The brain is the softest organ of the body, so when rigid Aβ plaques form, they stand in stark contrast to the squishiness surrounding them. Microglia sense this rigidity via Piezo1, a calcium channel that opens in response to mechanical stimuli. Two recent studies say as much. One, led by Wei Mo at Xiamen University in China, reported on November 8 in Neuron that rigid Aβ fibrils and plaques rev Piezo1 expression in microglia. Without this sensor, the cells do not surround and contain plaques, exacerbating synaptic damage and memory loss in a mouse model of amyloidosis. An earlier study, led by Tarja Malm at the University of Eastern Finland in Kuopio, had reported on June 15 in the Journal of Neuroinflammation that Piezo1 helps orchestrate the microglial response to plaques. Both groups found that Yoda1—an agonist of Piezo1—enhanced microglial recruitment to plaques and stimulated their phagocytosis, lessening Aβ burden.

  • Only stiff Aβ fibrils activate Piezo1 channels on microglia.
  • Without this sensor, mice accumulate more amyloid and lose more synapses.
  • A Piezo1 agonist improves microglial containment of plaques, lessens damage.

“These studies elegantly demonstrate that microglia are able to detect and respond to changes in tissue stiffness via the Piezo1 mechanoreceptor, and that this mechanism is essential for their detection and response to plaques,” wrote Kim Green of the University of California, Irvine (comment below). “Importantly, this further reinforces the protective effects of microglia in the early stages of disease, and their direct link to neuronal health and damage.”

Sanja Ivkovic of Serbia's University of Belgrade agreed, noting “The findings implicate Piezo1 as a necessary player in the defense system maintained by microglia.” Ivkovic's research focuses on lipids, microglia, and AD; she recently published a review article about how fatty acids tweak Piezo1 activation in AD (Ivkovic et al., 2022). 

Microglia respond to a plethora of environmental cues in the brain, including both molecular and mechanical signals. In a process called durotaxis, microglia reportedly migrate toward stiff surfaces, implying that mechanosensors are at work (Bollman et al., 2015; Moshayedi et al., 2014). Among the candidates are Piezo1 and Piezo2—mechanically activated, cell-surface cation channels that whisk calcium into the cell in response to sheering, stiffness, or pressure (Coste et al., 2010). 

In macrophages, Piezo1 helps orchestrate wound healing (Atcha et al., 2021). Recent studies suggest that Piezo1 also acts in the brain. One reported that astrocytes crank up the sensor near plaques, while another found that Aβ monomers, but not oligomers, squelched Piezo1 signaling in HEK293 cells (Velasco-Estevez et al., 2018; Maneshi et al., 2018). 

Might microglia call upon Piezo1 to detect and respond to Aβ aggregates? To address this, co-first authors of the Neuron paper—Jin Hu, Qiang Chen, Hongrui Zhu, and Lichao Hou—and colleagues first asked if brain tissue around plaques is indeed stiffer than regions without plaques. Using atomic force microscopy to scan slices of brain tissue from 5xFAD mice, they compared the rigidity of tissue near to, and far from, plaques. Lo and behold, they found that the tissue around plaques was almost twice as stiff. Perhaps unsurprisingly, in suspension, synthetic Aβ42 fibrils were stiffer than monomers of the peptide.

Feeling Stiff? Piezo1 senses the stiffness of plaques within softer brain tissue (top). Its activation triggers calcium influx, which activates microglial clustering, phagocytosis, and compaction of plaques (bottom). [Courtesy of Hu et al., Neuron, 2022.]

Next, Hu and colleagues asked if and how microglia respond to this stiffness. Searching published mouse and human gene-expression data, the researchers found that of all of the known mechanosensory ion channels expressed by microglia, Piezo1 reached the highest levels (Kim et al., 2019). When Hu exposed primary microglia to Aβ42 fibrils, or cultured the cells on rigid hydrogels, Piezo1 levels shot up, as did calcium influx. No such changes occurred in response to Aβ monomers, or when the cells were grown on soft hydrogels. When the stiffness signal was saturated by first growing microglia on a rigid hydrogel, neither Aβ monomers nor fibrils invoked a further uptick in Piezo1, suggesting it was the stiffness of the fibrils, rather than other molecular features, that had triggered Piezo1. The researchers pegged the lion’s share of the microglial stiffness response to Piezo1, because knocking it down or adding a Piezo1 inhibitor hobbled the calcium influx.

To investigate Piezo1 in the brain, the researchers crossed 5xFAD mice with transgenic mice producing fluorescently labeled Piezo1. They saw that microglia surrounding plaques expressed almost twice as much of this mechanosensor as their counterparts further afield. In postmortem brain samples from people with AD, as well, Piezo1 expression was highest in microglia surrounding plaques.

Plaques Encircled. In postmortem brain from a person with AD, Piezo1 (red) is expressed in microglia (green) that surround amyloid plaques (white). Nuclei are stained blue. [Courtesy of Hu et al., Neuron, 2022.]

Conditionally knocking out microglial Piezo1 in 5xFAD mice worsened their amyloidosis. They accumulated more plaques, sprung additional dystrophic neurites, lost more synapses, and had more trouble navigating water and Y mazes. Without Piezo1, microglial phagocytosis of Aβ waned, and fewer microglia surrounded plaques.

Those plaques were more diffuse and irregularly shaped, with fibrous structures extending outward. This finding jibes with previous studies implicating microglia as compactors, and even builders, of plaques (May 2016 news; Sep 2019 news; Apr 2021 news). Notably, phenotypes that had worsened without microglial Piezo1 were ameliorated when the scientists injected a small-molecule Piezo1 agonist called Yoda1 into the peritoneal cavity of 5xFAD mice (Syeda et al., 2015). 

Green pointed out that the phenotypes of Piezo1-deficient mice resemble those of TREM2 knockouts, in which microglia mount sluggish responses to plaques and cannot contain them. “The overlap suggests that microglia require both detection of changes in tissue stiffness as well as TREM2 ligands to react to plaques,” Green wrote. Interestingly, RNA-Seq revealed that Piezo1 deficiency did not affect the microglial transition to the disease-associated microglial (DAM) transcriptional state.

Over in Finland, in the Malm lab, co-first authors Henna Jäntti, Valeriia Sitnikova, Yevheniia Ishchenko, and colleagues also treated 5xFAD mice with Yoda1. They injected it directly into the brain, but reached essentially the same conclusions. Yoda1 bolstered microglial plaque containment and lowered Aβ burden. The scientists also combed through several single-cell RNA-Seq datasets but, unlike Hu et al., found Piezo1 expression to be negatively correlated with the DAM state. In postmortem brain samples from people, Piezo1 expression was highest among a cluster of microglia that were overrepresented in AD, suggesting some relationship between Piezo1 expression and AD pathogenesis.

On a related note, Oleg Butovsky of Brigham and Women’s Hospital wondered how Piezo1 could possibly help microglia migrate toward plaques without physically touching them first. He said the findings point to yet another cue microglia might use to sense and respond to plaques. Butovsky believes the expression pattern of Piezo1 needs further study, noting that in his published dataset of APP-PS1 mouse transcriptomes, microglia hardly expressed it, whether they were near plaques or not (Sep 2017 news). Indeed, Hu and colleagues also found no change in Piezo1 mRNA in microglia surrounding plaques—it was only the protein that doubled.

Andrew Stern of Brigham and Women’s Hospital, Boston, recently described diffusible Aβ fibrils within the AD brain (Nov 2022 news). He found both papers fascinating. “They add to the evidence that a protective function of microglia is to wall off plaques to limit Aβ access to cellular structures vulnerable to tau aggregation,” Stern wrote to Alzforum He is curious about the spatial scale at which Piezo1 detects changes in stiffness, asking “Can an individual fibril at the furthest edges of a plaque activate Piezo1, or does one need a thick ‘mesh’ that could influence the total stiffness of the surrounding parenchyma?”

Hu wrote to Alzforum that microglia may initially sense diffusible fibrils, such as those Stern et al. described, which could trigger the cells to start compacting them into plaques. “Since Piezo1 activation positively correlates with its protein level, we propose that there is positive feedback in which microglia that initially sense fibrillar Aβ42 stiffness through basal level Piezo1 upregulate the protein, mounting the protective response of microglia to Aβ plaques,” Hu wrote.

Adding a twist to that scenario, Malm and colleagues found that soluble Aβ counteracted Yoda1-induced activation of Piezo1 in induced human microglia. “Thus, microglia sense Aβ aggregates via Piezo1, but at the same time, blockage of Piezo1 by soluble Aβ may explain dysfunctional phagocytosis in microglia in the course of AD,” wrote Malm. This finding jibes with a previous study led by Philip Gottlieb of State University of New York in Buffalo, who reported that Aβ monomers, but not oligomers, inhibited Piezo1 signaling.

Several commentators expressed interest in activating Piezo1 channels as a therapeutic strategy for AD. “Both works indicated that Piezo1 ion channels on microglia cells are a new pharmacological target that can help in the removal of Aβ aggregates,” wrote Gottlieb (comment below). “At a more fundamental level, these publications show that cell mechanics affect brain function, and that we need to explore this issue in greater detail to better understand brain pathologies.”—Jessica Shugart

Comments

  1. These studies elegantly demonstrate that microglia are able to detect and respond to changes in tissue stiffness via the Piezo1 mechanoreceptor, and that this mechanism is essential for their detection and response to plaques. Furthermore, deletion of Piezo1 from microglia impairs their response to plaques, phenocopying Trem2 deficiency, and exacerbating plaque-induced damage. Importantly, this work further reinforces the protective effects of microglia in the early stages of disease, and their direct link to neuronal health and damage. It will be important to further evaluate this at later disease stages, and also in the context of tau pathology.

    The overlap between published Trem2 KO studies and the Piezo1 KO data are intriguing—it suggests that microglia require both detection of changes in tissue stiffness as well as TREM2 ligands to react to plaques. Notably, Piezo1 KO did not prevent the induction of TREM2-dependent, disease-associated microglia (DAM) gene induction, suggesting some non-overlapping roles, yet microglia in intact animals with reduced Piezo1 expression appeared to increase TREM2-dependent DAM gene expression.

    Both studies highlight the dramatic effects that the Piezo1 agonist Yoda1 has on microglial responses to plaques—increasing clustering, and decreasing plaque burden, dystrophic neurites, and cognitive impairments, etc. It will be intriguing to see how Piezo1 agonists develop and what effects they may have in other diseases and at different disease states.

  2. I think these two PIEZO1 papers are very interesting. They add to the evidence that a protective function of microglia is to wall off plaques to limit Aβ access to cellular structures vulnerable to tau aggregation.

    I wonder over what spatial scale PIEZO1 can detect changes in stiffness. Hu et al. report increased local changes in tissue stiffness near plaques, which are tens of microns in diameter. They also report that a suspension of synthetic Aβ fibrils, perhaps microns long, but presumably only nanometers in diameter, can provoke a PIEZO1-dependent microglial response. Can an individual fibril at the furthest edges of a plaque activate PIEZO1, or does one need a thick “mesh” that could influence the total stiffness of the surrounding parenchyma? Further, would a different (non-Aβ) extracellular amyloid have the same effect if it increased tissue stiffness, or is there something specific about Aβ fibrils?

    These will be interesting questions to address.

  3. Ever since its discovery in 2010, the function of Piezo1 has been widely characterized in different tissues and cell types, in both physiological and pathological contexts. However, little was known regarding its role in microglia.

    Here, Jin Hu et al. showed that amyloid plaques in mouse AD brains can increase tissue stiffness, and microglia sense and respond to tissue stiffness via Piezo1-mediated calcium signaling. They observed that 5xFAD mice with microglial-specific Piezo1 depletion sustained higher plaque load and more severe memory deficit, which was likely due to the impaired microglia phagocytosis; Yoda1 (a highly specific Piezo1 activator) administration nicely ameliorated AD pathology in 5xFAD mice, but not in Piezo1 KO 5xFAD mice. Transcriptomic analysis revealed that genes related to cytoskeleton dynamics were downregulated in Piezo1 KO microglia, which explained why they were less capable of engulfing and compacting Aβ plaques.

    The study from Jäntti et al. utilized iPSC-derived microglia-like cells and demonstrated that Yoda1 treatment increased microglia migration, phagocytosis, and lysosomal function. Similarly, they also apply Yoda1 in 5xFAD mice and observed a reduction of plaque burden.

    These discoveries uncover a novel aspect of how microglia sense Aβ plaques. Previous research in the field has focused on how innate sensing receptors (such as Trem2 and Clec7a) and phagocytic receptors (Mer, Axl, Tyro3) help to sense and engulf Aβ. These two papers showed that despite its biochemical properties, Aβ plaques can also provide mechanical stimuli to trigger microglial phagocytosis through an entirely different machinery.

    One of the remaining questions about these Piezo1-expressing microglia is where they fit in with the current microglia transcriptional landscape. Although Piezo1+ microglia are more enriched in plaque-associated regions (Figure 2 in Hu et al.) and have a more phagocytic and activated phenotype, they don’t seem to acquire disease-associated microglia (DAM) signatures. Jäntti et al. even showed that Piezo1 mRNA level negatively correlated with DAM marker genes. It would be interesting to further characterize their transcriptional signatures.

    It will also be interesting to know how Piezo1 expression is regulated in vivo—is it triggered by Aβ fibrils (as shown in the in vitro experiments by Hu et al.) or by other cues in the microenvironment? Another intriguing point to follow up is the interplay between mechanosensory signaling and innate receptor/phagocytic receptor signaling pathways during microglia activation (i.e., do they have a synergetic or redundant role), since calcium signaling is broadly involved in multiple pathways. These studies definitely open up a lot of avenues for further investigation.

  4. I read through both these papers and the findings are exciting. Both studies indicated that PIEZO1 ion channels on microglia are a new pharmacological target that can help in the removal of Aβ aggregates. These works showed that activation of PIEZO1 channels by a small molecule called Yoda1 reduced aggregate population. At a more fundamental level, these publications show that cell mechanics affect brain function and that we need to explore this issue in greater detail to better understand brain pathologies.

  5. Piezo 1 research has gained interest lately, especially in connection with Alzheimer’s disease pathological hallmarks, such as plaque formation and the ensuing changes in the stiffness in the tissue surrounding plaques. These changes in stiffness induce the activation and alteration in behavior of resident brain immune cells—the microglia. The role of microglia in Alzheimer’s disease (AD) has been studied by many researchers over the last decades. The recent findings deciphering the importance of Piezo1 in specific processes characteristic of microglia—the engulfment of Aβ plaques and phagocytosis of Aβ—implicate Piezo1 as a necessary player in the defense system maintained by microglia.

    Of particular importance is the potential to regulate Piezo1 with the agonist Yoda1, which has direct implications for the number, morphology, and distribution of Aβ plaques, resulting in changes in cognitive capacity in the 5xFAD AD transgenic mouse model. These two recent studies analyzed the effects of Yoda 1 in the presymptomatic and symptomatic phase of the disease, obtaining encouraging results in both cases. The presymptomatic phase of AD is particularly important because it is assumed that the disease starts in people several decades before the onset of the symptoms. It is possible that current therapies are so ineffective because by the time symptoms emerge the disease is so developed that the treatments are futile. It would be interesting to compare the effects of Piezo1 activation (or loss) on microglia in particular phases of the disease. The 5xFAD mouse aggressively produces Aβ under the Thy1 promoter, and the upregulation of Piezo1 may have even stronger effects in milder, more humanized AD mouse models, such as APP knock-ins. Although Yoda 1 effectively activated Piezo1 and diminished Aβ plaque burden when delivered into the brain, either intraperitoneally or through cerebroventricular infusion, the broader effects of activated Piezo1 warrant further investigations.

    Piezo1 function is regulated via force, from lipids or through the lipid composition of the membrane, including incorporated polyunsaturated fatty acids (PUFAs), which can affect Piezo1 by altering mechanosensitive properties of the cell. While PUFAs dietary supplementation can alter microglial polarization, the envelopment of amyloid plaques, and the immune response, Piezo1 activity was implicated in similar modulations of microglia behavior (for review see Ivkovic et al., 2022). It will be interesting to investigate the alterations of Piezo1 activity in microglia (in parallel with microglial behavior) under the treatment with PUFAs. Importantly, PUFAs treatment is currently in use in medical trials as the therapy for sickle cell anemia, a disease linked with the mutations in Piezo1.

    References:

    . Fatty acids as biomodulators of Piezo1 mediated glial mechanosensitivity in Alzheimer's disease. Life Sci. 2022 May 15;297:120470. Epub 2022 Mar 10 PubMed.

Make a Comment

To make a comment you must login or register.

References

Research Models Citations

  1. 5xFAD (C57BL6)
  2. APPPS1

News Citations

  1. Barrier Function: TREM2 Helps Microglia to Compact Amyloid Plaques
  2. Are Microglia Plaque Factories?
  3. Microglia Build Plaques to Protect the Brain
  4. ApoE and Trem2 Flip a Microglial Switch in Neurodegenerative Disease
  5. Short Aβ Fibrils Easily Isolated from Alzheimer's Brain Fluid

Paper Citations

  1. . Fatty acids as biomodulators of Piezo1 mediated glial mechanosensitivity in Alzheimer's disease. Life Sci. 2022 May 15;297:120470. Epub 2022 Mar 10 PubMed.
  2. . Microglia mechanics: immune activation alters traction forces and durotaxis. Front Cell Neurosci. 2015;9:363. Epub 2015 Sep 23 PubMed.
  3. . The relationship between glial cell mechanosensitivity and foreign body reactions in the central nervous system. Biomaterials. 2014 Apr;35(13):3919-25. Epub 2014 Feb 11 PubMed.
  4. . Piezo1 and Piezo2 are essential components of distinct mechanically activated cation channels. Science. 2010 Oct 1;330(6000):55-60. Epub 2010 Sep 2 PubMed.
  5. . Mechanically activated ion channel Piezo1 modulates macrophage polarization and stiffness sensing. Nat Commun. 2021 May 31;12(1):3256. PubMed.
  6. . Infection Augments Expression of Mechanosensing Piezo1 Channels in Amyloid Plaque-Reactive Astrocytes. Front Aging Neurosci. 2018;10:332. Epub 2018 Oct 22 PubMed.
  7. . Enantiomeric Aβ peptides inhibit the fluid shear stress response of PIEZO1. Sci Rep. 2018 Sep 24;8(1):14267. PubMed.
  8. . Deep proteome profiling of the hippocampus in the 5XFAD mouse model reveals biological process alterations and a novel biomarker of Alzheimer's disease. Exp Mol Med. 2019 Nov 15;51(11):1-17. PubMed.
  9. . Chemical activation of the mechanotransduction channel Piezo1. Elife. 2015 May 22;4 PubMed.

Further Reading

No Available Further Reading

Primary Papers

  1. . Microglial Piezo1 senses Aβ fibril stiffness to restrict Alzheimer's disease. Neuron. 2022 Nov 8; PubMed.
  2. . Microglial amyloid beta clearance is driven by PIEZO1 channels. J Neuroinflammation. 2022 Jun 15;19(1):147. PubMed.