Editor's evaluation: Atmospheric particulate matter aggravates CNS demyelination through involvement of TLR-4/NF-kB signaling and microglial activation
Year: 2021
Type: peer-review
Abstract: Article Figures and data Abstract Editor's evaluation Introduction Results Discussion Materials and methods Data availability References Decision letter Author response Article and author information Metrics Abstract Atmospheric Particulate Matter (PM) is one of the leading environmental risk factors for the global burden of disease. Increasing epidemiological studies demonstrated that PM plays a significant role in CNS demyelinating disorders; however, there is no direct testimony of this, and yet the molecular mechanism by which the occurrence remains unclear. Using multiple in vivo and in vitro strategies, in the present study we demonstrate that PM exposure aggravates neuroinflammation, myelin injury, and dysfunction of movement coordination ability via boosting microglial pro-inflammatory activities, in both the pathological demyelination and physiological myelinogenesis animal models. Indeed, pharmacological disturbance combined with RNA-seq and ChIP-seq suggests that TLR-4/NF-kB signaling mediated a core network of genes that control PM-triggered microglia pathogenicity. In summary, our study defines a novel atmospheric environmental mechanism that mediates PM-aggravated microglia pathogenic activities, and establishes a systematic approach for the investigation of the effects of environmental exposure in neurologic disorders. Editor's evaluation The study by Han et al. uses a rodent model of demyelination to investigate the effects of atmospheric particulate matter (PM) on CNS demyelination. The authors provide evidence that PM can promote and exacerbate demyelination which is associated with increased microglial activation and inflammation in the rodent central nervous system. These findings further our understanding of how environmental factors can influence human diseases. https://doi.org/10.7554/eLife.72247.sa0 Decision letter Reviews on Sciety eLife's review process Introduction Air pollution has become the prominent environmental risk factor that affects public health (Caplin et al., 2019). Among its heterogeneous composition, atmospheric particulate matter (PM) appears to be one of the most harmful components contribute to the pathogenesis of diseases (Babadjouni et al., 2017; Wu et al., 2018). Although a majority of PM research focuses on the respiratory and cardiovascular diseases, increasing epidemiological evidence has linked the irreversible adverse roles of continuous PM exposure with the morbidity and mortality of central nervous system (CNS) diseases (Boda et al., 2020; de Prado Bert et al., 2018; Younan et al., 2020). Studies have confirmed that nanosized PM can physically penetrate the blood-brain barrier and placental barrier, and therefore invade the brain parenchyma of adults and fetuses to directly induce effects in the brain (Bové et al., 2019; Maher et al., 2016). In 2018, the editorial of The Lancet Neurology reported that air pollution has been associated with increased risk of neurological disorders (The Lancet Neurology, 2018). Growing epidemiological reports have indisputably demonstrated that long-time PM exposure affects the development of the nervous system, thus increases the incidence and/or severity of Alzheimer's disease (AD), stroke, brain atrophy, anxiety, and multiple sclerosis (MS), among others (Babadjouni et al., 2017; Boda et al., 2020; de Prado Bert et al., 2018; Khan et al., 2019; Younan et al., 2020). In addition, due to the specific physiology of fetuses (relatively high ratio of respiration rate to body size and underdeveloped lung), PM exposure during pregnancy or early pregnancy causes long-term and irreversible adverse effects on neurodevelopment of fetus (Chandrakumar and 't Jong, 2019; Sripada, 2017). However, the pathogenic mechanism of PM leading to different/specific neurological diseases needs to be elucidated in depth. Myelin sheaths represent key structures for saltatory conduction of nerve impulse and trophic support of axons, and their loss or dysfunction leads to demyelination and impaired neurological function (Hughes and Appel, 2020). Clinically, demyelination occurs in a range of human neurological disorders as diverse as MS, optic neuromyelitis, leukodystrophy, spinal cord injury, AD, Parkinson's disease (PD), white matter stroke, as well as schizophrenia (Goldman and Kuypers, 2015; Goldman et al., 2012; Malpass, 2012; Marin and Carmichael, 2019; Mitew et al., 2010; Pukos et al., 2019). The exact etiology and pathogenesis of demyelination diseases are unknown so far, involving many complex factors such as environment exposures, autoimmune response, and genetic susceptibility (Kumar and Abboud, 2019). Klocke et al. found that exposure to concentrated fine and ultrafine particles during embryonic development affected oligodendrocyte maturation and brain myelination in adulthood (Klocke et al., 2017). Taking MS, a representative inflammatory demyelination disease, as an example, epidemiologic evidence highlights the effect of PM exposure on the risk of incidence, relapse, and deterioration of MS (Bai et al., 2018; Tateo et al., 2019; Zhao et al., 2019), suggesting strong correlations connecting PM exposure to demyelination and remyelination failure. However, the direct evidence of PM action on demyelinating disease is lack, and the cellular/molecular mechanisms leading to this disease process remain unclear. To investigate this important question, here we develop the complementary pathological and physiological demyelination/myelination models to identify PM that boost microglia-driven neuroinflammation, and define the TLR-4/NF-kB signaling pathways involved in the regulation of microglia pro-inflammatory activities as well as PM-related demyelinating diseases. Results PM exposure aggravates myelin injury in EAE, a CNS inflammatory demyelination model Neuroinflammation is one of the major pathogenic factors resulting in CNS injury. To assess the impact of PM aspiration on inflammation-induced demyelination in vivo, we first employed experimental autoimmune encephalomyelitis (EAE) model to recapitulate human demyelinating diseases based on immune response. Experimental design and treatment strategies are shown in Figure 1A. We found that PM exposure (nasopharyngeal inhalation, 5.0 mg/kg/d) exacerbated EAE progression with an earlier onset compared with the PBS-treated control (Figure 1B and C). In PBS group, the average day of onset was day 12 post immunization (p.i.), while in PM-treated group disease onset began at~ day 8 p.i., deteriorated rapidly, and no recovery after reaching the peak (Figure 1B). The majority of EAE mice in PBS group displayed moderate signs (limp tail, wadding gait, or paralysis of one limb), while almost all PM-treated mice exhibited severe signs (complete paralysis of both hind limbs or moribund) at the end of observation (Figure 1C). Figure 1 with 1 supplement see all Download asset Open asset PM exposure aggravates myelin injury in an inflammatory demyelination model. (A) Schematics of treatment strategies for EAE. Female, 8–10 week-old C57BL/6 mice were immunized with MOG35-55 and pre-treated with PBS or PM (nasopharyngeal inhalation, 5.0 mg/kg/d) daily, starting at day –30 (before immunization) until 30 p.i. All mice were sacrificed and their tissues (brain, spinal cord, draining lymph nodes, and spleen) were harvested for Q-PCR, flow cytometry, ELISA, or immunohistochemistry analysis at day 30 p.i. (B) EAE development was evaluated daily by two researchers blindly, following to a 0–5 scale. (C) Distribution of disease status at the end points of experiment (day 30 p.i.). (D) Thoracic spinal cord sections were assayed for inflammation by H&E and demyelination by Luxol fast blue (LFB), and CNS pathology was scored on a 0–3 scale. (E) Absolute number of CNS mononuclear cells (MNCs) in cell suspension of each mouse (brain and spinal cord) was counted. (F) Representative images of spinal cord sections of PBS- and PM-treated EAE mice in the dorsal funiculus. Quantitative analysis of MBP, IBA1, A2B5, and GFAP expression was assessed by using Image-Pro. The measured areas included 8–10 fields and covered virtually all the white matter of the spinal cord. Dorsal column at the thoracic spinal cord is shown as representative images. (G) Effects of PM treatment on the various inflammatory cells in the CNS. MNCs from spinal cords and brains were isolated at day 30 p.i., stimulated with MOG35–55 (10 μg/mL) for 24 h, and analyzed by flow cytometry. Cells were gated as CD45+CD11b+ (microglia and infiltrating macrophages) and CD45+CD11b- (other infiltrating immune cells), and their subsets were further defined. One representative of three independent experiments is shown. Symbols represent mean ± SD; n = 4–5 mice in each group. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001, compared to PBS-treated group, two-way ANOVA comparison with Multiple t' tests. Scale bar = 40 µm in D, Scale bar = 10 µm in F. To evaluate the effect of PM on EAE-associated CNS pathology, we isolated thoracic spinal cord sections of EAE mice for histology and immunohistochemistry evaluation. Consistent with the clinical finding, histological analysis revealed significantly enhanced inflammatory and demyelinating foci in the white matter of the spinal cord (Figure 1D) in PM-treated animals when compared with the PBS control. PM-treated mice also had a remarkably increased number of mononuclear cells (MNCs) in the CNS (Figure 1E). The total number of MNCs per mouse in the PM-treated group was 13.9 ± 1.1 × 106, which is ~2.2 fold of that PBS control (6.24 ± 0.8 × 106, p < 0.001, Figure 1E). We then determined the effect of PM on myelin loss using anti-MBP (myelin marker) staining. As shown in Figure 1F and a significant degree of MBP loss (demyelination) had occurred in PM-treated mice, indicating disease progression. Furthermore, significantly accumulation of IBA1+ activated microglia (~2.5 fold in PM-treated EAE mice compared with PBS controls, p < 0.001) and A2B5+ OPCs (~2.7 fold in PM-treated EAE mice compared with PBS controls, p < 0.001) in the area of demyelination injury were detected in PM-treated mice, while PM aspiration did not significantly impact the expression of the astrocyte marker GFAP (Figure 1F). To further evaluate the effects of PM treatment on the various inflammatory cells in CNS, isolated MNCs from CNS were analyzed by flow cytometry. The percentage and absolute number of CD45+ CD11b + cells (microglia and infiltrating macrophages) and other infiltrated immune cells (CD45+ CD11b− cells) were increased obviously in PM group, compared with the PBS control (Figure 1G). Increased expression of CD80 and MHCII was also observed in CD11b + cells (microglia/infiltrating macrophages) and CD11c + cells (dendritic cells; DCs) (Figure 1G), indicating an enhanced activation of these cells. The total numbers of CD4+ and CD8+ T cells and percentages of Th17 (CD4+ IL17+) and Th1 (CD4+ IFN-γ+) cells were also significantly increased under PM aspiration (all p < 0.05; Figure 1G). To study the effect of nasopharyngeal inhaled PM on the peripheral immune response, splenocytes of EAE mice were stimulated with MOG35-55 peptide and analyzed by flow cytometry. As shown in Figure 1—figure supplement 1, splenocytes in PM-treated group exhibited significantly increased numbers and expression of co-stimulatory molecules (e.g. MHC class II, CD80, CD86) of antigen-presenting cells (CD11c+), and higher percentage of IFN-γ+, IL-17+, and TNF-α+ CD4+ T cells. We also examined the effect of PM on Th17 and Th1 cell differentiation in vitro, and similar to LPS stimulation, PM treatment significantly increased Th1 cells but had no significant effects on Th17 cells (Figure 4—figure supplement 1A). Overall, these results showed that PM exposure possibly aggravates inflammatory demyelination of the CNS. PM aspiration exacerbates demyelination and prevents remyelination in a toxin-induced demyelination model We then investigated the impact of PM aspiration on coprizone-induced demyelination, another classical mammalian animal model that for studying pathological processes associated with demyelinating diseases. Experimental design and treatment paradigms were shown in Figure 2A. As shown in Figure 2B, body weight of mice in PBS group showed progressive loss during the 4 weeks of cuprizone fed. After cuprizone withdrawn and returning to normal diet for 2 weeks, weight loss was gradual recovery and remyelination was observed in the corpus callosum area of mice in the PBS group (Figure 2B and C). However, continuous PM aspiration obviously inhibited the weight gain (Figure 2B). PM exposure significantly increased the susceptibility to demyelination injury caused by cuprizone (oral administration for 3 weeks), and prevented CNS spontaneous remyelination in the corpus callosum even after cuprizone withdrawn for 2 weeks, as detected by FluoroMyelin staining (Figure 2C). The numbers of CNS resident immune cells, GFAP+ astrocytes and IBA1+ microglia, were also significantly increased in the PM-treated group both at demyelination and remyelination process, indicating that PM exposure exacerbated neuroinflammation and demyelination in toxic-induced demyelinating mice, and significantly inhibited myelin repair (Figure 2C). Under 4-week-cuprizone treatment and 2-week normal chow, most of the axons in the corpus callosum re-wrapped by thin myelin sheath (spontaneous remyelination) in PBS-treated group, while in PM group only few axons with loosely wrapped myelin were observed as evaluated by ultrastructual electron microscopy (EM; Figure 2D). PM aspiration not only significantly reduced the number of myelination axon, but also increased G-ratio of the remyelinated axons, indicating a poor recovery from demyelination (Figure 2E). These findings suggest that PM aspiration exacerbates demyelination and prevents remyelination in toxin-induced demyelination model. Figure 2 Download asset Open asset PM aspiration exacerbates demyelination and prevents remyelination in a toxin-induced demyelination model. (A) Treatment paradigms. Male, 8–10 week-old C57BL/6 mice were pre-treated with PBS or PM (nasopharyngeal inhalation, 5.0 mg/kg/d) daily from week –4 to week 6. Standard rodent diet containing 0.2% copper chelator cuprizone (CPZ), which causes CNS demyelination, were fed for 4 weeks to achieve complete demyelination in the corpus callosum; cuprizone was then withdrawn and mice were again fed normal chow, allowing for spontaneous remyelination to occur within the next 2 weeks. (B) Body weights of mice from different groups were recorded every 2 days. (C) Representative images and quantitative analysis of LFB, FluoroMyelin, and immunohistochemistry (GFAP, IBA1) stains in the body of the corpus callosum of PM- or PBS-treated mice at different timepoints. (D) Representative electron microscopy images of the corpus callosum region isolated from PM- or PBS-treated mice at 2 weeks after cuprizone withdrawal. Arrows heads mark demyelinated axons. (E) Quantification of the myelinated axons shown in (D). The G-ratios (axon diameter divided by entire myelinated fiber diameter) of myelinated fibers and number of myelination axon were assessed by using Image-Pro. One representative of three independent experiments is shown. Symbols represent mean ± SD; n = 5–8 mice in each group. *p < 0.05; **p < 0.01; ***p < 0.001, compared to PBS-treated group, two-way ANOVA comparison with Tukey's multiple comparisons test. Scale bar = 50 µm in C, Scale bar = 2 µm in D. PM exposure during pregnant and postnatal delays myelinogenesis in the developing nervous system Given that the respirable PM aggravated myelin injury under pathological conditions (immune- and toxicity-induced demyelination), we then investigated whether maternal PM exposure during pregnancy affected the postnatal myelination in the developing nervous system under nonpathological conditions by employing the neonatal mouse myelinogenesis model. We hypothesized that PM aspiration during pregnant and neonatal period would inhibit precocious oligodendrocyte differentiation and myelination; therefore, we chose the postnatal day 14 to examine the extent of myelination in the developing corpus callosum, because the establishment of the myelin sheath completes within the first 3 weeks postnatally in rodents (Osorio-Querejeta et al., 2017; Figure 3A). As shown in Figure 3 B, C, 14 days after birth the extent of MBP intensity in the corpus callosum of PBS-treated group were achieved 33.4% ± 8.5% per field, while PM treatment significantly decreased the intensity of MBP staining to 10.1% ± 2.6%, indicative of a 70% reduction in myelination compared with the PBS-treated littermates at this time point (p < 0.001). The numbers of IBA1+ microglia and GFAP+ astrocytes were also increased significantly following the PM treatment, indicating the widespread inflammation in the brains of offspring mice and the critical role of micro- and astrogliosis in myelination/demyelination (Figure 3B and C). Lower numbers of myelination axons as well as thinner myelin sheathes were found in the PM treatment group than those of the PBS-treated littermates at postnatal Day 14 by EM (Figure 3D and E). The expression levels of pro-inflammatory cytokines such as Tnfa, Il1b, Il6, and Inos were remarkably increased in PM-treated mice, with significantly inhibited myelination and myelin protein gene expression compared with the PBS control (Figure 3F). These pathological changes prompt us to find out if the delayed myelin development induced by PM exposure resulted in any adverse effects in neurological function. Balance and motor coordination were evaluated in a rotating rod, beam walking, and tight rope test. In the rotating rod test at accelerating speed, PM-treated mice exhibited a significant shorter latency to fall off the rod than that of PBS-treated mice (decreased –32.6% in the time of staying on the rotating rod, Figure 3G, p < 0.001). Disturbances of motor dysfunction in PM-treated mice were also observed in the beam walking and tight rope test. Mice exposed to PM had significantly worse motor performance, as the time taken by mice to traverse a narrow beam or cotton rope increased remarkably compared to PBS-treated control mice (Figure 3G, p < 0.05). These results clearly indicate that maternal exposure to PM during pregnancy have adverse effect on fetal and neonate mouse, which lead to pathological and structural changes in the myelin development, activation of astrocytes and microglia, thus directly contributes to dysfunction of movement coordination ability of mouse offspring. Figure 3 Download asset Open asset PM exposure during pregnant and postnatal delays myelinogenesis in the developing nervous system. (A) Schematics of treatment strategies. In order to model a maternal PM exposure, pregnant mice were pre-treated with PBS or PM (nasopharyngeal inhalation, 5.0 mg/kg/d) daily until parturition. Pups from PBS- or PM-treated group with similar weights were subsequently exposed to PBS or PM at postnatal Days 4–21. Brain was harvested for Q-PCR, immunohistochemistry, or TEM analysis at postnatal day 14, and behavioral evaluation was processed at postnatal Days 20–22. (B) Representative images of myelin content (MBP staining), microgliosis (IBA1 staining), and astrogliosis (GFAP staining) in the body of the corpus callosum of PM- or PBS-treated mice. (C) Quantitative analysis of MBP, IBA1, and GFAP expression using Image-Pro. (D) Representative electron microscopy images of the corpus callosum region isolated from PM- or PBS-treated mice. Arrows heads mark myelin thinning and dysplasia. (E) Quantification of the G-ratios (axon diameter divided by entire myelinated fiber diameter) of myelinated fibers and number of myelination axon by Image-Pro. (F) mRNA relative expression of pro-inflammatory cytokines and myelin protein in corpus callosum of PM- or PBS-treated mice was detected by real-time PCR. (G) Effect of PM on motor balance and motor coordination were determined by behavioral evaluation (rotating rod, beam walking, and tight rope test). Data were collected from three to five separate mouse litters. Symbols represent mean ± SD; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001, compared to PBS-treated group, two-way ANOVA comparison with Tukey's multiple comparisons test. Scale bar = 50 µm in B, Scale bar = 1 µm in D. PM exposure boosts microglial activation and pathogenicity Although little is known about the mechanism of PM exacerbating myelin damage, we have observed a significant increase of activated IBA1+ microglia in the demyelination/myelinogenesis animal models (Figures 1F, G—3B and C). There are similar pathological features in human autopsy study with high levels of air pollution et al., suggesting that activation of microglia be the and pathogenic factor in the process of PM neuroinflammation and demyelination. As the resident immune cells and the CNS microglia to a of environmental or et al., 2017). The activation of microglia under PM exposure in the demyelination/myelinogenesis animal models us to that the have critical roles in the To test this, microglia were stimulated by exposure to using as a control. is well that LPS is a to induce microglia of PM with the microglia even the effect of LPS (Figure of the cells in PBS group were with while under LPS or PM treatment, the cell body and with decreased and activation (Figure to LPS stimulation, PM-treated microglia higher levels of MHC and (Figure of pro-inflammatory factors or is one of the of microglial we also examined the by gene expression and As shown in Figure and D, both PM- and microglia are by expression of pro-inflammatory cytokines and Tnfa, Il6, Il1b, and with remarkably decreased gene expression of and indicating that PM stimulation, similar to microglia activation and cell from a to a pro-inflammatory Figure 4 with 1 supplement see all Download asset Open asset PM exposure boosts microglial activation and pathogenicity. (A) of microglia under the treatment of PM or microglia from C57BL/6 mice were stimulated for with LPS or PM were with and with (B) Effects of PM treatment on the expression of MHC and of microglia were stimulated with LPS or PM for 24 h, and analyzed by flow cytometry. (C) mRNA relative expression of cytokines and cell of or PM-treated microglia was detected by real-time PCR. (D) analysis for of pro-inflammatory and in under PM (E) CD4+ cells were with or microglia under The percentage of Th17 and Th1 cells in the were analyzed by staining of and (F) of OPCs to of in OPCs were in differentiation for 3 and of was by of microglia with PBS LPS or PM for another 4 days. were by specific MBP Quantitative analysis of numbers or of was assessed by using analysis of the of mRNA relative expression of pro-inflammatory cytokines Il6, and of or microglia pre-treated with (10 or (10 for 4 analysis for of in or microglia pre-treated with (10 and (10 for 4 EAE development was evaluated daily for PM- or PBS-treated group with or and administration by two researchers blindly, to a 0–5 scale. Female, 8–10 week-old C57BL/6 mice were immunized with MOG35-55 and pre-treated with PBS or PM (nasopharyngeal inhalation, 5.0 mg/kg/d) daily, starting at day –30 (before immunization) until the end of the mg/kg/d) or mg/kg/d) was by at day p.i. Data were collected from 3 to mice each group. Symbols represent mean ± SD; *p < 0.05; **p < 0.01; ***p < 0.001; two-way ANOVA comparison with Tukey's multiple comparisons test. Scale bar = µm in Scale bar = 50 µm in These us to whether PM exposure of microglia resulted in changes as well as To test this, we assessed the of T cell as well as OPCs to of in As shown in Figure or microglia were with CD4+ T cells in the of Th17 and microglia to LPS stimulation, PM-treated microglia to of and by CD4+ T cells (Figure In addition, OPCs with control differentiation or for with normal of the while the of or PM- by the significantly decreased numbers and of cells (Figure These data indicate that PM and LPS have activities and on the of microglia In addition, we the effects of PM on astrocytes and in Quantitative showed that the expression of and of activated was significantly under LPS treatment compared with the group, but not in the PM-treated group, indicating that PM promote astrocyte activation LPS (Figure 4—figure supplement 1B). compared with the control group, the number of decreased and the number of increased in the PM treatment significantly (Figure 4—figure supplement and and poor structures were observed in the PM treatment (Figure 4—figure supplement 1E). The effect of PM on OPCs differentiation and of processes was as the of is a critical of microglia After activation of on the microglia a of the factor leading to increased expression of pro-inflammatory cytokines such as Tnfa, Il6, As shown in Figure PM treatment, similar to significantly the of and the expression of pro-inflammatory cytokines Il6, and in However, the was by and that the activities of and in the signaling (Figure and In EAE administration of and pharmacological of and the deterioration of the disease resulted from PM exposure (Figure these findings suggest that TLR-4/NF-kB signaling microglia of microglia activation via TLR-4/NF-kB signaling plays important roles in the pathological demyelination, as well as myelination or remyelination under physiological data of PM action on microglia TLR-4/NF-kB signaling as a of microglia pathogenic To evaluate the role of TLR-4/NF-kB in microglia, we first to investigate the effects on gene on microglia by using Figure Results shown that genes were and were under PM compared with PBS control rate < 0.001; Figure As the in PM-treated group a of LPS response genes and inflammatory and with the microglia activation (Figure real-time analysis confirmed a in TLR-4/NF-kB signaling associated such as and (Figure analysis showed that the genes
Source: eLife
Author: Brandon K Harvey
Language: English
Subfield: Neurology
Field: Neuroscience
Domain: Life Sciences
Open Access status: gold
Estimated APC paid: $2,500
OpenAlex ID: W4214874457