Flury A, Aljayousi L, Park HJ, Khakpour M, Mechler J, Aziz S, McGrath JD, Deme P, Sandberg C, González Ibáñez F, Braniff O, Ngo T, Smith S, Velez M, Ramirez DM, Avnon-Klein D, Murray JW, Liu J, Parent M, Mingote S, Haughey NJ, Werneburg S, Tremblay MÈ, Ayata P. A neurodegenerative cellular stress response linked to dark microglia and toxic lipid secretion. Neuron. 2024 Dec 19; Epub 2024 Dec 19 PubMed.
Recommends
Please login to recommend the paper.
Comments
Columbia University Medical Center
In the years following their first description in 2016, dark microglia were relegated to the backwaters of scientific discourse. Flury et al. brought this curious microglia phenotype back into the mainstream of glia research by establishing its relationship with the human and mouse microglia subsets identified in single-cell RNA sequencing studies, in particular the DAM phenotype. In a tour de force of mechanistic studies, they linked dark microglia to the integrated stress response (ISR) pathway and identified several toxic lipid species as effectors of this gain-of-function neurodegenerative microglia phenotype.
Given 1) the variable outcomes of global ISR inhibition in preclinical models of aging, Alzheimer’s, and other neurodegenerative diseases, 2) the unknown safety profile and 3) the likely narrow therapeutic dosing window of ISR inhibition, the development of approaches that specifically target the acquired effector function of neurotoxic lipid secretion by ISR/dark microglia subset is highly desirable. Additionally, while this study made great progress in further characterizing these once- obscure microglia, the comprehensive molecular phenotyping of this mainly ultrastructurally defined subset is still lacking. Efforts to gain in-depth understanding of the epigenetic, genetic, transcriptomic, proteomic, and metabolomic characteristics of the ISR/dark microglia, whether utilizing spatial multi-omic approaches, or cell-isolation techniques, could provide potentially novel therapeutic targets. This will undoubtedly move the field forward.
Curiously enough, the specific, upstream, microglia cell-autonomous sensome components that trigger the ISR/dark microglia phenotype are similarly unknown. This makes one wonder whether a smartly designed xenotransplantation experiment, combined with an in vivo CRISPR screen, could identify key sensome molecules/genes upstream of the canonical ISR response that are responsible for the development of this neurodegenerative microglia phenotype.
Federal University of Rio de Janeiro
Federal University of Rio de Janeiro
Federal University of Rio de Janeiro
Loss of homeostasis and toxic gain of function by microglia have been causatively implicated in Alzheimer’s disease pathogenesis. Studies over the past decade have uncovered a number of microglial subtypes (or subsets identified by their transcriptional profiles) found more prominently in neurodegenerative diseases, mainly in AD. In this study, Pinar Ataya, Marie-Eve Tremblay, and collaborators identified that dark microglia, an ultrastructurally distinct subtype of neurodegeneration-associated microglia, expand in response to activation of the integrated stress response.
The ISR is a ubiquitous cellular stress response mechanism activated in response to various stimuli, such as nutrient deprivation, endoplasmic reticulum dysfunction, and viral infections. These stimuli culminate in increased phosphorylation of the eukaryotic initiation factor 2α (eIF2α-P), which shuts down global mRNA translation, reduces cellular energy expenditure, and aims to restore homeostasis. Nonetheless, prolonged or aberrant ISR activation has been associated with neurodegenerative diseases.
We and others have shown that ISR and eIF2α-P are abnormally increased in AD brains and that targeting neuronal ISR blocks AD-associated synapse and memory impairments in mouse models (Lourenco et al., 2013; Ma et al., 2013; Oliveira et al., 2021). Here, the authors identified ISR activation in dark microglia in AD patients and in mouse models of Aβ pathology (5xFAD) and tauopathy (PS19) .
Flury, Aljayousi et al. generated genetically engineered mouse models to induce ISR (through an inducible version of the eIF2α kinase PKR) or to inhibit it (by introducing a non-phosphorylatable version of eIF2α, eIF2αA). Interestingly, the authors demonstrated that microglial PKR induction expands the pool of stressed microglia and impairs Aβ and tau clearance while disrupting synaptic homeostasis. Their findings further highlight that microglial ISR regulates lipid biosynthesis, contributing to neurotoxicity through secreted toxic lipids, which might also affect other brain cell types.
This very interesting study raises intriguing questions. First, while the authors report that elevated levels of eIF2α-P are found in disease-associated microglia (DAM) that are proximal (but not distal) to amyloid plaques in the brains of 5xFAD mice, they also show that a robust ISR activation is only induced by exposure of microglia to tau aggregates, but not to Aβ aggregates.
Second, the lack of difference in global translation levels in DAM compared to homeostatic microglia, despite the increase in eIF2α-P in DAM, is unexpected. This might suggest that, although somewhat elevated in relation to control microglia, eIF2α-P in DAM is not sufficient to cause a global arrest of translation.
Third, dark microglia are not eliminated by ISR inhibition and, conversely, induction of ISR in mice harboring the inducible form of PKR causes an accumulation of dark microglia that is fivefold less than the density of dark microglia found in the brains of 5xFAD mice. Importantly, even though accumulation of dense-core plaques, and dystrophic neurites, were increased in 5xFAD mice harboring the inducible PKR, these pathologies were not mitigated by ISR inhibition.
Fourth, it is not clear whether microglial ISR activation is a contributing factor or a consequence of neurodegeneration. Lastly, the specific contribution of ISR activation and lipid secretion by glial cells and neurotoxicity is intriguing. It would be interesting to clarify, for example, the nature of the secreted lipid mediators involved in toxicity: Are such toxic lipids released in exosomes/microvesicles or in free form/associated with carrier proteins?
Overall, this exciting study showcases the importance of ISR for microglial homeostasis, with potential implications for AD. Further research appears warranted to elucidate the roles of other cellular stress pathways, such as the unfolded protein response branches, and their interplay with ISR in mediating transcriptional alterations and microglial toxicity.
References:
Lourenco MV, Clarke JR, Frozza RL, Bomfim TR, Forny-Germano L, Batista AF, Sathler LB, Brito-Moreira J, Amaral OB, Silva CA, Freitas-Correa L, Espírito-Santo S, Campello-Costa P, Houzel JC, Klein WL, Holscher C, Carvalheira JB, Silva AM, Velloso LA, Munoz DP, Ferreira ST, De Felice FG. TNF-α mediates PKR-dependent memory impairment and brain IRS-1 inhibition induced by Alzheimer's β-amyloid oligomers in mice and monkeys. Cell Metab. 2013 Dec 3;18(6):831-43. PubMed.
Ma T, Trinh MA, Wexler AJ, Bourbon C, Gatti E, Pierre P, Cavener DR, Klann E. Suppression of eIF2α kinases alleviates Alzheimer's disease-related plasticity and memory deficits. Nat Neurosci. 2013 Sep;16(9):1299-305. PubMed.
Oliveira MM, Lourenco MV, Longo F, Kasica NP, Yang W, Ureta G, Ferreira DD, Mendonça PH, Bernales S, Ma T, De Felice FG, Klann E, Ferreira ST. Correction of eIF2-dependent defects in brain protein synthesis, synaptic plasticity, and memory in mouse models of Alzheimer's disease. Sci Signal. 2021 Feb 2;14(668) PubMed.
University of Washington
Microglia play both protective and toxic roles in AD pathogenesis, and genetic studies have supported the premise that inherent microglial dysfunction contributes to neurodegeneration. While numerous recent studies highlight the diversity of microglial states, including the emergence of "dark microglia" in disease, what drives the transition from homeostatic to a disease state remains a mystery. Toxic effects of microglial stress have been observed in AD tissues and tested in vitro, and the work here by Flury et al. provides insight into a specific stress pathway, the Integrated Stress Response (ISR) driving a microglial toxic phenotype.
Having identified evidence of ISR induction in AD patient brain, 5XFAD mouse brain, and in vitro with a microglia cell line, the group generated a series of genetic models. Using a chemogenetic system, they created crosses with 5XFAD mice that could be used to assess the effect of microglia-specific ISR induction (5xFADiPKR) or microglia-specific constitutive ISR inhibition (5xFADEif2A ). Transcriptomic and pathological studies of the 5xFADiPKR mice demonstrated the expansion of cell populations expressing genes often attributed to “DAMs,” and the shift in proportions of DAM states correlated to ISR induction or inhibition.
While ISR induction exacerbated 5XFAD pathology, inhibition of ISR did not alter plaque and axonal dystrophy. However, ISR inhibition did seem to rescue presynaptic puncta.
In contrast, crossing their ISR-manipulating lines to PS19 tau pathology mice revealed that ISR induction also worsened tau pathology while ISR inhibition did, in fact, decrease tau pathology.
Finding differential regulation of lipid synthesis genes in ISR-manipulated animals, the authors pursued lipid synthesis and secretion as candidate mechanisms of neurotoxicity. Activation of the ISR in vitro led to increased lipid synthesis and secretion.
Importantly, this ISR-mediated cell stress could be transmitted to neurons in cultured media, suggesting another mechanism by which microglia, challenged by pathological components of AD, can propagate cell stress and vulnerability to neurons.
Novel aspects
This study shows a potential link between the ISR and “dark microglia” and possible disease-relevant consequences to microglial ISR induction. The study describes innovative genetic tools to specifically manipulate the ISR in microglia for use in mice models. The group generated a chemogenetically inducible activation (iPKR) system and a constitutive inhibition (Eif2A) of ISR that could be crossed to established genetic AD pathology models. The identification of increased lipid synthesis and secretion, as well as transmittal of ISR stress—manifested as ISR-related gene expression—raises another potential target at the nexus of microglial stress and lipid metabolism in AD pathogenesis.
Another particularly interesting finding is the difference in ISR induction or inhibition with regard to Aβ or tau pathology. While ISR induction exacerbated both pathologies, ISR inhibition specifically reduced tau but not amyloid burden, underscoring the differences in microglial stress responses, or thresholds, to various proteinopathies.
Questions
One question that naturally arises is what is the relationship between ISR-activated microglia, dark microglia, and the described “lipid droplet-accumulating microglia,” which also show impairment of phagocytosis and lipid accumulation. Do these phenotypes, or states, lie along a spectrum of stress-response, or are they subsets of one another? One big challenge in mapping AD pathogenesis is establishing the temporal dynamics and triggers of observed microglial phenotypes in order to identify the most relevant and targetable states. Could the lipid-stress nexus be one such pathway? Could blocking ISR induction prevent LDAM development, and does impaired secretion of lipids contribute to the lipid accumulation phenotype?
Further studies into the relative effect of tau and Aβ (in all their various species) on microglia will likely facilitate interpretation of the myriad omics profiles in AD tissues and the heterogenous microglial signatures.
Make a Comment
To make a comment you must login or register.