly impossible to draw any valid conclusion about the role of microglial Rhoa signaling (or any other small GTPase) in microglial function, inflammation, or during AD pathogenesis, progression, and/or memory deficits. We feel that elucidating the specific functions of RhoGTPase signaling in microglia in health or disease requires stringent, careful, and thorough examination of the individual functions of RhoGTPases using different methodological approaches, including inducible conditional transgenesis as we have done in our publication. To understand RhoGTPase function in human microglia and in AD will require direct assessment of the activation status of the different RhoGTPase proteins, which rely on posttranslational modifications, an information that is unavailable in human brain transcriptomic datasets. Unfortunately, the present lack of reliable antibodies recognizing active, GTP-bound RhoGTPases has hampered our capacity to investigate microglial RhoGTPase activation in human tissue by immunohistochemistry or related methodologies.

    2. It is not very clear if statins will cause inhibition or activation of RhoA-dependent signaling, and how this would compare to our reported RhoA phenotype. As a matter of fact, in microglia-like cells, statins increase the GTP load of RhoA (Cordle et al., 2005), and even though statins can prevent RhoA prenylation and some form of membrane targeting, unprenylated (but GTP-loaded) RhoA can still be partially functional and activate downstream targets (Turner et al., 2008). Moreover, statins regulate microglia phagocytosis of Aβ, likely via inhibition of Rac1 function (Cordle et al., 2005), whereas we showed that activation of RhoA increased Aβ engulfment. On the other hand, statins have also been demonstrated to cause microglia inflammation (TNF production) in hippocampal slices by a mechanism requiring Rho prenylation (Bi et al., 2004). 

    3. Statins are not cell-type specific and, upon entering the brain, will modulate RhoA and the activity of many other small GTPases in different cell types (including oligodendrocytes, astrocytes, and neurons). As pointed out by Dr. Butovsky, this is particularly problematic if one considers that RhoA exerts nonredundant and context-specific roles in different CNS cell types that interact and crosstalk continuously.

    4. From an AD point of view, several reports have indicated that prolonged use of statins can lead to reversible cognitive impairment, whereas other studies showed beneficial effects of statins in decreasing the risk of developing AD (Schultz et al., 2018). 


    . Inhibition of geranylgeranylation mediates the effects of 3-hydroxy-3-methylglutaryl (HMG)-CoA reductase inhibitors on microglia. J Biol Chem. 2004 Nov 12;279(46):48238-45. PubMed.

    . Inhibition of geranylgeranylation mediates the effects of 3-hydroxy-3-methylglutaryl (HMG)-CoA reductase inhibitors on microglia. J Biol Chem. 2004 Nov 12;279(46):48238-45. PubMed.

    . The role of statins in both cognitive impairment and protection against dementia: a tale of two mechanisms. Transl Neurodegener. 2018;7:5. Epub 2018 Feb 27 PubMed.

    . Microglia Dysfunction Caused by the Loss of Rhoa Disrupts Neuronal Physiology and Leads to Neurodegeneration. Cell Rep. 2020 Jun 23;31(12):107796. PubMed.

    . Effects of lovastatin on Rho isoform expression, activity, and association with guanine nucleotide dissociation inhibitors. Biochem Pharmacol. 2008 Jan 15;75(2):405-13. Epub 2007 Sep 1 PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Paper Alert: Synaptic Activity Increases Aβ Release
  2. Human and Mouse Microglia Look Alike, but Age Differently
  3. When It Comes to Alzheimer’s Disease, Do Human Microglia Even Give a DAM?
  4. Human and Mouse Microglia React Differently to Amyloid

Research Models Citations

  1. APPSwe/PSEN1(A246E)

Paper Citations

  1. . The emerging role of guanine nucleotide exchange factors in ALS and other neurodegenerative diseases. Front Cell Neurosci. 2014;8:282. Epub 2014 Sep 10 PubMed.
  2. . Aβ modulates actin cytoskeleton via SHIP2-mediated phosphoinositide metabolism. Sci Rep. 2019 Oct 29;9(1):15557. PubMed.
  3. . Pro-Nerve Growth Factor Induces Activation of RhoA Kinase and Neuronal Cell Death. Brain Sci. 2019 Aug 19;9(8) PubMed.
  4. . Involvement of RhoA/ROCK Signaling in Aβ-Induced Chemotaxis, Cytotoxicity and Inflammatory Response of Microglial BV2 Cells. Cell Mol Neurobiol. 2019 Jul;39(5):637-650. Epub 2019 Mar 9 PubMed.
  5. . Pyk2 Signaling through Graf1 and RhoA GTPase Is Required for Amyloid-β Oligomer-Triggered Synapse Loss. J Neurosci. 2019 Mar 6;39(10):1910-1929. Epub 2019 Jan 9 PubMed.
  6. . NMDA receptor activation inhibits alpha-secretase and promotes neuronal amyloid-beta production. J Neurosci. 2005 Oct 12;25(41):9367-77. PubMed.
  7. . Rho GTPases as therapeutic targets in Alzheimer's disease. Alzheimers Res Ther. 2017 Dec 15;9(1):97. PubMed.
  8. . The TREM2-APOE Pathway Drives the Transcriptional Phenotype of Dysfunctional Microglia in Neurodegenerative Diseases. Immunity. 2017 Sep 19;47(3):566-581.e9. PubMed.
  9. . Caspase-2 is required for dendritic spine and behavioural alterations in J20 APP transgenic mice. Nat Commun. 2013;4:1939. PubMed.
  10. . Annexin A1 restores Aβ1-42 -induced blood-brain barrier disruption through the inhibition of RhoA-ROCK signaling pathway. Aging Cell. 2017 Feb;16(1):149-161. Epub 2016 Sep 16 PubMed.
  11. . Altered Distribution of RhoA in Alzheimer's Disease and AbetaPP Overexpressing Mice. J Alzheimers Dis. 2009 Sep 11; PubMed.
  12. . Protein prenylation restrains innate immunity by inhibiting Rac1 effector interactions. Nat Commun. 2019 Sep 4;10(1):3975. PubMed.

Further Reading


  1. . Rho GTPases in neurodegeneration diseases. Exp Cell Res. 2013 Sep 10;319(15):2384-94. Epub 2013 Jul 2 PubMed.

Primary Papers

  1. . Microglia Dysfunction Caused by the Loss of Rhoa Disrupts Neuronal Physiology and Leads to Neurodegeneration. Cell Rep. 2020 Jun 23;31(12):107796. PubMed.