Skip to main content
Download PDF

Senolytic treatment diminishes microglia and decreases severity of experimental autoimmune encephalomyelitis

Journal of Neuroinflammation volume 21, Article number: 283 (2024) Cite this article

Abstract

Background

The role of senescence in disease contexts is complex, however there is considerable evidence that depletion of senescent cells improves outcomes in a variety of contexts particularly related to aging, cognition, and neurodegeneration. Much research has shown previously that inflammation can promote cellular senescence. Microglia are a central nervous system innate immune cell that undergo senescence with aging and during neurodegeneration. The contribution of senescent microglia to multiple sclerosis, an inflammatory neurodegenerative disease, is not clear, but microglia are strongly implicated in chronic active lesion pathology, tissue injury, and disease progression. Drugs that could specifically eliminate dysregulated microglia in multiple sclerosis are therefore of great interest to the field.

Results

A single-cell analysis of brain tissue from mice subjected to experimental autoimmune encephalomyelitis (EAE), a mouse model of CNS inflammation that models aspects of multiple sclerosis (MS), identified microglia with a strong transcriptional signature of senescence including the presence of BCL2-family gene transcripts. Microglia expressing Bcl2l1 had higher expression of pro-inflammatory and senescence associated genes than their Bcl2l1 negative counterparts in EAE, suggesting they may exacerbate inflammation. Notably, in human single-nucleus sequencing from MS, BCL2L1 positive microglia were enriched in lesions with active inflammatory pathology, and likewise demonstrated increased expression of immune genes suggesting they may be proinflammatory and contribute to disease processes in chronic active lesions. Employing a small molecule BCL2-family inhibitor, Navitoclax (ABT-263), significantly reduced the presence of microglia and macrophages in the EAE spinal cord, suggesting that these cells can be targeted by senolytic treatment. ABT-263 treatment had a profound effect on EAE mice: decreasing motor symptom severity, improving visual acuity, promoting neuronal survival, and decreasing white matter inflammation.

Conclusion

These results support the hypothesis that microglia and macrophages exhibit transcriptional features of cellular senescence in EAE and MS, and that microglia expressing Bcl2l1 demonstrate a proinflammatory signature that may exacerbate inflammation resulting in negative outcomes in neuroinflammatory disease. Depleting microglia and macrophages using a senolytic results in robust improvement in EAE disease severity, including across measures of neurodegeneration, inflammation, and demyelination, and may therefore represent a novel strategy to address disease progression in multiple sclerosis.

Introduction

Multiple sclerosis (MS) is a complex immune and neurodegenerative disease. The classic pathological feature of MS is the inflammatory demyelinating lesion which consists of myelin sheath degeneration and axonal injury. Extravasation and activation of T-cells, B-cells, and macrophages in the central nervous system (CNS), alongside gliosis of resident astrocytes and microglia contribute to demyelination, axonal injury, and neuronal and oligodendrocyte cell death [1]. Transcriptional studies of CNS glial cells have identified numerous dysregulated pathways, including complement signaling, interferon signaling, and antigen presentation in astrocytes, microglia, and oligodendrocytes respectively [2].

Of the CNS glial cells, microglia are particularly implicated in disease development and progression. Microglia and macrophages are a main contributor to the pathology of chronic active or smouldering lesions, which are characterized by the presence of a rim of reactive microglia and macrophages that contribute to gradually expanding tissue injury [3, 4]. Chronic active lesions and microglia activation and reactivity are also associated with MS disease progression and severity, even in the absence of new or ongoing peripheral immune cell invasion and inflammation [4, 5]. Reactive microglia can produce high amounts of the cytokine TNFα which is toxic to oligodendrocytes thus likely contributing to demyelination and oligodendrocyte cell death [4]. Treatments that deplete microglia are likewise effective at reducing disease severity in rodent models of MS [4]. However, microglial cell states in the CNS are highly complex, being influenced by their microenvironment, the molecules they express, and their function [6]. In some contexts, microglia promote tissue injury and inflammation, while in other contexts, microglia assume a phagocytic role in clearing myelin debris to promote remyelination and repair [4, 7,8,9]. Thus, strategies that specifically target injurious microglia could have an important role in reducing disease activity within the CNS and in tackling progression in MS.

The continuum of microglial states exists across the axes of homeostatic to injurious and young to old [10]. As a dividing cell population, microglia are susceptible to cellular senescence, and senescent microglia can perpetuate CNS inflammation and injury [11]. Importantly, aged microglia demonstrate decreased turnover and limited myelin debris phagocytosis suggesting impaired homeostatic functions [12, 13]. Like an ouroboros, CNS injury and inflammation can likewise accelerate microglial senescence. This has been particularly demonstrated in models of traumatic brain injury which induce a senescent profile in microglia; depleting these microglia subsequently improves outcomes related to cognition and tissue repair [14,15,16]. Further evidence for the relationship between aging and inflammation in microglia comes from an elegant study which conducted adoptive transfer EAE (AT-EAE) in young and middle-aged mice [17]. Here, AT-EAE in middle aged mice produces a significantly more severe and progressive disease course. Single-cell RNA-sequencing demonstrated that, compared to microglia from young AT-EAE mice, microglia from middle aged AT-EAE mice downregulate homeostatic marker genes, and upregulate genes related to proinflammatory interferon signaling, antigen presentation, and proteasome assembly [17].

Since biological age is correlated with conversion from relapsing-remitting to progressive MS, it is possible that disease progression without peripheral immune involvement is driven in part by the persistent accumulation of proinflammatory, senescent cells [18]. Senescent cells canonically upregulate anti-apoptotic genes including the BCL2 family genes, enabling them to evade cell death and persist longer in tissue, exacerbating inflammation and tissue injury [19]. Thus, drugs that can specifically target senescent cells may interrupt this vicious cycle. Senolytics are one such class of drugs that induce cell death in senescent cells and penetrate the CNS to eliminate senescent glia [16, 20, 21]. One notable study found that senolytic treatment prevented the upregulation of senescence genes and attenuated tau phosphorylation in a tauopathy mouse model [22].

Given the strong relationship between MS progression, aging, and microglia, this study aimed to investigate whether microglia in EAE and in MS exhibit a senescent profile, and whether they might be eliminated by senolytic treatment. Here, analysis of single-cell RNA-sequencing data from EAE and control mice establishes a proinflammatory and senescent microglial transcriptome in the disease. Assessment specifically of cells expressing the anti-apoptotic gene Bcl2l1 demonstrates that these cells exhibit even higher levels of proinflammatory and senescence-associated genes in EAE compared to their Bcl2l1 negative counterparts. In human MS single-nucleus sequencing, BCL2L1 was enriched in microglia in lesions with chronic active pathology, and BCL2L1 positive cells demonstrated concurrent upregulation of genes relevant to proinflammatory processes. Navitoclax (ABT-263), a small molecule BCL-2 family inhibitor and senolytic drug, led to significant depletion of microglia and macrophages in the spinal cord of EAE mice without notably reducing other immune cells. Remarkably, mice that received ABT-263 at disease onset exhibited significantly reduced motor symptom severity, improved neuronal survival, and diminished markers of inflammation in the white matter. Together, the data presented in this study provide evidence that senolytic therapy through BCL2 inhibition may be a promising avenue to reduce CNS inflammation and prevent disease worsening in MS.

Results

Analysis of single-cell RNA sequencing and immunohistochemistry of EAE microglia implicates cellular senescence

To characterize microglial gene expression in the context of pathological inflammation, single cell RNA-sequencing data from EAE at the peak timepoint and control mice was analyzed, and microglia were identified based on annotations from the published dataset [23] (Fig. 1A). In this cell cluster, high expression of microglial transcripts such as Aif1, P2ry12, Csf1r, and Cx3cr1 confirmed the microglial identity (Fig. 1B-E). Microglial gene expression was aggregated across biological replicates (n = 3 Control, n = 3 EAE) and differential gene expression analysis followed by KEGG pathway level analysis was conducted. KEGG pathways strongly prioritized pathways associated with neurodegeneration, including “Alzheimer’s disease” and “Parkinson’s disease”. Other biological pathways pointed towards cellular injury and senescence, including “Oxidative phosphorylation”, “Cell cycle”, “Cellular senescence”, and “Necroptosis”. Further pathways highlighted were related to DNA damage and repair, and proinflammatory signaling cascades including “Antigen processing and presentation”, “TNF signaling pathway”, and “NF-kappa B signaling pathway” (Fig. 1F). Thus, pathway level analysis suggests EAE microglia have a proinflammatory, neurodegenerative, and senescent phenotype. Additionally, genes commonly associated with senescence including Cdkn1a (padj < 0.01), Cdkn2a (padj < 0.01), and Rb1 (p < 0.05) were upregulated in EAE microglia (S. Table 1).

Fig. 1
figure 1

Single-cell RNA-seq of EAE microglia implicates senescence and identifies inflammatory subpopulation expressing ABT-263 targets. A-E) Microglial cells from Fournier et al. 2023 from EAE and Control mice, as confirmed by expression in the cluster of microglial specific genes Aif1 (B), P2ry12 (C), Csf1r (D), and Cx3cr1 (E). F) Kegg pathway enrichment for genes upregulated in EAE microglial demonstrates an enrichment for immune signaling pathways, cellular senescence, and neurodegenerative diseases. G) Percent of EAE microglia expressing various targets of ABT-263 and log2FC between EAE and control microglia. H) Top 20 upregulated genes in Bcl2l1 positive EAE microglia compared to Bcl2l1 negative EAE microglia, colour represents the difference in % positive cells between Bcl2l1 + population and Bcl2l1- population. I) Genes belonging to the SenMayo dataset significantly dysregulated (p < 0.05) in EAE vs. Control microglia (n = 47 genes, up = 42,down = 5, gene labels removed for clarity of figure). J) Significant dysregulation (p < 0.05) of genes from the 47 SenMayo genes dysregulated in EAE in Bcl2l1 positive EAE microglia, colour represents the difference in % positive cells between Bcl2l1 + and Bcl2l1- populations, positive number/red indicates higher percent positive cells in Bcl2l1 + population, negative number/blue indicates higher percent positive cells in Bcl2l1- population (n = 29 genes, Log2FC up = 28, Log2FC down = 1). K) GO biological process pathway enrichment for significantly upregulated genes (padj < 0.1, log2FC > 0) in Bcl2l1 positive EAE microglia

Full size image

A canonical feature of senescent cells is the upregulation of anti-apoptotic Bcl2 family genes, which can be targeted by a class of senolytic drugs that function through their inhibition [19]. These include Bcl2/BCL2, Bcl2l1/BCL-xL, and Bcl2l2/BCL-w. Given the strong inflammatory and senescent transcriptional phenotype in EAE microglia, the three Bcl2 family genes were assessed to evaluate whether senescent EAE microglia might be targetable by BCL2 inhibiting senolytics. Indeed, all markers were higher in EAE compared to control microglia, with Bcl2l1 notably being significantly upregulated and detected in over 20% of EAE microglia (Fig. 1G, Fig. S1). Differential expression of the Bcl2 family genes was assessed in other CNS cell types, however no other cell type demonstrated significant consistent upregulation of any of the genes (S.Table 2). Given its significant upregulation in EAE and overall higher proportion of positive microglia in EAE, Bcl2l1 was subsequently used to partition EAE microglia, and differential gene expression between Bcl2l1 positive and Bcl2l1 negative EAE microglia was conducted. Here, EAE microglia expressing Bcl2l1 showed significantly higher expression of many genes related to proinflammatory processes including Tnf, Tlr2, and Icam1 (Fig. 1H). Additional upregulated genes associated with aging, senescence, and inflammation notably upregulated in Bcl2l1 + EAE microglia included Ccl2 and Cdkn1a (S.Table 3). To better assess the specific contribution of senescence related genes to EAE microglia, genes from a curated senescence gene set were visualized based on significance and differential expression [24]. In EAE microglia, 47 of the 101 detected genes from this gene-set were dysregulated, with 42 being significantly upregulated in EAE (Fig. 1I). The 47 genes from the senescence gene set upregulated in EAE were then assessed in the Bcl2l1 positive EAE microglia. Here, compared to Bcl2l1 negative EAE microglia, Bcl2l1 positive EAE microglia showed further significant upregulation of 28 of the 47 genes, supporting the hypothesis that Bcl2l1 positive microglia exhibit higher levels of senescence associated genes than their negative counterparts (Fig. 1J). To further characterize the Bcl2l1 positive microglial subset in EAE, GO term analysis of significantly upregulated genes was conducted. This analysis revealed a strong inflammatory and cellular stress signature, with terms such as “activation of innate immune response”, “regulation of autophagy”, “response to ER stress”, “negative regulation of apoptotic signaling pathway”, and “myeloid cell differentiation” (Fig. 1K). Together, these data suggest that EAE microglia display a strong inflammatory and senescent signature, with a subset of microglia expressing anti-apoptotic Bcl2l1 and exhibiting an enhanced proinflammatory gene signature. To validate the presence of senescent microglia in EAE white matter, immunohistochemistry of EAE optic nerve was conducted for BCL2, IBA1, and senescence associated beta-galactosidase (SABG) (Fig.S2A-D). Confocal imaging demonstrated the unequivocal presence of numerous cells triply labeled for these markers. Indeed, BCL2 staining particularly labeled cells with a microglial like morphology (Fig. S2E-I). Altogether, these data provide evidence that microglia demonstrate markers of senescence and express BCL2 in EAE.

Human MS microglia display enrichment of BCL2 and BCL2L1 in chronic active lesions

As chronic active lesions are strongly associated with disease severity, this study first aimed to evaluate the microglial transcriptome in chronic active lesions. Two previously published single-nucleus sequencing datasets of cortical brain tissue from patients with progressive MS and age-matched controls were reanalyzed to assess microglial transcriptomes [25, 26]. The first dataset consisted of cells from healthy control tissue, MS patients with chronic active lesion pathology, and MS patients with chronic inactive lesion pathology [25] (Fig. 2A). The published cell type annotations were used to identify the microglial cell cluster, which demonstrated enrichment for microglial genes (Fig. 2B). Notably, comparing the microglia cluster across the annotated pathologies demonstrated an enrichment for BCL2L1 expression in chronic active samples, but not from control or chronic inactive samples (Fig. 2C). This is notable as inactive lesions are associated with resolved inflammation, whereas chronic active lesions are associated with ongoing inflammation and tissue damage. The second dataset was processed similarly, however here annotations from the original publication labeled a group of cell clusters as ‘immune’. Cluster 7 in this immune annotation was used in downstream analyses due to its stronger enrichment for microglial genes compared to other immune labeled clusters (Fig. 2D, E). Again, BCL2L1 expression was the highest in microglia from the chronic active labeled lesion pathology, suggesting that these cells may be contributing to ongoing inflammatory activity and tissue injury in this type of lesion (Fig. 2F). Similar enrichment was found for BCL2 (Fig. S3A-B). Analysis of differentially expressed genes between the BCL2L1 + and BCL2L1- microglia within this dataset identified the significant upregulation of several genes, including NFKB1 which is an integral component of the NFKB signaling pathway, ID2 which was detected in Bcl2l1 + mouse microglia as well and has previously been shown to be enriched in human aged microglia, and CH25H, which is involved in cholesterol processing and is also upregulated in human aged microglia [27,28,29] (Fig. 2G, S.Table 4). GO term analysis of the significantly upregulated genes prioritized pathways related to inflammation and cholesterol processing, including “myeloid cell differentiation”, “response to type II interferon signaling”, and “lipid storage” (Fig. 2H). To validate that microglia express BCL-xL in chronic active lesions, immunohistochemistry (IHC) for the protein was conducted on post-mortem MS patient brain tissue sections (Fig. 3, Fig. S4). Here, two non-MS control patients (control WM) and two MS patient sections with chronic active lesions were assessed for IBA1+/BCL-xL + cells (Fig. 3A, B). No double positive cells were observed in the control WM (Fig. 3A). However, multiple IBA1+/BCL-xL + cells were observed in the chronic active lesion edge (Fig. 3B). Additional IHC was conducted for pRB, a marker of cellular senescence. Here as well, control WM microglia did not display reactivity for pRB (Fig. 3C) whereas in chronic active lesions pRB + microglia were detected (Fig. 3D). Together, these data suggest that MS microglia from chronic active lesions exhibit features of senescence, are enriched for BCL2L1 expression at both the gene and protein level, and BCL2L1 positive human microglia, like EAE microglia, have an altered gene profile with heightened expression of known proinflammatory and aging related genes.

Fig. 2
figure 2

Single-nucleus RNA-seq of human MS microglia demonstrate enrichment for BCL2L1 genes in chronic active lesions and dysregulation of immune genes in BCL2L1 + cells. (A) Microglia from Schirmer et al. 2019 clustered based on microglial annotation accompanying the original published manuscript across control samples (Ctl), MS tissue with chronic active lesion pathology (CA) and chronic inactive lesion pathology (CI). (B) Microglial marker gene expression in the microglia annotated cluster. (C) Expression and percent cells expressing BCL2L1 across different conditions. (D) Microglia from Absinta et al. 2022, obtained from ‘immune’ annotation cluster 7 across different lesion pathologies including chronic active lesion edge (CA), chronic inactive lesion edge (CI), lesion core (LC), control non-lesion white matter (CWM), and periplaque white matter (PPWM). (E) Microglial marker gene expression in cells from (D). (F) Expression and percent cells expressing BCL2L1 across the different lesion classifications in cells from (D). (G) Visualization of top 50 upregulated genes (padj < 0.1, log2FC > 0) ordered by log2FC in BCL2L1 + microglia from the data in (D). (H) Significantly enriched GO terms from genes upregulated (padj < 0.1, log2FC > 0) in BCL2L1 + microglia

Full size image
Fig. 3
figure 3

Immunohistochemistry of MS lesions reveals BCL-xL + and pRB + microglia. (A) IBA1 + microglia in control white matter (tissue samples 1 and 2) do not co-label with BCL-xL. (B) IBA1 + microglia in MS chronic active lesions from two separate patients (tissue samples 3 and 4) co-label with BCL-xL in some cells. (C) IBA1 and pRB labeling in control white matter do not label the same cells. (D) IBA1 and pRB labeling in chronic active MS lesion tissue reveals overlap of the signals. Arrows drawn to indicate double positive cells. Scale bars = 20 μm

Full size image

Senolytic treatment depletes microglia and macrophages in EAE spinal cord

Given the presence of senescent-like microglia expressing BCL2 in EAE, the small molecule BCL2 inhibitor, Navitoclax (ABT-263), was administered to EAE mice and immune cell flow cytometry was conducted on spinal cord immune cell isolates (Fig. 4A). ABT-263 was selected as it is a high affinity inhibitor for all three members of the BCL2 family proteins (BCL2; Bcl2, BCL-xL; Bcl2l1, and BCL-w; Bcl2l2), is a known senolytic, and penetrates the CNS [30, 31]. Drug administration was conducted daily beginning at EAE symptom onset to not interfere with disease induction, and flow cytometry analysis was conducted at peak disease at least 3 days post treatment start. Cells were gated to enable analysis of percent population of viable single cells (Fig. 4B-I). CD45 and Cd11b were used to identify immune cells and differentiate the innate immune cells from lymphocytes (Fig. 4E-I). Notably, there was a significant reduction in CD45+/CD11b + cells, which consist of mostly innate immune cells (Fig. 4J). Thus, to narrow down the cells most likely being targeted, further gating was conducted. Microglia were characterized based on their medium expression of CD45, and low percent positivity for Ly6C and Ly6G (Fig. 4F, H). CD45 high Cd11b positive cells were similarly grouped based on Ly6G and Ly6C expression to subset different innate immune cell types (Fig. 4G). As microglia can upregulate CD45 in EAE and other neuroinflammatory conditions, with the markers used in this experiment, three broad categories were defined within the CD45high/CD11b positive group: monocytes/macrophages/CD45high-microglia (Ly6C-,Ly6G-), neutrophils (Ly6G+), and proinflammatory monocytes (Ly6C+) [32, 33]. Within the macrophage/monocyte/neutrophil populations, there was only a significant decrease in the Ly6C and Ly6G negative monocyte/macrophage/CD45high-microglia population, with no change in the neutrophil or proinflammatory monocyte populations (Fig. 4K-M). Similarly, in the CD45mid microglial population, there was a significant decrease with ABT-263 treatment (Fig. 4N). Finally, CD45 positive CD11b negative lymphocytes were stratified by CD4 and CD8 expression to identify the two major classes of T-cells (Fig. 4I). Neither lymphocytes broadly, nor the subpopulations of CD4 + and CD8 + lymphocytes demonstrated a change in percentage of the cell population (Fig. 4O-Q). Thus, these data demonstrate that senolytic treatment with a small molecule BCL2 inhibitor depletes microglia and macrophages in EAE spinal cord seemingly without altering the proportion of other major cell types of the innate and adaptive immune system.

Fig. 4
figure 4

ABT-263 treatment decreases proportion of microglia in spinal cord. A) Experimental design demonstrating induction of EAE followed by ABT-263 treatment at symptom onset and spinal cord flow cytometry at 18 d.p.i., and at least 4 days post-treatment start. B-D) Cell gating parameters for evaluation of viable single cell population. E) Gating strategy for isolating CD45-Cd11b- cells (R1) from CD45 + cells (R2). F) Division of R2 into three cell subsets, consisting of CD45 high Cd11b positive innate immune cells (R3), CD45 mid Cd11b positive microglia (R4) and CD45 + Cd11b- lymphocytes (R5). G) Division of R3 by Ly6C and Ly6G fluorescence to differentiate monocytes/macrophages (Ly6C-Ly6G-) from other innate immune cells (Ly6C + proinflammatory monocytes, and Ly6G + neutrophils). H) Division of microglial population based on Ly6C and Ly6G positivity demonstrating most cells (> 85%) in the R4 microglial population are negative for both. I) Division of the R5 lymphocyte population by CD4 and CD8 positivity. J) Quantification of the percent of viable cells within the innate immune cell population (R3 + R4) population demonstrating a significant decrease in population in ABT-263 treated animals. K-M) Division of the R3 population (CD45 high Cd11b+) based on Ly6G (K) and Ly6C (L) positivity, and the R3 population negative for both (M), demonstrating significant decrease only in the Ly6G and Ly6C negative population. N) Evaluation of the microglial subset in vehicle treated and ABT-263 treated mice as a percent of total viable cells, demonstrating a significant decrease in treated mice in both the whole population. O-Q) Analysis of the R5 lymphocyte population (O) based on subdivision of CD4 (P) and CD8 (Q) positivity showing no change in the number of CD4 + or CD8 + lymphocytes present in the spinal cords. * < 0.05, all comparison’s conducted with one-tailed unpaired t-test, vehicle n = 3, ABT-263 n = 4

Full size image

Senolytic treatment significantly improves EAE severity, neuronal survival, visual function, and white matter inflammation

To evaluate potential therapeutic effects of ABT-263 in the EAE model, mice were treated daily with ABT-263 beginning at disease onset (Fig. 5A). EAE motor symptoms were significantly milder in ABT-263 treated animals beginning at 2 days post treatment and continuing until experiment end compared to vehicle treated animals, with a trend towards improved weight maintenance (Fig. 5B, C). Likewise, visual acuity, a measure of visual system function, was better maintained in treated animals compared to vehicle controls (Fig. 5D). Next, optic nerve sections were assessed by immunohistochemistry to evaluate the extent of inflammation and demyelination in a uniform white matter tract (Fig. 5E-K). Here, immunohistochemistry determined that ABT-263 animals exhibited less Hoechst-stained area in the optic nerve suggestive of decreased inflammation and gliosis (Fig. 5E-G). This correlated with a significant decrease in IBA1 positive stained area (Fig. 5H). GFAP positive area was also assessed to determine whether astrocyte gliosis was similarly affected, but there was no difference between control and ABT-263 treatment (data not shown). Next, fluoromyelin labeling of lipid membranes in the optic nerve sections demonstrated a trend towards higher myelin area (Fig. 5I-K). Particularly, approximately 50% of control treated optic nerves demonstrated a significant decrease myelin area, while none of the ABT-263 optic nerves demonstrated significant myelin loss (Fig. 5K). Lastly, neuronal survival was assessed through BRN3A counts in retinal flatmounts to quantify surviving RGCs in vehicle control and ABT-263 treated EAE mice (Fig. 5L, M). Remarkably, there was a significant RGC survival effect in the ABT-263 condition (Fig. 5N). To assess the contribution of BCL-xL positive microglia, IHC for IBA1 and BCL-xL was conducted, and the fluorescence intensity of BCL-xL signal was evaluated within the IBA1 + area. Here, mice receiving ABT-263 demonstrated decreased fluorescent intensity of BCL-xL within IBA1 + optic nerve area but not significant within the total optic nerve, suggesting that ABT-263 may indeed deplete microglia with higher expression of BCL-xL. (Fig. 5O-R). Together, these data provide evidence that senolytic treatment inhibiting BCL2 family proteins in EAE mice decreases disease severity, improves outcomes related to visual acuity and neuronal survival, and reduces white matter demyelination and inflammation.

Fig. 5
figure 5

ABT-263 treatment from disease onset reduces severity, inflammation, promotes neuronal survival. (A) Experimental design demonstrating induction of EAE followed by ABT-263 treatment at symptom onset, optomotor testing at peak disease, and tissue collection at 18 days post induction (d.p.i.). (B) EAE score from day starting ABT-263 or vehicle treatment, data are combined numbers from two separate EAE cohorts. Bars = SEM. Two-way Anova with Sidak’s multiple comparison’s test with adjusted pvalues, * < 0.05, ** < 0.01. (C) Weight calculated as difference from baseline of day before symptom onset/treatment start (day 0). Bars = SEM. Two-way Anova with Sidak’s multiple comparison’s test with adjusted pvalues, * < 0.05. (D) Optomotor response data for right and left eyes of EAE mice at peak treated with either vehicle or ABT-263 for at least 3 days prior to evaluation. 3 stripe sizes were tested (0.3 c/d, 0.35 c/d, and 0.4 c/d) and data were binned into two groups no response (n.r.)/0.3 and 0.35/0.4. Number in pie slices represents % of responses in that category. Binomial test for observed versus expected distribution. E, F) Thresholded images of EAE optic nerves from vehicle treated (E) or ABT-263 treated (F) mice labeled with IBA1 (red) and Hoechst (white). Scale bar = 250 μm. G) Quantification of the percent area of optic nerve positive for Hoescht dye from thresholded images. Error bars = SD, two-tailed unpaired t-test. H) Quantification of the percent area of optic nerve positive for IBA1 immunostaining from thresholded images. Error bars = SD. Two-tailed unpaired t-test. I, J) Thresholded images of EAE optic nerves from vehicle treated (I) or ABT-263 treated (J) mice labeled with fluoromyelin (FM-red) and Hoechst (white). Scale bar = 250 μm. K) Quantification of the percent area of optic nerve positive for Fluoromyelin dye from thresholded images. Error bars = SD, one-tailed Welch’s t-test. L, M) Representative images of flatmount retinas labeled with RGC marker BRN3A in vehicle treated (L) and ABT-263 treated (M) EAE mice. Scale = 50 μm. N) Quantification of BRN3A positive cells per area in retinas from vehicle or ABT-263 treated mice. Bars = SD. Two-tailed unpaired t-test. O,P) Representative images of EAE optic nerves labeled with IBA1 and BCL-xL from control and ABT-263 treated animals. Scale bar = 200 μm. Q) Fluorescence intensity of BCL-xL in IBA1 + area. R) Fluorescence intensity of BCL-xL in the entire optic nerve area. Error bars = SD, two-tailed unpaired t-test performed. For E-R, * = pval < 0.05, ** = pval < 0.01, and *** = pval < 0.001

Full size image

Conclusion

The data presented here demonstrate that in EAE, microglia develop a senescent phenotype based on transcriptomics and IHC, and this subpopulation of microglia and potentially macrophages may be depleted through senolytic treatment. However, senescent microglia and macrophages likely constitute only a small proportion of the total phagocytes, with around 20% of EAE microglia expressing transcripts for Bcl2l1, 5% expressing transcripts for Bcl2, and senolytic treatment resulting in an approximately 5% reduction of spinal cord microglia and macrophages by flow cytometry and similar reduction in IBA1 positive area in the optic nerve white matter. That said, single-cell RNA-sequencing suggests that the subpopulation of EAE microglia expressing Bcl2l1 express higher levels of secreted proinflammatory factors. Thus, it is possible that senescent microglia may have an outsized effect on inflammation and injury by exacerbating proinflammatory signaling in their local microenvironment and reducing this cell population may broadly reduce CNS glia reactivity and inflammation (Fig. 6). This would be in line with the known function of the senescence-associated secretory phenotype, where senescent cells secrete a collection of proinflammatory factors that lead to persistent low-grade tissue inflammation and injury [19].

While there is a consensus that microglia occupy a continuum of states and functions dependent on their intrinsic gene and protein expression, and cues from their local microenvironment, equally there is significant evidence for certain genes consistently dysregulated in microglia across numerous different diseases and in aging [6, 10]. It is important to note that many studies on human microglia are often limited by the amount and quality of microglial nuclei, and systematic analysis of microglial gene expression changes compared to mouse gene expression changes in aging and disease finds that microglial signatures comparing across mice and human are not highly similar [2, 34]. Still, when considering within humans, microglia demonstrate consistent gene expression changes between aging and neurodegenerative diseases, particularly demonstrating a shared expression profile between microglia from AD patients and MS patients, and between genes downregulated in aging and MS [34]. Some of the genes common to aged human microglia were identified here in BCL2L1 + microglia, such as ID2 and CH25H. Equally, within mouse models of aging and neurodegeneration, Spp1, Ccl2, Ccl3, and Ccl4 are frequently upregulated and associated with proinflammatory microglial states [2, 28, 35,36,37]. It seems likely that this gene signature may also be associated with a senescent state among subsets of microglia, as these genes are highly enriched in senescent cells, and senescent microglia have been found in mouse models of Alzheimer’s disease and traumatic brain injury where these genes are upregulated [14, 15, 22, 24, 38]. Indeed, p16ink + microglia have been previously identified in MS brain lesions, implicating senescent microglia in the pathology of the disease [26]. Thus, the current body of research on microglia in aging and disease create a compelling argument for the presence and contribution of senescent microglia to disease activity in mice and humans [6, 28, 39]. Notably, microglia are essential for myelin maintenance, and aging related changes to autophagic pathways in microglia can interfere with their capacity to properly clear myelin debris [13, 36]. Interestingly, ‘regulation of autophagy’ and ‘macroautophagy’ were two of the top enriched GO terms specific to Bcl2l1 + EAE microglia suggesting these processes may be altered or impaired in senescent microglia.

Fig. 6
figure 6

Hypothesis for effectiveness of ABT-263 in EAE. Initial inflammatory insult and injury induces microglia to undergo reactivity and gliosis and may induce senescence in a subset of microglia (2). Senescent microglia may secrete more proinflammatory factors like TNFα, exacerbating local inflammatory activity (3a, top). Treatment with ABT-263 may induce apoptosis in the subset of senescent, proinflammatory microglia and macrophages, reducing the secretion of inflammatory factors into the local milieu, and therefore limiting further inflammatory activity (3b, bottom)

Full size image

Currently, microglia are a major target of interest for development of new MS therapies due to their capacity to perpetuate chronic inflammation and tissue injury in the CNS in the absence of ongoing blood-brain-barrier dysfunction and peripheral immune cell invasion [3, 5, 9, 26, 40]. For example, the in clinical trial Bruton’s tyrosine kinase (BTK) inhibitors are an exciting new development for MS treatment due to their capacity to pass through the blood-brain-barrier and potentially modulate microglial states therein [41]. Indeed, BTK inhibitors may decrease expression of proinflammatory cytokines including TNF, overexpression of which is associated with the proinflammatory, senescent transcriptomic profile seen here in EAE microglia [41]. Although EAE does not perfectly recapitulate the lesion pathology and disease progression seen in MS patients, both diseases are driven in part by the inflammatory activity of microglia and macrophages in the CNS, and all currently available MS therapies are efficacious in treating EAE suggesting some shared disease mechanisms between them. Targeting microglia is particularly relevant in the case of progressive MS, for which few therapies are currently available and effective at preventing CNS degeneration occurring independent of inflammatory relapses [42]. In MS, reactive microglia and macrophages contribute to the pathology of chronic active lesions, which are associated with heightened disease progression in patients [3, 26, 43]. Remarkably, the data in this manuscript show BCL2L1 positive microglia are most strongly enriched in chronic active lesions, suggesting their potential relevance to this pathology.

While these data suggest ABT-263 may be a viable approach to modulate microglial states in MS, further investigation is needed before reaching any clinical application. One caveat is the difficulty to truly distinguish between brain borne microglia and infiltrating phagocytes in neuroinflammatory conditions due to the strong overlap of markers at the gene and protein levels between these cell groups [32, 33]. Thus, the specific contribution of macrophages versus microglia here is not certain. However, the possibility that ABT-263 is eliminating both senescent macrophages and microglia may be of further benefit as both cell types exacerbate tissue injury in EAE and MS [44]. Additionally, the data presented in this manuscript do not rule out the possibility that other senescent cells not profiled in these analyses are also being targeted by ABT-263 and contributing to the observed phenotype, as ABT-263 has been demonstrated in other studies to have profound effects on various senescent cell populations (Table 1 [22, 45,46,47,48,49,50,51,52,53,54,55]). Since some immune cells may also express high levels of BCL2 at baseline, a more granular approach would be warranted in the future to assess whether senolytics affect these cells and their sub-populations. However, it is also possible that due to their high expression of BCL2, the low dose BCL2 inhibitors used to drive senescent cell death are insufficient to drive apoptosis in these populations.

Table 1 Summary of studies assessing ABT-263 in aging, CNS injury and neurodegenerative diseases [22, 36,37,38,39,40,41,42,43,44,45,46]
Full size table

Assessment of ABT-263 in other EAE models, particularly those that evoke a more progressive disease course such as the NOD-EAE model would be integral to understanding whether depletion of senescent cells ameliorates progressive tissue injury. Additionally, using such a model would enable more efficient evaluation for later treatment start and evaluation of the outcomes of such, better reflecting the treatment intervention timeline for progressive MS patients who may have lived with the disease for many years and accumulated substantial disability and disease activity by the time they may receive a therapeutic intervention. Indeed, better characterization as well of senescent microglial states in human MS patient tissue would further solidify potential therapeutic applicability of this approach. Finally, ABT-263 is a single senolytic, but many exist that may prove more effective in the clinic. For instance, the senolytic drug combination dasatinib and quercetin is already in clinical trials for Alzheimer’s disease, and it is feasible that if successful in AD, this combination therapy may also have off-label success in MS given what seems to be a common feature to many diseases, that senescent cells perpetuate neurodegenerative disease activity [56].

Materials and methods

Animal use approval

All in vivo animal experiments were performed using 6–8 week old C57Bl6 mice and experiments received ethical approval according to the guidelines set by the Canadian Council on Animal Care (CCAC).

Human tissue sample approval

Human brain tissue sections were obtained from patients with a clinical and neuropathological MS diagnosis according to the revised 2010 McDonald’s criteria [57]. Tissue samples were collected from healthy donors and MS patients with ethical approval (BH07.001, Nagano 20.332-YP) and informed consent as approved by the local ethics committee. Autopsy samples were preserved, and lesions classified by Luxol Fast Blue/Haematoxylin and Eosin staining and Oil Red O staining as previously reported [58, 59].

Single-cell RNA sequencing analysis

EAE RNA-sequencing data was obtained from the processed files published in Fournier et al. 2023, identified as microglia according to original publication annotations, pseudo-bulked and analysed for differential expression using the FindMarkers function with DESeq2 in Seurat v.5 [23, 60]. Upregulated genes (p < 0.05 and log2FC > 0) from differential expression analysis were input for KEGG pathway enrichment using the cluster Profiler package in R [61]. Microglia were subset based on EAE annotation, and each EAE microglial cell was annotated to reflect Bcl2l1 transcript presence or absence (counts > 0, counts = 0). Single-cell differential gene expression was conducted using SCT transformed values and the FindMarkers function on EAE microglia grouped by Bcl2l1 positivity, and differentially regulated genes were visualized in R. Pathway analysis was run using the goEnrich function from clusterProfiler on significantly upregulated genes (padj < 0.1 and log2FC > 0). Human single-nucleus datasets were obtained from Schirmer et al. 2019 and Absinta et al. 2022 with original annotations [25, 26]. Cluster 7 from the immune subset in the Absinta et al. 2022 dataset was used in downstream analyses due to high presence of microglial transcripts and was re-clustered to improve UMAP visualization. Percent positive cells for BCL2L1 was determined in R for each lesion pathology in both datasets. Due to higher number of cells in Absinta dataset, this dataset was used for downstream differential expression between BCL2L1 positive and negative cells, using SCT transformed counts and FindMarkers in Seurat. Pathway analysis was run using the goEnrich function from clusterProfiler on significantly upregulated genes (padj < 0.1 and log2FC > 0).

Experimental autoimmune encephalomyelitis

Active induction of MOG-EAE was achieved as previously described [62,63,64]. Briefly, 8-week-old female C57Bl6 mice received subcutaneous MOG peptide emulsified in complete Freund’s adjuvant (CFA) in 100 uL volume (50 uL injected bilaterally per mouse). Two days later, mice received intraperitoneal injection of 400 ng pertussis toxin. Mice were monitored daily in the morning for weight change and motor symptom development from 7 days post induction until experiment end. Beginning at symptom onset, ABT-263 or vehicle control was administered by intraperitoneal injection daily at 1.5 mg/kg, volume matched for controls based on weight. ABT-263 was prepared in vehicle solution containing 45% PEG-400, 45% DDH20, and 10% DMSO to a final concentration of 0.5 ug/mL. Mice were randomly assigned to vehicle and ABT-263 treatment groups.

Spinal cord immune cell isolation

Mice were anesthetized, then intracardially perfused with ice cold Hanks Balanced Salt Solution (HBSS w/o Ca2+ and Mg2+, herein HBSS), followed by decapacitation. Spinal cords were carefully isolated and immediately transferred into 1X HBSS with 20 mM HEPES. Spinal cord tissue was then transferred to a glass homogenizer containing 1 mL of prewarmed enzyme mix (1X HBSS, DNase I 0.025 U/mL, TLCK 0.1ug/mL, HEPES 20 mM, Collagenase type IV 2.5 mg/mL and Elastase 1U/mL) and gently homogenized fifteen times up and down. The tissue and enzyme mix were transferred into a tube and the homogenizer was rinsed twice with enzyme mix that was added to the same tube. Tissue and enzyme mix tubes were incubated for 30 min at 37 °C, with a homogenization step conducted after 15 min of incubation using a P1000 pipette to pipette the tissue and enzyme mix up and down ten times before incubating for the last 15 min. After, tubes were spun down at 300 g for 3 min, resuspended in the pre-existing supernatant, and filtered through a 70 μm cell strainer into a new tube. 10 mL of room-temperature wash buffer (1X HBSS,20 mM HEPES, 2 mM EDTA) was added to this mix and the tubes were immediately centrifuged at 300 g for 15 min. Percoll gradient solutions were prepared in HBSS for 37% and 70%. After centrifugation, supernatant was discarded and the pellet was resuspended in 5 mL of 37% Percoll. 5 mL of the 70% Percoll was carefully underlaid the 37% Percoll/cell suspension, and tubes were centrifuged for 20 min at 1000 g. Myelin debris was carefully removed from the top layer of the Percoll gradient, and ~ 3 mL of the 30–70% interphase was collected into a clean tube and diluted 3X with wash buffer. Tubes were centrifuged at 450 g for 10 min at 4 °C, then the cell pellet was resuspended in 300 uL of cold FACS wash buffer (1X HBSS, 0.5% fetal bovine serum (FBS)) and counted.

Flow cytometry

Cells were washed twice with cold FACS wash buffer, then blocked with Fc block antibody (4 uL / 100k cells) on ice for 15 min. Surface staining mix was prepared with fluorophore conjugated antibodies against Cd11b, CD45, Ly6C, Ly6G, CD4, CD8, and Zombie Aqua for cell viability (S. Table 5). Compensation controls were prepared on beads, while fluorescence minus one (FMO) controls were prepared on separate cell aliquots. Cells were incubated in a plate with 50 uL of surface staining mix at 4 °C in the dark for 30 min, then washed three times with cold FACS wash buffer. For flow cytometry settings and parameters, voltages were setup according to optimal PMT sensitivity using the peak 2 (Spherotech) voltration technique described previously by Maeker and Trotter [65]. Compensation control was performed using Ultracomp beads (Thermo Fisher) using optimal antibody concentration determined by titration. All data was acquired on Attune NxT (Thermo-Fisher). Analysis was conducted on all samples in floreada.io in order to obtain percent of viable population of relevant immune cell populations, following similar published gating methods, and using the FMO samples to identify and group the positive and negatively labeled cell populations [66, 67].

Optomotor response

Optomotor response was tested as previously described, in a custom-built drum with striped paper sheets rotated around the animal at 2 rpm in standard lighting [63, 68]. Stripe sizes tested included 0.3, 0.35, and 0.4 cycles/degree (c/d). Animals were acclimated to the apparatus for 10 min per animal at least one day prior to collecting testing data. Animals that were unable to balance themselves on the platform were excluded from testing. Video data was analysed blinded to mouse identity and treatment condition. Clockwise and counterclockwise responses were measured and compared as a group, such that each animal contributes two datapoints (one result per direction), since each eye is differentially affected in the disease course and contributes the clockwise vs. counterclockwise response. Data were aggregated into groups consisting of 0.4 and 0.35, and 0.3 and n.r., as 100% of presymptomatic EAE mice in our protocol respond at 0.35 or 0.4 c/d, thus a response below this threshold indicates decreased visual acuity.

Tissue collection and processing

Mice were anesthetized then intracardially perfused sequentially with ice-cold PBS and 4% paraformaldehyde (PFA). Eyes and optic nerves were carefully dissected out and post-fixed in 4% PFA overnight, before washing with PBS. For tissue sectioning, tissues were cryoprotected in 30% sucrose followed by mounting and flash freezing in optimal cutting temperature reagent (OCT). A cryostat was used to collect 12 μm sections on slides.

Immunohistochemistry of mouse tissue

For retinal flatmounts, retinas were washed 3x in phosphate buffer solution (PBS) then blocked overnight in 5% donkey serum and 0.2% Triton-X (blocking solution). After blocking, retinas were incubated in primary mouse anti-Brn3a antibody at 1:500 in 1% donkey serum and 0.2% Triton-X (staining solution) for 3 days at 4 °C (S.Table 5). Afterwards, tissue was washed 3X with PBS then incubated in secondary antibodies, AF568 donkey anti-mouse at 1:500 and Hoechst at 1:1000, in staining solution over night at 4 °C. Tissue was then washed 3x with PBS then mounted on slides with coverslips and Fluoromount mounting reagent. For tissue sections, sections were rehydrated with PBS and washed 3x with PBS. Then, sections were blocked with blocking solution for 1 h at room temperature, followed by primary antibodies overnight in staining solution at 4 °C (S. Table 5). Next, tissue was washed 3x with PBS and incubated with secondary antibodies in staining solution at room temperature for 1 to 3 h. Finally, sections were washed 3x with PBS and coverslips were mounted on slides with Fluoromount mounting reagent. Senescence associated beta galactosidase fluorescence was conducted using the SpiderGAL kit according to manufacturer’s instructions prior to IHC.

Immunohistofluorescence of human brain tissue

For immunofluorescence staining of paraffin sections, sections were deparaffinized in toluene for 10 min, followed by a second toluene wash of 5 min. Sections were rehydrated with 100% ethanol for 5 min, followed by 95% ethanol for 5 min. After 2 × 5 min washes with PBS, antigen retrieval was performed by placing slides in a decloaking chamber (Biocare Medical) containing 1mM EDTA buffer (pH 8.0) for 20 min at 95 °C. Slides were then washed 2 × 5 min with PBS before a hydrophobic barrier was applied around sections. Sections were washed with 0.1% Tween 20 in PBS (PBST) and endogenous biotin was blocked using the Endogenous Biotin Blocking Kit (Thermo-Fisher), following manufacturer’s instructions. Briefly, sections were incubated with streptavidin reagent (Component A) for 15 min at room temperature and washed 3 × 5 min with PBST, followed by incubation with biotin reagent (Component B) for 15 min at room temperature and 3 × 5 min washes with PBST. Sections were then blocked with 10% Normal Goat Serum in PBS for 30 min at room temperature. Primary antibodies were diluted in 0.1% Triton X-100 and 3% Normal Goat Serum before being incubated with sections overnight at 4 °C (S. Table 5). The following day, sections were equilibrated to room temperature for 1 h and washed 3 × 5 min with PBST. Secondary antibodies were applied at room temperature in 3 successive rounds of 40 min each. First, sections were incubated with a biotinylated goat anti-rabbit IgG antibody (1:500 in PBS), followed by 3 × 5 min washes with PBST. Second, sections were incubated with AF488-conjugated streptavidin (1:1000 in PBS), followed by 3 × 5 min washes with PBST. Third, sections were incubated with AF568 goat anti-mouse IgG (1:500 in PBS) and Hoechst (1:1000 in PBS), followed by 3 × 5 min washes with PBST. Finally, sections were washed 2 × 5 min with 1% Triton X-100 and 3 × 3 min with PBST before being coverslipped with Fluoromount Mounting Medium and stored at 4 °C, protected from light.

Imaging and analysis

Images were acquired on LSM 880 and LSM 900 confocal microscopes using z-stacking and tiling functions to image the entire length of optic nerve sections with the same laser intensity and settings across all samples. Images were imported into ImageJ and regions of interest were drawn manually to isolate the optic nerve. Default threshold parameters were used in ImageJ to generate a thresholded areas of markers of interest, and %area covered or fluorescence intensity within the region was measured in ImageJ.

Statistical analysis

Statistical tests were performed in R for sequencing data analysis, and in Graphpad Prism 8.0 for all other experiments. Details of the tests performed for each comparison and the values for statistical significance are reported in the figure captions.

Data availability

The datasets used in this manuscript are available online under accession numbers GSE199460 and GSE180759 on the GEO, and PRJNA544731 on SRA.

Abbreviations

ABT-263:

Navitoclax

AD:

Alzheimer’s disease

AT-EAE:

Adoptive transfer experimental autoimmune encephalomyelitis

BTK:

Bruton tyrosine kinase

CIHR:

Canadian institutes for health research

CNS:

Central nervous system

EAE:

Experimental autoimmune encephalomyelitis

ER:

Endoplasmic reticulum

GEO:

Gene expression omnibus

GO:

Gene ontology

MNI:

Montreal neurological institute

MS:

Multiple sclerosis

RGC:

Retinal ganglion cell

SABG:

Senescence associated beta galactosidase

References

  1. Filippi M, et al. Multiple sclerosis. Nat Reviews: Disease Primers. 2018;4. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41572-018-0041-4.

  2. Drake SS, Zaman A, Simas T, Fournier AE. Comparing RNA-sequencing datasets from astrocytes, oligodendrocytes, and microglia in multiple sclerosis identifies novel dysregulated genes relevant to inflammation and myelination. WIREs Mech Dis. 2023;15:e1594. https://doiorg.publicaciones.saludcastillayleon.es/10.1002/wsbm.1594.

    Article  CAS  PubMed  Google Scholar 

  3. Bagnato F, et al. Imaging chronic active lesions in multiple sclerosis: a consensus statement. Brain. 2024. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/brain/awae013.

    Article  PubMed  Google Scholar 

  4. Yong VW. Microglia in multiple sclerosis: protectors turn destroyers. Neuron. 2022;110:3534–48. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.neuron.2022.06.023.

    Article  CAS  PubMed  Google Scholar 

  5. Kuhlmann T, et al. Multiple sclerosis progression: time for a new mechanism-driven framework. Lancet Neurol. 2023;22:78–88. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/s1474-4422(22)00289-7.

    Article  PubMed  Google Scholar 

  6. Paolicelli RC, et al. Microglia states and nomenclature: a field at its crossroads. Neuron. 2022;110:3458–83. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.neuron.2022.10.020.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  7. Lampron A, et al. Inefficient clearance of myelin debris by microglia impairs remyelinating processes. J Exp Med. 2015;212:481–95. https://doiorg.publicaciones.saludcastillayleon.es/10.1084/jem.20141656.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  8. Rawji KS, et al. Niacin-mediated rejuvenation of macrophage/microglia enhances remyelination of the aging central nervous system. Acta Neuropathol. 2020;139:893–909. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s00401-020-02129-7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  9. Dong Y, et al. Oxidized phosphatidylcholines found in multiple sclerosis lesions mediate neurodegeneration and are neutralized by microglia. Nat Neurosci. 2021;24:489–503. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41593-021-00801-z.

    Article  CAS  PubMed  Google Scholar 

  10. Arutyunov A, Klein RS. Microglia at the scene of the crime: what their transcriptomics reveal about brain health. Curr Opin Neurol. 2023;36:207–13. https://doiorg.publicaciones.saludcastillayleon.es/10.1097/wco.0000000000001151.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  11. Rawji KS, et al. Immunosenescence of microglia and macrophages: impact on the ageing central nervous system. Brain. 2016;139:653–61. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/brain/awv395.

    Article  PubMed  PubMed Central  Google Scholar 

  12. Berglund R, et al. The aging mouse CNS is protected by an autophagy-dependent microglia population promoted by IL-34. Nat Commun. 2024;15:383. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41467-023-44556-6.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  13. Berglund R, et al. Microglial autophagy-associated phagocytosis is essential for recovery from neuroinflammation. Sci Immunol. 2020;5. https://doiorg.publicaciones.saludcastillayleon.es/10.1126/sciimmunol.abb5077.

  14. Ritzel RM, et al. Brain injury accelerates the onset of a reversible age-related microglial phenotype associated with inflammatory neurodegeneration. Sci Adv. 2023;9:eadd1101. https://doiorg.publicaciones.saludcastillayleon.es/10.1126/sciadv.add1101.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  15. Schwab N, Ju Y, Hazrati LN. Early onset senescence and cognitive impairment in a murine model of repeated mTBI. Acta Neuropathol Commun. 2021;9:82. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s40478-021-01190-x.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  16. Wang J, Lu Y, Carr C, Dhandapani KM, Brann DW. Senolytic therapy is neuroprotective and improves functional outcome long-term after traumatic brain injury in mice. Front Neurosci. 2023;17:1227705. https://doiorg.publicaciones.saludcastillayleon.es/10.3389/fnins.2023.1227705.

    Article  PubMed  PubMed Central  Google Scholar 

  17. Atkinson JR, et al. Biological aging of CNS-resident cells alters the clinical course and immunopathology of autoimmune demyelinating disease. JCI Insight. 2022;7. https://doiorg.publicaciones.saludcastillayleon.es/10.1172/jci.insight.158153.

  18. Graves JS, et al. Ageing and multiple sclerosis. Lancet Neurol. 2023;22:66–77. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/s1474-4422(22)00184-3.

    Article  PubMed  Google Scholar 

  19. López-Otín C, Blasco MA, Partridge L, Serrano M, Kroemer G. Hallmarks of aging: an expanding universe. Cell. 2023;186:243–78. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.cell.2022.11.001.

    Article  CAS  PubMed  Google Scholar 

  20. Oost W, Talma N, Meilof JF, Laman JD. Targeting senescence to delay progression of multiple sclerosis. J Mol Med (Berl). 2018;96:1153–66. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s00109-018-1686-x.

    Article  CAS  PubMed  Google Scholar 

  21. Sahu MR, Rani L, Subba R, Mondal AC. Cellular senescence in the aging brain: a promising target for neurodegenerative diseases. Mech Ageing Dev. 2022;204:111675. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.mad.2022.111675.

    Article  CAS  PubMed  Google Scholar 

  22. Bussian TJ, et al. Clearance of senescent glial cells prevents tau-dependent pathology and cognitive decline. Nature. 2018;562:578–82. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41586-018-0543-y.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  23. Fournier AP, et al. Single-cell Transcriptomics identifies Brain Endothelium Inflammatory Networks in Experimental Autoimmune Encephalomyelitis. Neurol Neuroimmunol Neuroinflamm. 2023;10. https://doiorg.publicaciones.saludcastillayleon.es/10.1212/nxi.0000000000200046.

  24. Saul D, et al. A new gene set identifies senescent cells and predicts senescence-associated pathways across tissues. Nat Commun. 2022;13:4827. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41467-022-32552-1.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  25. Schirmer L, et al. Neuronal vulnerability and multilineage diversity in multiple sclerosis. Nature. 2019;573:75–82. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41586-019-1404-z.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  26. Absinta M, et al. A lymphocyte-microglia-astrocyte axis in chronic active multiple sclerosis. Nature. 2021;597:709–14. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41586-021-03892-7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  27. Rolova T, et al. Complex regulation of acute and chronic neuroinflammatory responses in mouse models deficient for nuclear factor kappa B p50 subunit. Neurobiol Dis. 2014;64:16–29. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.nbd.2013.12.003.

    Article  CAS  PubMed  Google Scholar 

  28. Hammond TR, et al. Single-cell RNA sequencing of Microglia throughout the mouse lifespan and in the injured brain reveals Complex Cell-State changes. Immunity. 2019;50:253–e271256. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.immuni.2018.11.004.

    Article  CAS  PubMed  Google Scholar 

  29. Olah M, et al. A transcriptomic atlas of aged human microglia. Nat Commun. 2018;9:539. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41467-018-02926-5.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  30. Tse C, et al. ABT-263: a potent and orally bioavailable Bcl-2 family inhibitor. Cancer Res. 2008;68:3421–8. https://doiorg.publicaciones.saludcastillayleon.es/10.1158/0008-5472.Can-07-5836.

    Article  CAS  PubMed  Google Scholar 

  31. Zhu Y, et al. Identification of a novel senolytic agent, navitoclax, targeting the Bcl-2 family of anti-apoptotic factors. Aging Cell. 2016;15:428–35. https://doiorg.publicaciones.saludcastillayleon.es/10.1111/acel.12445.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  32. Greter M, Lelios I, Croxford AL. Microglia Versus Myeloid Cell nomenclature during brain inflammation. Front Immunol. 2015;6:249. https://doiorg.publicaciones.saludcastillayleon.es/10.3389/fimmu.2015.00249.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  33. Honarpisheh P, et al. Potential caveats of putative microglia-specific markers for assessment of age-related cerebrovascular neuroinflammation. J Neuroinflamm. 2020;17. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12974-020-02019-5.

  34. Srinivasan K, et al. Alzheimer’s patient microglia exhibit enhanced aging and unique transcriptional activation. Cell Rep. 2020;31:107843. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.celrep.2020.107843.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  35. Dong Y, et al. Single-cell and spatial RNA sequencing identify perturbators of microglial functions with aging. Nat Aging. 2022;2:508–25. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s43587-022-00205-z.

    Article  CAS  PubMed  Google Scholar 

  36. Marschallinger J, et al. Lipid-droplet-accumulating microglia represent a dysfunctional and proinflammatory state in the aging brain. Nat Neurosci. 2020;23:194–208. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41593-019-0566-1.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  37. Rangaraju S, et al. Identification and therapeutic modulation of a pro-inflammatory subset of disease-associated-microglia in Alzheimer’s disease. Mol Neurodegener. 2018;13. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13024-018-0254-8.

  38. Ng PY, Zhang C, Li H, Baker DJ. Senescent microglia represent a subset of Disease-Associated Microglia in P301S mice. J Alzheimers Dis. 2023;95:493–507. https://doiorg.publicaciones.saludcastillayleon.es/10.3233/jad-230109.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  39. Han T, Xu Y, Sun L, Hashimoto M, Wei J. Microglial response to aging and neuroinflammation in the development of neurodegenerative diseases. Neural Regen Res. 2024;19:1241–8. https://doiorg.publicaciones.saludcastillayleon.es/10.4103/1673-5374.385845.

    Article  CAS  PubMed  Google Scholar 

  40. Kamma E, Lasisi W, Libner C, Ng HS, Plemel JR. Central nervous system macrophages in progressive multiple sclerosis: relationship to neurodegeneration and therapeutics. J Neuroinflamm. 2022;19:45. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12974-022-02408-y.

    Article  CAS  Google Scholar 

  41. Krämer J, Bar-Or A, Turner TJ, Wiendl H. Bruton tyrosine kinase inhibitors for multiple sclerosis. Nat Rev Neurol. 2023;19:289–304. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41582-023-00800-7.

    Article  PubMed  PubMed Central  Google Scholar 

  42. Sutter PA, McKenna MG, Imitola J, Pijewski RS, Crocker SJ. Therapeutic opportunities for targeting cellular senescence in progressive multiple sclerosis. Curr Opin Pharmacol. 2022;63:102184. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.coph.2022.102184.

    Article  CAS  PubMed  Google Scholar 

  43. Absinta M, et al. Association of chronic active multiple sclerosis lesions with disability in vivo. JAMA Neurol. 2019;76:1474–83. https://doiorg.publicaciones.saludcastillayleon.es/10.1001/jamaneurol.2019.2399.

    Article  PubMed  PubMed Central  Google Scholar 

  44. Prineas JW, Parratt JDE. Multiple sclerosis: Microglia, Monocytes, and macrophage-mediated demyelination. J Neuropathol Exp Neurol. 2021;80:975–96. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/jnen/nlab083.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  45. Acklin S, et al. Depletion of senescent-like neuronal cells alleviates cisplatin-induced peripheral neuropathy in mice. Sci Rep. 2020;10:14170. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41598-020-71042-6.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  46. Mehdipour M, et al. Plasma dilution improves cognition and attenuates neuroinflammation in old mice. Geroscience. 2021;43:1–18. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s11357-020-00297-8.

    Article  CAS  PubMed  Google Scholar 

  47. Paramos-de-Carvalho D, et al. Targeting senescent cells improves functional recovery after spinal cord injury. Cell Rep. 2021;36:109334. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.celrep.2021.109334.

    Article  CAS  PubMed  Google Scholar 

  48. Ahire C, et al. Accelerated cerebromicrovascular senescence contributes to cognitive decline in a mouse model of paclitaxel (Taxol)-induced chemobrain. Aging Cell. 2023;22:e13832. https://doiorg.publicaciones.saludcastillayleon.es/10.1111/acel.13832.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  49. Faakye J, et al. Preventing spontaneous cerebral microhemorrhages in aging mice: a novel approach targeting cellular senescence with ABT263/navitoclax. Geroscience. 2023. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s11357-023-01024-9.

    Article  PubMed  PubMed Central  Google Scholar 

  50. Gulej R, et al. Elimination of senescent cells by treatment with Navitoclax/ABT263 reverses whole brain irradiation-induced blood-brain barrier disruption in the mouse brain. Geroscience. 2023;45:2983–3002. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s11357-023-00870-x.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  51. Tarantini S, et al. Treatment with the BCL-2/BCL-xL inhibitor senolytic drug ABT263/Navitoclax improves functional hyperemia in aged mice. Geroscience. 2021;43:2427–40. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s11357-021-00440-z.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  52. Fatt MP, et al. Restoration of hippocampal neural precursor function by ablation of senescent cells in the aging stem cell niche. Stem Cell Rep. 2022;17:259–75. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.stemcr.2021.12.010.

    Article  CAS  Google Scholar 

  53. Torres P, et al. A motor neuron disease mouse model reveals a non-canonical profile of senescence biomarkers. Dis Model Mech. 2022;15. https://doiorg.publicaciones.saludcastillayleon.es/10.1242/dmm.049059.

  54. Budamagunta V, et al. Effect of peripheral cellular senescence on brain aging and cognitive decline. Aging Cell. 2023;22:e13817. https://doiorg.publicaciones.saludcastillayleon.es/10.1111/acel.13817.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  55. Lu KJ, et al. Ability of Local Clearance of Senescent Cells in Ipsilateral Hemisphere to Mitigate Acute Ischemic Brain Injury in mice. Int J Biol Sci. 2023;19:2835–47. https://doiorg.publicaciones.saludcastillayleon.es/10.7150/ijbs.84060.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  56. Gonzales MM, et al. Senolytic therapy to modulate the progression of Alzheimer’s Disease (SToMP-AD) - outcomes from the first clinical trial of senolytic therapy for Alzheimer’s disease. Res Sq. 2023. https://doiorg.publicaciones.saludcastillayleon.es/10.21203/rs.3.rs-2809973/v1.

    Article  PubMed  PubMed Central  Google Scholar 

  57. Polman CH, et al. Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonald criteria. Ann Neurol. 2011;69:292–302. https://doiorg.publicaciones.saludcastillayleon.es/10.1002/ana.22366.

    Article  PubMed  PubMed Central  Google Scholar 

  58. Broux B, et al. Interleukin-26, preferentially produced by T(H)17 lymphocytes, regulates CNS barrier function. Neurol Neuroimmunol Neuroinflamm. 2020;7. https://doiorg.publicaciones.saludcastillayleon.es/10.1212/nxi.0000000000000870.

  59. Kuhlmann T, et al. An updated histological classification system for multiple sclerosis lesions. Acta Neuropathol. 2017;133:13–24. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s00401-016-1653-y.

    Article  CAS  PubMed  Google Scholar 

  60. Hao Y, et al. Dictionary learning for integrative, multimodal and scalable single-cell analysis. Nat Biotechnol. 2023. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41587-023-01767-y.

    Article  PubMed  PubMed Central  Google Scholar 

  61. Wu T, et al. clusterProfiler 4.0: a universal enrichment tool for interpreting omics data. Innov (Camb). 2021;2:100141. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.xinn.2021.100141.

    Article  CAS  Google Scholar 

  62. Drake SS, et al. 3-Dimensional immunostaining and automated deep-learning based analysis of nerve degeneration. Int J Mol Sci. 2022;23. https://doiorg.publicaciones.saludcastillayleon.es/10.3390/ijms232314811.

  63. Drake SS et al. Cellular rejuvenation protects neurons from inflammation mediated cell death. bioRxiv, 2023.2009.2030.560301, https://doiorg.publicaciones.saludcastillayleon.es/10.1101/2023.09.30.560301 (2023).

  64. Morquette B, et al. MicroRNA-223 protects neurons from degeneration in experimental autoimmune encephalomyelitis. Brain. 2019;142:2979–95. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/brain/awz245.

    Article  PubMed  Google Scholar 

  65. Maecker HT, Frey T, Nomura LE, Trotter J. Selecting fluorochrome conjugates for maximum sensitivity. Cytometry A. 2004;62:169–73. https://doiorg.publicaciones.saludcastillayleon.es/10.1002/cyto.a.20092.

    Article  PubMed  Google Scholar 

  66. Caravagna C, et al. Diversity of innate immune cell subsets across spatial and temporal scales in an EAE mouse model. Sci Rep. 2018;8:5146. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41598-018-22872-y.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  67. Montilla A, et al. Microglia and meningeal macrophages depletion delays the onset of experimental autoimmune encephalomyelitis. Cell Death Dis. 2023;14:16. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41419-023-05551-3.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  68. Caves EM, Troscianko J, Kelley LA. A customizable, low-cost optomotor apparatus: a powerful tool for behaviourally measuring visual capability. Methods Ecol Evol. 2020;11:1319–24. https://doiorg.publicaciones.saludcastillayleon.es/10.1111/2041-210X.13449.

    Article  Google Scholar 

Download references

Acknowledgements

We would like to acknowledge the essential technical support from Thomas Stroh and the Montreal Neurological Institute (MNI) Microscopy Core Facility, and Julien Sirois and the MNI Flow Cytometry Core Facility. We would also like to thank Steve Lacroix for providing advice and a protocol for immune cell flow cytometry.

Funding

S.S.D. received funding from the Canadian Institutes for Health Research (CIHR) Vanier Canada Graduate Scholarship. E.M.-L.H. is funded by the Fonds de Recherche du Quebec-Santé Master’s Training Scholarship. A.Z. is funded by the Fonds de Recherche du Quebec-Santé PhD Training Scholarship and MS Canada. A.E.F. receives funding from the CIHR and MS Canada.

Author information

Authors and Affiliations

  1. Montréal Neurological Institute, McGill University, Montréal, Québec, Canada

    Sienna S. Drake, Aliyah Zaman, Christine Gianfelice, Elizabeth M.-L. Hua, Kali Heale, Elia Afanasiev, Jo Anne Stratton, Stephanie Zandee & Alyson E. Fournier

  2. Neuroimmunology Research Laboratory, Centre de Recherche du Centre Hospitalier de l’, Université de Montréal (CRCHUM), Montreal, Québec, H2X 0A9, Canada

    Wendy Klement & Alexandre Prat

Authors
  1. Sienna S. Drake
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Aliyah Zaman
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Christine Gianfelice
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Elizabeth M.-L. Hua
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Kali Heale
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Elia Afanasiev
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Wendy Klement
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Jo Anne Stratton
    View author publications

    You can also search for this author inPubMed Google Scholar

  9. Alexandre Prat
    View author publications

    You can also search for this author inPubMed Google Scholar

  10. Stephanie Zandee
    View author publications

    You can also search for this author inPubMed Google Scholar

  11. Alyson E. Fournier
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

S.S.D. conceptualized the project, conducted all experiments, analysed the sequencing data, produced the figures, and wrote the manuscript. A.Z. oversaw the blinding of the data files and treatment conditions and conducted IHC for the human patient tissue sections. C.G conducted tissue processing and IHC of optic nerves. E. M-L. H., and K.H. assisted with EAE tissue processing. E.A. assisted with preparing single-cell sequencing objects and data analysis pipelines. J.A.S. oversaw E.A. contributions. W.K., A.P. and S.Z. assisted with human tissue processing. A.E.F. oversaw the project, provided funding and materials, and edited the manuscript.

Corresponding author

Correspondence to Alyson E. Fournier.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Supplementary Material 2

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Drake, S.S., Zaman, A., Gianfelice, C. et al. Senolytic treatment diminishes microglia and decreases severity of experimental autoimmune encephalomyelitis. J Neuroinflammation 21, 283 (2024). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12974-024-03278-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12974-024-03278-2

Keywords

Journal of Neuroinflammation

ISSN: 1742-2094