Traffic-Generated Air Pollution– Exposure Mediated Expression of Factors Associated with Demyelination in a Female Apolipoprotein E^−/− Mouse Model

Abstract

Epidemiology studies suggest that exposure to ambient air pollution is associated with demyelinating diseases in the central nervous system (CNS), including multiple sclerosis (MS). The pathophysiology of MS results from an autoimmune response involving increased inflammation and demyelination in the CNS, which is higher in young (adult) females. Exposure to traffic-generated air pollution is associated with neuroinflammation and other detrimental outcomes in the CNS; however, its role in the progression of pathologies associated with demyelinating diseases has not yet been fully characterized in a female model. Thus, we investigated the effects of inhalation exposure to mixed vehicle emissions (MVE) in the brains of both ovary-intact (ov+) and ovariectomized (ov−) female Apolipoprotein (ApoE^−/−) mice. Ov+ and ov− ApoE^−/− mice were exposed via whole-body inhalation to either filtered air (FA, controls) or mixed gasoline and diesel vehicle emissions (MVE: 200 PM μg/m³) for 6 hr/d, 7 d/wk, for 30 d. We then analyzed MVE-exposure mediated alterations in myelination, the presence of CD4+ and CD8+ T cells, reactive oxygen species (ROS), myelin oligodendrocyte protein (MOG), and expression of estrogen (ERα and ERβ) and progesterone (PROA/B) receptors in the CNS. MVE-exposure mediated significant alterations in myelination across multiple regions in the cerebrum, as well as increased CD4+ and CD8+ staining. There was also an increase in ROS production in the CNS of MVE-exposed ov− and ov+ ApoE^−/− mice. Ov− mice displayed a reduction in cerebral ERα mRNA expression, compared to ov+ mice; however, MVE exposure resulted in an even further decrease in ERα expression, while ERβ and PRO A/B were unchanged across groups. These findings collectively suggest that inhaled MVE-exposure may mediate estrogen receptor expression alterations associated with increased CD4+/CD8+ infiltration, regional demyelination, and ROS production in the CNS of female ApoE^−/− mice.

Graphical Abstract

1. Introduction.

Outcomes from both human studies and animal models have established an association between exposure to environmental air pollution and detrimental CNS consequences, including neuroinflammation, neurodegeneration, and blood-brain barrier (BBB) disruption (Block et al., 2009, Block et al., 2015; Suwannasual et al., 2018; Cacciottolo et al., 2020; Milani et al., 2020). Air pollution is associated with neurodegenerative diseases such as Alzheimer’s disease (AD) and exacerbation of autoimmune disorders of the CNS, including multiple sclerosis (MS) (Block and Calderón-Garcidueñas, 2009; Calderón-Garcidueñas et al., 2019; Armstrong et al., 2020; Abbaszadeh et al., 2021). Multiple epidemiologic studies have reported that exposure to components of air pollution, including those derived from traffic-generated sources, are strongly correlated to incidence and relapse of MS (Roux et al., 2017; Scartezzini et al., 2020; Elgabsi et al., 2021); however, there are also conflicting studies that report no significant correlation between certain regional components of air pollution and MS incidence (Palacios et al., 2017; Bai et al., 2018; Kazemi Moghadam et al., 2021). While laboratory studies are minimal, exposure to urban particulate matter (PM) has also been shown to promote demyelination through oxidative stress-associated mechanisms in both in vitro and in vivo studies in male mice (Kim et al., 2020).

MS is a chronic autoimmune disease associated with T-cell infiltration and inflammation in the CNS. The immune cells attack the myelin sheath, a protective and insulating fatty layer that originates from oligodendrocytes, resulting in neuronal damage. Myelin oligodendrocyte glycoprotein (MOG) is a minor component of the myelin sheath, believed to act as a cellular adhesion molecule or surface receptor involved in the regulation of oligodendrocyte microtubules to facilitate a complement cascade (Narayan et al., 2018; Wynford-Thomas et al., 2019). MOG has been associated with the destruction of myelin sheath in the CNS of people with MOG antibody disease. While MOG antibody disease is now diagnosed as a distinct demyelinating disease from MS, the progression of MS and relapse rates is reported to increase in patients with antibodies to MOG (Spadaro et al., 2016).

Reactive oxygen species (ROS) and inflammatory signaling are associated with demyelination in the CNS (Ortiz et al., 2013). Overproduction of ROS can accelerate the initiation of a lipid peroxidation cascade, which results in demyelination and the death of oligodendrocytes (Smith et al., 1999). ROS plays a vital function in signaling the molecules that target T cell activation and differentiation, whereas overproduction of ROS causes damage to these biomolecules and cellular organelles, resulting in abnormal function (Rashida Gnanaprakasam et al., 2018).

T cells, CD4+ and CD8+, are part of the adaptive immune system. CD4+ cells bind to major histocompatibility complex (MHC) class II, which are antigen-presenting cells that act as macrophage cells present in human MS lesions, while CD8+ bind to MHC class I and promote a cytotoxic immune response (Denic et al., 2013). During MS, the integrity of the BBB is often impaired, leading to increased translocation of activated T cells, such as CD8+ and CD4+, from the circulation into the parenchyma (Ortiz et al., 2013). The increase in the infiltration of activated T cells is associated with an inflammatory response and activation of microglia, damaging the myelin sheath and promoting neuronal apoptosis (Volpe et al., 2016).

Sex hormones are also believed to contribute to the onset or progression of MS. The highest prevalence of MS is reported in young females (ages 20–40 yrs.), with diagnosis rates approximately 2–3 times higher than that reported in men (Noonan et al., 2002). Also, relapses are reported to decline by 80% in the third trimester of pregnant MS patients when estrogen and progesterone levels are elevated (Harbo et al., 2013). Increased expression of estrogen receptor alpha (ERα) and estrogen receptor beta (ERβ) receptors has been associated with reduced demyelination and axonal loss in the EAE animal model of MS (Maglione, 2019). Moreover, estrogen has also been reported to have neuroprotective effects, as treatment with an ERβ agonist resulted in improved clinical outcomes in MS and provided neuroprotection in EAE (Spence et al., 2013). Together, these outcomes suggest that alterations in sex-steroid hormone production and signaling likely play an essential role in the etiology and pathology of the disease. This is further confirmed by the premise that pregnancy appears to provide “protective effects” from relapse in MS patients via an estrogen-mediated reduction in proinflammatory cytokine expression (Soldan et al., 2003).

To date, very little information exists on the effects of traffic-generated pollutant-exposures in the CNS of females. We have recently reported that exposure to a mixture of gasoline and diesel vehicle emissions (MVE) mediates alterations in BBB integrity and increased inflammation in the CNS of both ovariectomized (ov−) and ovary-intact (ov+) female Apolipoprotein (Apo) E^−/− mice (Adivi et al., 2021). Furthermore, we have also reported MVE-exposure mediates neuroinflammation and elevated ROS production in male ApoE^−/− and C57Bl/6 mice (Oppenheim et al., 2013; Lucero et al., 2017; Suwannasual et al., 2018). As the pathogenesis of multiple CNS disorders, including MS, is associated with BBB disruption, we investigated the hypothesis that MVE-exposure mediates CD4+/CD8+ immune cell infiltration, ROS production, and demyelination in female ApoE^−/− mice. Thus, the current study aims to characterize whether exposure to traffic-generated air pollutants promotes outcomes in the CNS associated with the progression or exacerbation of demyelinating disease states. Notably, we also utilized both ov− and ov+ animals to investigate sex-hormone-associated responses in the CNS related to these outcomes.

2. Materials and methods.

2.1. Animals and Inhalation Exposure Protocol

Thirty-two female Apo E^−/− mice, aged 6–8 weeks old, were obtained from Taconic (Albany, NY); 16 were ovariectomized (ov−), and 16 were ovary-intact (ov+). All mice were placed on a high-fat “Western” diet (TD88137 Custom Research Diet, 21.2% fat, 1.5g/kg cholesterol diet; Harlan Teklad, Madison, WI) for two weeks before the beginning of exposures and were maintained on the same diet throughout the 30-day exposure. Mice were separated and designated to receive either mixed vehicle emissions (MVE: 200 PM μg/m³), which was created from a mixture of vehicle exhaust generated by GM gasoline engine (50 PM μg/m³) and a Yanmar diesel generator system (~150 PM μg/m³) or filtered air (FA, controls) for 6 hr/d, 7 d/wk, for 30 d, as previously described (Lund et al., 2011). Briefly, exhausts from both engines were combined after their primary dilutions into a 2m³ mixing chamber. The proportion of the diesel engine exhaust was balanced based on particle mass concentration, since the majority of the particle mass came from that engine source, while the gasoline engine exhaust was balanced by measuring the carbon monoxide (CO) concentration (Lund et al., 2011). Characterization and measurement of gases and PM were conducted daily throughout the entire exposure protocol, and particle mass concentration was measured gravimetrically using Teflon membrane filters (Lund et al., 2011). Particle size distribution was measured with a fast mobility particle sizer (TSI, St. Paul, MN). The median MVE PM size for this study was approximately 60nm, while the PM mass size distribution median was approximately 1μm (range: < 0.5–20μm). The general characteristics of the MVE exposures are shown in Table 1; MVE chemical characteristics have been previously reported (Mumaw et al., 2016). While higher than most environmental exposure scenarios would likely be, the 200 μg/m³ PM concentration of MVE was chosen for the current study and a “midpoint” dose between previously reported findings in our laboratory from both human exposure studies (Lund et al., 2011) and male mice at the 100 μg/m³ PM and 300 μg/m³ PM concentration of MVE (Lund et al., 2011; Oppenheim et al., 2013; Suwannasual et al., 2018). Furthermore, while this concentration would be higher than environmental levels in most urban regions, it is not outside of concentrations often seen in heavily populated and industrial areas worldwide (Pronk et al., 2009; Costa et al., 2017; IQAir).

Table 1.

General characteristics of exposure atmosphere.

	PM (μg/m3)	CO (ppm)	NOx (ppm)	NO2 (ppm)
Filtered air control (FA)	10.1 ± 7.7	0.2 ± 0.3	0.1 ± 0.1	0.0 ± 0.0
Mixed gasoline and diesel engine exhausts (MVE)	204.9 ± 32.9	67.4 ± 7.2	19.6 ± 2.2	3.2 ± 0.7

Data represented as exposure averages ± standard deviation (SD) over the entire exposure period. PM, particulate matter; CO, carbon monoxide; NO_x, nitrogen oxides; NO₂, nitrogen dioxide; ppm, particles per million.

Mice were singly housed to monitor food intake and track estrous stages via vaginal swab cytology; mice were then randomly assigned study groups while ensuring similarities across the estrous cycles in control vs. exposure groups. All mice were housed in standard shoebox cages within an Association for Assessment and Accreditation of Laboratory Animal Care International approved rodent housing facility (2m³ exposure chambers) for the duration of the study, which were maintained within a constant temperature (20–24°C) and humidity range (30–50% relative humidity). Mice had access to chow and water ad libitum throughout the study period, except during daily exposures when chow was removed. All animal protocols were approved by the Lovelace Respiratory Research Institute’s Animal Care and Use Committee (AAALAC-accredited Assurance #A3083-01; USDA-registered facility #85-R-003) and followed the Guide for the Care and Use of Laboratory Animals issued by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996).

2.2. Tissue Collection

At the end of the assigned exposure protocol, the animals were weighed (Table S1), anesthetized with Euthasol®, euthanized by exsanguination, decapitated, and brain tissues were collected. The meninges were gently removed from the brains, the brains weighed, dissected (coronal plane/cut at roughly Bregma 0 – Bregma −2.92 mm), and a portion was fixed in HISTOCHOICE (VWR, Irving, TX) at 4°C overnight. Fixed tissue was then rehydrated in 30% sucrose/PBS (weight/vol) at 4°C overnight and embedded in Tissue Freezing Medium (TBS, IMEB Inc., San Marcos, CA) for sectioning. The remaining regions of the brain not fixed for histology were frozen in liquid nitrogen and stored at −80°C for future molecular assays.

2.3. Cerebral Myelin Staining

Embedded frozen brain tissues were cut on a cryostat in serial 10μm sections (approximate bregma −1.65mm to −2.12 mm), placed on Superfrost™ Plus slides (ThermoFisher, Richardson, TX), and stored at −80°C. A Brain – Stain™ Imaging Kit (ThermoFisher #B34650) was utilized for myelin staining, following the manufacturer’s recommended protocol. Green (GFP) was used to stain myelin, and red (RFP) was used for Nissl staining. Stained slides were imaged at 10x, 20x, and 40x using the inverted fluorescent microscope with epifluorescence optics (EVOS Fl, ThermoFisher) and analyzed with ImageJ software (NIH, Bethesda, MD). Slides with no primary antibody treatment were used as negative controls. N = 5 animals for each group, 2 slides each animal, with 2 sections on each slide were used for quantification and analysis. For microscopy and analysis, brains were divided into six regions (Fig. 1). Region 6 was excluded from myelin staining quantification. The same area and size were measured in each region for quantification in Image J. All myelin staining is reported as total fluorescence staining per unit area for the defined region (Fig. 1).

Figure 1.

Schematic of regions used for imaging/analysis of the cerebrum of ApoE^−/− mice for histological endpoints.

2.4. CD4+/CD8+ Staining and Quantification

Serial sections of 10 μm frozen cerebral tissue (approximate bregma −1.65mm to −2.12 mm) were processed through immunohistochemistry for CD4+ or CD8+ cell expression. Briefly, sections were treated in cold acetone for 30 min at RT and rinsed with 1X PBS 3 times. Then slides were blocked with a solution of 1X PBS-T, and Bovine Serum Albumin (60 mg/2 ml vol/vol) for 1 hr at RT. Tissue sections were then incubated with biotinylated primary antibodies CD4+, and CD8+ (Affymetrix eBioscience/ThermoFisher #36-0041-85 and 36-0081-85, respectively) in a dark chamber overnight at 4°C. Samples were rinsed 3 times with 1X PBS, and the substrate was added using an ABC detection kit (#PK-6100; Vector Laboratories, Burlingame, CA), following the manufacturer protocol, and incubated for 1 hr in the dark chamber at RT. Slides were developed by using a Vector Red substrate kit (#SP-5100 0; Vector Laboratories), following the manufacturer protocol. Slides were then counterstained with hematoxylin and cover-slipped. Sections were imaged by microscopy at 10X and 40X, from an n=5 from each group, 2–3 slides per animal, with 2 sections each. Histological analysis was conducted in regions of the brain, as described in the previous section and shown in Fig. 2-1. Image J cell counter software was used to quantify the expression of CD8+ and CD4+ (red staining) in each defined region. Samples without primary antibodies were used as a negative staining control to confirm staining specificity.

Figure 2.

Representative staining of myelination in the cerebral cortex (regions 1–4) of female ApoE^−/− mice, on a high fat diet, with either ovaries (ov+) or ovariectomized (ov−) and exposed to either filtered air (FA, control A - C ovary intact; and G - I ovariectomized) or mixed gasoline and diesel vehicle exhaust (MVE: exposed; 200 PM μg/m³ for 6 hrs/d, 7d/wk for 30 d); (D - F ovary intact; J - L ovariectomized). Green stain represents myelin, red stain represents Nissl staining, and the overlay represents neurons with the myelin sheath. (M) Graph of normalized myelination in cortex represented; results represent mean ± S.E. Quantification per unit area; the unit of the area measured was kept consistent within each region quantified, across all animals in the study. Scale bar = 100 um. N = 5 animals for each group, 2 slides for each animal, 2 sections on each slide were used for quantification and analysis. *p≤0.050 compared to FA ov+ group; †p≤0.050 compared to MVE ov+; ‡ p≤0.050 compared to FA ov−.

2.5. Real-time RT-qPCR

RNA was isolated from cerebral tissue using a RNeasy Mini kit (Qiagen, Valencia, CA), per kit instructions. cDNA was synthesized using an iScript cDNA Synthesis kit (BIORAD, Hercules, CA; Cat. #170-8891). Real-time PCR analyses of ERα, ERβ, progesterone receptors A/B (PRO A/B), GPx, or GAPDH (house-keeping gene) were conducted using specific primers (Table 2) and SYBR Green detection (SSo Advanced Universal SYBR Green Supermix, Biorad, Hercules, CA; Cat #172-5271), following the manufacturer’s protocol. Real-time PCR analyses were run on a Biorad CFX96 Touch™ Real-Time PCR Detection System (BioRad). ΔΔCT values measured using CFX Manager™ Software and normalized to GAPDH on triplicate samples, as previously described by our laboratory (Lund et al., 2009). Results are expressed as mean normalized gene expression as a percentage of GAPDH controls.

Table 2.

Primer sets utilized for real-time RT-qPCR.

Gene/Primer	Sequence (5’ – 3”)
Mouse ERα FP	TGGGCTTATTGACCAACCTAGCA
Mouse ERα RP	AGAATCTCCAGCCAGGCACAC
Mouse ERβ FP	GACTGTAGAACGGTGTGGTCATCAA
Mouse ERβ RP	CCTGTGGAGGTAGGAATGCGAAAC
Mouse GAPDH FP	CATGGCCTTCCGTGTTCCTA
Mouse GAPDH RP	GCGGCAGTCAGATCCA
Mouse GPx FP	AAGCAACCCAGCCTTTTCTC
Mouse GPx RP	TGAGCATTCTCCTTTGGAC
Mouse MOG FP	GCTAATTGAGACCTATTTCTC
Mouse MOG RP	AGCAATAAACAGGTGGAAGGTC
Mouse PRO A/B FP	GGTGGGCTTCCTAACGAG
Mouse PRO A/B RP	GACCACATCAGGCTCAACGAG

FP, forward primer; RP, reverse primer; ERα, estrogen receptor alpha; ERβ, estrogen receptor beta, PRO A/B, progesterone receptors A/B, GAPDH, Glyceraldehyde 3-phosphate dehydrogenase; MOG, myelin oligodendrocyte glycoprotein, GPx, glutathione peroxidase

2.6. Dihydroethidium (ROS) Staining

Dihydroethidium (DHE) was conducted on 20μm thick cerebral sections (approximate bregma −0.9 mm to −1.64 mm), imaged, and quantified, as previously described by our laboratory (Oppenheim et al., 2013). Sections were imaged by fluorescent microscopy at 10X from an N=5 from each group, 2–3 slides per animal, with 2 sections each through all 6 regions. We quantified average DHE expression in the cortex of female ApoE^−/− (per unit area), using Image J software, in all 6 regions.

2.7. Double Immunofluorescence

10 μm frozen serial sections of the cerebrum (approximate bregma −1.65mm to −2.12 mm) for immunofluorescent labeling, as previously described by our laboratory (Suwannasual et al., 2018), utilizing primary antibody estrogen (1:500; Abcam, Cambridge, MA, #168986), and von Willebrand factor (vWF: 1:1000; Abcam #11713). Alexa-Fluor 488 (1:250, Thermo Fisher Scientific #A32731) and Alexa Fluor 555 (1:250) were used for the secondary antibody. Slides were imaged by EVOS fluorescent microscopy (EVOS Fl, Thermo Fisher Scientific) at 40x with the proper excitation/emission filter and digitally recorded. Images were analyzed by Image J software (NIH, Bethesda, MD) by a blinded technician. Colocalization was measured by analyzing total fluorescence from at least 4–5 vessels (less than 50 μm in size) from the overlaid images (2 sections per slide, 2 slides per animal, and n= 5 per group).

2.8. Statistical Analysis

Data are shown as mean ± SEM. Sigma Plot 10.0 was used to analyze all statistical endpoints. A two-way ANOVA with Tukey post-hoc analysis was used for statistical comparisons between the exposure and/or ovary +/− groups and the exposure × ovary status interaction, as indicated in the results and figure legends. Data is represented as mean ± standard error (S.E.). A p≤0.050 was considered a statistically significant difference for all measured endpoints.

3. Results.

3.1. Quantification of myelination in the cerebrum of female ovary-intact and ovariectomized ApoE^−/− mice exposed to MVE.

As demyelination is a hallmark of MS, we assessed the effects of MVE-exposure on myelination in the cerebrum (regions 1–5, as indicated in Fig.1) of our female animals. Compared to ovary intact (ov+) FA controls (Figs. 2A–C), we observed a decrease in myelin staining in the cortex region of MVE-exposed female ov+ ApoE^−/− mice (Figs. 2D–F). Statistical analysis for demyelination in the cortex related to exposure, F(1,27)= 19.095; p<0.001, ovary status F(1,27) = 38.950; p<0.001; and, for exposure × ovary interaction F(1,27)=0.112; p= 0.742. Furthermore, we also observed a decrease in myelination in the cerebrums of both FA ov− (Figs. 2 G–I; p= 0.030) and MVE ov+ (Figs. 2 J – L; p<0.001) animals, compared to FA ov+. Graphic representation of myelination in the cortex of ApoE^−/− mice are shown in Fig. 2M. Compared to FA ov+ (Figs. 3A–C), quantification of myelination in the corpus callosum (region 5, as indicated in Fig.1) of female ApoE^−/− mice showed no significant alterations in myelination in the MVE-exposed ov + (Figs. 3 D–F), FA ov− (Figs.3 G–I), nor MVE ov− (Figs. 3J–L) animals, as indicated in Fig. 3M. Analysis of myelin staining across different regions of the cerebrum revealed a significant decrease in myelination in the MVE-exposed groups, regardless of ovary status, compared to the FA groups, across regions 3 and 4 (Figs. 4A and 4B, respectively).

Figure 3.

Representative myelination in the corpus callosum of female ApoE^−/− mice (region 5) on a high-fat diet, with either ovaries (ov+) or ovariectomized (ov−) and exposed to either filtered air (FA, control A - C ovary intact; and G - I ovariectomized) or mixed gasoline and diesel vehicle exhaust (MVE: exposed; 200 PM μg/m³ for 6 hrs/d,7d/wk for 30 d); (D - F ovary intact; J - L ovariectomized). Green stain represents myelin, red stain represents Nissl staining, and the overlay represents neurons with the myelin sheath. Quantification per unit area; the unit of the area measured was kept consistent within each region quantified, across all animals in the study. (M) Graph of normalized myelination in cortex represented; results represent mean ± S.E. Scale bar = 100 um. N = 5 animals for each group, 2 slides for each animal, 2 sections on each slide were used for quantification and analysis.

Figure 4.

Quantification of normalized myelination in (A) region 3, and (B) region 4 in the cerebrum of female ApoE^−/− mice, on a high-fat diet, with ovaries (ov+) or ovariectomized (ov−) and exposed to either filtered air (FA control) or mixed exhaust (MVE: mixed gasoline and diesel emissions, 200 PM μg/m³ for 6 hrs/day, 7 d/wk, for 30 d). Scale bar = 100 um. N = 5 animals for each group, 2 slides each animal, 2 sections on each slide were used for quantification and analysis. Results represent mean ± S.E. * p≤0.050 compared to FA ov+; †p≤0.050 compared to MVE; ‡ p≤0.050 compared to FA ov−.

3.2. Expression of CD4+ and CD8+ cells in the cerebrum of female ApoE^−/− mice exposed to MVE.

MS lesions in the brain are typically associated with increased expression of CD4+ and CD8+ cells. Therefore, we analyzed the expression of the T lymphocyte subtypes via immunohistochemistry. Compared to FA ov+ (Fig. 5A), we observed a significant increase of CD4+ cells in the cerebrum in both the MVE ov+ (Fig. 5B) and MVE ov− (Fig. 5D) female ApoE^−/− animals, as presented in Fig. 5E. Statistical comparisons for exposure was F(1,31)=32.989; p<0.001; for ovary status was F(1,31)=2.498; p=0.130; for exposure × ovary status interaction F(1,31)=0.0917; p=0.765. No differences were noted in cerebral CD4+ levels between FA ov+ (Fig. 5A) and FA ov− (Fig. 5C) female ApoE^−/− mice. Immunohistochemical staining of cerebral tissue was also performed to analyze the expression of CD8+. Compared to FA ov+ (Fig. 6A), there was a significant increase of CD8+ cells in the cerebrum with MVE-exposure in both the MVE ov+ (Fig. 6B, p=0.010) and MVE ov− (Fig. 6D; p<0.001) female Apo E^−/− animals, as presented in Fig. 6E. Statistical comparisons for the exposure was F(1,47)= 9.564, p=<0.001; for ovary status F(1,47)= 0.773, p=0.525; for the exposure × ovary interaction F(1,47)=0.0746, p=0.788. No differences were noted in cerebral CD8+ levels between FA ov+ (Fig. 6A) and FA ov− (Fig. 6C) female ApoE^−/− mice.

Figure 5.

Representative CD4+ staining in the cerebrum of female ApoE^−/− mice with ovaries (ov+; A, B) or ovariectomized (ov−; C, D), and exposed to either filtered air (FA: A, C) or mixed vehicle exhaust (MVE: B, D; mixture of gasoline and diesel emissions at 200 PM μg/m³ for 6 hr/d, 7 d/wk for 30 d). (E) Displays quantification of CD4+ expression per unit area in the cerebrum, results represent mean ± S.E. N = 5 animals for each group, 2 slides each animal, 2 sections on each slide were used for quantification and analysis. Scale bar = 100 μm. *p≤0.050 compared to FA ov+; ‡ p≤0.050 compared to FA ov−.

Figure 6.

Representative CD8+ staining in the cerebrum of female ApoE^−/− mice with ovaries (ov+; A, B) or ovariectomized (ov−; C, D), and exposed to either filtered air (FA: A, C) or mixed vehicle exhaust (MVE: B, D; mixture of gasoline and diesel emissions at 200 PM μg/m³ for 6 hr/d, 7 d/wk for 30 d). (E) Displays quantification of CD8+ expression per unit area in the cerebrum, results represent mean ± S.E. N = 5 animals for each group, 2 slides each animal, 2 sections on each slide were used for quantification and analysis. Scale bar = 100 μm. *p≤0.050 compared to FA ov+; ‡ p≤0.050 compared to FA ov−.

3.3. Expression of estrogen and progesterone receptors in ovary-intact and ovariectomized female ApoE^−/− exposed to MVE.

Female sex hormone signaling may be a factor associated with the increased occurrence of MS in females. As such, we analyzed the expression of estrogen and progesterone receptors in the cerebrum of our study animals. ERα receptor expression in the cerebral microvasculature of female ApoE^−/− mice was not significantly altered in response to MVE-exposure (Figs. 7A–L), as quantified in Fig. 7M. However, real-time RT-qPCR analysis of ERα mRNA expression from the cerebral tissue of MVE-exposed female ApoE^−/− mice revealed a decrease in expression of ERα receptor in MVE ov+ and MVE ov− animals, which was also observed in the FA ov− animals, compared to FA ov+ controls (Fig. 7N). Statistical analysis showed for exposure F(1,27) = 4.671, p=0.042; for ovary status F(1,27)= 17.680, p <0.001. and for the exposure × ovary status interaction F(1,27)= 3.641, p=0.070. Neither ERβ nor progesterone A/B receptors showed any statistical change in cerebral mRNA expression across exposure and/or control groups (Figs. 8A and 8B, respectively).

Figure 7.

Representative staining of cerebral expression of estrogen receptor (ER) α in the cerebral microvasculature of female ApoE^−/− mice on a high-fat diet, with either ovaries (ov+) or ovariectomized (ov−) and exposed to either filtered air (FA, control A - C ovary intact; and G - I ovariectomized) or mixed gasoline and diesel vehicle exhaust (MVE: exposed; 200 PM μg/m³ for 6 hrs/d,7d/wk for 30 d); (D - F ovary intact; J - L ovariectomized). Green fluorescence indicates ERα and red fluorescence represents vonWillebrand factor (vWF), used as an endothelial cell marker to identify microvessels in the cerebrum. Merged panel = microvascularspecific ERα expression. (M) Quantification of cerebral microvascular expression of ERα, n=5, 2 sections each, 5 vessels per section were used for analyses (only vessels smaller than 100μm were used for analyses). Scale bar = 100 μm. (N) Mean normalized gene expression of cerebral ERα mRNA expression, as quantified by real time RT-qPCR (n=6 per group). *p≤0.050 compared to FA ov+ group, †p≤0.050 compared to MVE ov+.

Figure 8.

Mean normalized gene expression of (A) estrogen receptor (ER) β, and (B) progesterone receptors (PRO) A/B in female ApoE^−/− mice with ovaries (ov+) or ovariectomized (ov−), on the high-fat diet, exposed to either mixed vehicle emissions (MVE: 200 PM ug/m³ for 6 hrs/day, 7 d/wk, 30 d) or filtered air (FA), as quantified by real time RT-qPCR. N=6 per group.

3.4. Expression of ROS in ovary-intact and ovariectomized female ApoE^−/− exposed to MVE.

Since previous studies have reported demyelination associated with increased ROS production, dihydroethidium (DHE) staining of cerebral tissue was performed to analyze ROS production resulting from MVE-exposure. Compared to FA ov+ (Fig. 9A) and FA ov−(Fig. 9C), we observed a significant increase of ROS production in the cerebrum in both the MVE ov+ (Fig. 9B) and MVE ov− (Fig. 9 D) groups, as presented in Fig. 9E. Statistical analysis show for exposure F(2,38) = 6.506, p=0.014; for ovary status F(2,38)= 1.333, p=0.254 and the for exposure × ovary status interaction F(2,38)= 1.242, p=0.271 No differences were noted in cerebral ROS production between FA ov+ (Fig. 9 A) and ov− (Fig. 9C) female Apo E^−/− mice. In agreement with these findings, we also quantified antioxidant glutathione peroxidase (GPx) mRNA levels, which showed a significant decrease in expression in the cerebrum of MVE-exposed females, regardless of ovary status (Fig S1).

Figure 9.

Representative DHE staining in the cerebrum of female ApoE^−/− mice with ovaries (ov+; A, B) or ovariectomized (ov−; C, D), and exposed to either filtered air (FA: A, C) or mixed vehicle exhaust (MVE: B, D; 200 PM μg/m³ for 6 hr/d, 7 d/wk for 30 d). (E) Quantification of DHE fluorescence per unit area; results represent mean ± S.E. N= 6 per study group, 2 slides per animal, 2 sections on each slide. Scale bar = 400 μm. *p≤0.050 compared to FA ov+ group.

3.5. Cerebral mRNA expression of myelin oligodendrocyte in female ApoE^−/− mice.

An increase in MOG expression is thought to be involved in the process of demyelination; therefore, we quantified MOG mRNA levels in the brains of our study animals. Our results showed the MOG was increased at the transcript level in the cerebrum of the MVE -exposed groups compared to FA groups, regardless of the ovary status (Fig. 10). Statistical analysis shows for exposure F(1,20)= 4.794, p=0.045; for ovary status F(1,20)= 0.199, p=0.662, and for the exposure × ovary status interaction F(1,20) = 0.0820, p=0.779.

Figure 10.

Mean normalized cerebral expression of MOG in female ApoE^−/− mice with ovaries (ov+) or ovariectomized (ov−), on the high-fat diet, exposed to either mixed vehicle emissions (MVE: 200 PM ug/m³ for 6 hrs/day, 7 d/wk, 30 d) or filtered air (FA), as determined by real time RT-qPCR. N=6 per study group. *p≤0.050 compared to FA ov+ group.

4. Discussion.

The World Health Organization (WHO) has reported that air pollution exposure poses a significant public health risk, as it is estimated to be responsible for 7 million deaths per year (WHO, 2018). Exposure to traffic-generated air pollution has previously has been characterized to contribute to detrimental outcomes in the CNS, including stroke, neuroinflammation, and neurodegeneration (Block and Calderón-Garcidueñas, 2009). However, there are currently conflicting reports in the literature about whether exposure to air pollution is associated with the incidence or exacerbation of demyelinating diseases, such as MS (Angelici et al., 2016; Palacios et al., 2017; Roux et al., 2017; Bai et al., 2018; Scartezzinni et al., 2020; Elgabsi et al., 2021; Kazemi Moghadam et al., 2021). Moreover, even less is understood regarding the role of steroid hormone-receptor contributions in these air pollution exposure-mediated outcomes. Therefore, we investigated whether inhalation exposure to traffic-generated pollutants promoted the induction of factors in the CNS associated with the demyelination in a female mouse model. The rationale for utilizing female ApoE^−/− mice (+/− ovaries) in the current study is that we have recently reported that MVE-exposure mediates alterations in BBB integrity and increased neuroinflammation in the CNS of female ApoE^−/− mice, regardless of the presence of female hormones (ov+ vs. ov−) (Adivi et al., 2021). Moreover, since alterations in BBB have been reported to mediate the pathogenesis of MS, we wanted to investigate whether further pathophysiological changes also occur in the female CNS following MVE-exposure. Furthermore, these studies allow us to characterize whether sex-specific differences are occurring in the CNS of females vs. males, as we have previously characterized these same exposure scenarios and endpoints in the brains of male ApoE^−/− mice (Oppenheim et al., 2013; Lucero et al., 2017). Notably, we observed a significant increase in the body weight of both the FA and MVE ov− groups compared to the FA ov+ group; however, there were no differences in brain weights across any groups (Table S1). While we have previously reported that MVE exposure promotes increased adiposity and body weights in male C57Bl/6 mice (Phipps et al., 2021), the increase in body weight observed in this study appears to primarily result from loss of ovarian hormone signaling, as the increased body weight noted in the MVE-exposed animals was not statistically different from their respective FA controls. The difference in outcomes of MVE exposure on body weight across studies may be due to multiple factors, including different exposure durations, mouse strains, and sex-specific responses. However, weight gain in ovariectomized mouse models, especially when coupled with a HF diet, has been noted across several previous studies (Riant et al., 2009; Camporez et al., 2013; Della Torre et al., 2021; Varghese et al., 2021).

Findings from our study suggest that inhalation exposure to traffic-generated air pollutants may promote demyelination in certain regions of the brain. Furthermore, the presence or absence of ovaries, and presumably, sex hormone signaling, appeared to play a role in the degree of MVE-mediated demyelination observed, as evidenced by decreased levels of myelination observed in the brains of MVE-exposed ov−, compared to MVE-exposed ov+ female mice. Additionally, within the same regions that we observed a decrease in myelination, there was a concurrent increase in CD4+ and CD8+ cells, which are known to be upregulated in MS lesions (Chitnis, 2007; Pirko et al., 2012). CD8+ are cytotoxic cells that stimulate apoptotic signaling of the myelin sheath, while CD4+ cells are reported to be involved in the direct destruction of the myelin sheath (Patel and Balabanov, 2012). Interestingly, the reduction in myelination was more evident in the brains of ovariectomized mice (MVE ov− and FA ov−), while the increase in the presence of cerebral CD4+ and CD8+ cells appeared to be more strongly correlated to the exposure itself, as there were no statistical differences noted between these endpoints in the cerebrums of MVE ov− vs. MVE ov+ mice. As such, additional studies are necessary to determine the time course of CD4+/CD8+ infiltration into the cerebrum and resulting demyelination from MVE exposures.

MVE-mediated demyelination varied across different regions of the brain in our study. We measured higher degrees of demyelination across regions 3 and 4 of the brain (as indicated in Figs. 1 and 2) compared to other areas of the brain. Importantly, regions 3 and 4 are associated with structures including the basolateral and amygdaloid nucleus, pyriform (olfactory), lateral olfactory tract (LOT), lateral cortex ventricle, and subthalamic nucleus. Likewise, outcomes of neuroinflammation studies in the CNS, resulting from prolonged exposure to diesel exhaust, report that regions of the brain, including the frontal cortex, hippocampus, cerebellum, striatum, and olfactory bulb, are more susceptible to induced proinflammatory signaling with exposure to air pollution (Gerlofs-Nijland et al., 2010). Thus, it is plausible that different brain regions are more “susceptible” to detrimental outcomes in the CNS, associated with the progression of MS, resulting from inhalation exposure to environmental air pollutants.

In addition to demyelination and increased infiltration of CD4+ and CD8 + cells in the CNS, MVE-exposure also significantly increases cerebral MOG mRNA expression, compared to FA control groups, regardless of ovary status. As previously mentioned, MOG is exclusively present in the CNS and has been reported to play a role in demyelination. The presence of MOG antibody (MOG-Abs) is a biomarker used for both the diagnosis of neuromyelitis optica spectrum disorder (NMOSD) and MS (Hacohen and Banwell, 2019). Thus, MVE-mediated increases in the cerebral expression of MOG mRNA expression may be correlated with the altered myelination observed associated with these exposures.

We also observed a significant increase in ROS production in the cerebrum of MVE-exposed ov− and ov+ female ApoE^−/− mice. In MS, increased ROS production is associated with demyelination and axonal damage. This premise is further confirmed by reports showing elevated ROS in the CNS of both EAE mice and MS patients (Witherick et al., 2011). Increased ROS production has been correlated with the incidence and severity of MS, likely due to the activation of phagocytosis of the myelin sheath (Van der Goes et al., 1998). One possible source of ROS production and inflammatory signaling in the CNS is microglial activation (Lassmann and Hossen, 2011). While we have previously reported MVE-exposure mediates increased inflammatory signaling in the CNS of female ApoE^−/− mice (Adivi et al., 2021), and exposure to environmental air pollutants have been shown to directly result in the activation of microglia in both animal and human studies (Campbell et al., 2009; Santiago-López et al., 2010; Mumaw et al., 2016), further studies are required to determine if microglial activation is involved in the ROS production observed in the current study.

Analysis of female steroid hormone receptors in the cerebrum of our study animals showed a reduction in the expression of ERα mRNA in our ovariectomized groups (both FA and MVE). Interestingly, we also observed a significant decrease in ERα in the MVE-exposed ov+ mice. However, ERβ expression in cerebral microvascular of female ApoE^−/− mice did not show a significant change in response to MVE inhalation or even though the presence of ovaries. Estrogen receptors are nuclear transcriptional proteins that regulate the immune system and are implicated in altering innate and adaptive immune responses via the activation of dendritic cells and toll-like receptor (TLR) signaling (Kovats, 2015). Downregulation of ERα is associated with increased tumor necrosis factor (TNF)-α expression and macrophage infiltration, which may contribute to the increased occurrence of autoimmune diseases in females compared to males (Panchanathan et al., 2010). Signaling via the ERα has been reported to provide protective effects in the CNS via inhibiting inflammatory cell recruitment in EAE mice (Subramanian et al., 2003). It is plausible that the observed decrease in ERα receptor mRNA expression in the brain of MVE-exposed ov+ mice may result from negative feedback due to the presence and signaling of estrogen-mimetics in environmental air pollutants. These include compounds such as polycyclic aromatic hydrocarbons (PAHs), metalloestrogens, phthalates, and alkylphenols (Fucic et al., 2012), many of which have been shown to act as ligands at the estrogen receptors (Carpenter et al., 2002). Although we did not observe any MVE exposure-mediated change in expression of PRO A/B receptor mRNA, progesterone can also contribute to the pathophysiology of MS through inhibition of T helper cells (Hughes, 2012). Additional mechanistic/inhibitor studies can help characterize the role of the female sex hormones and/or receptor-mediated signaling in mediating effects of inhaled environmental air pollutants in autoimmune and demyelinated disorders in the CNS.

It is important to note that when exposed to environmental air pollutants via inhalation, there are at least two plausible routes by which pollutants can exert negative outcomes in the CNS, (1) indirectly (systemically), by crossing the respiratory barrier and BBB, as well as (2) directly, through interactions at the olfactory bulb epithelium and olfactory nerve (Jankowska-Kieltyka et al., 2021). Previous studies from our laboratory, as well as others, show that exposure to MVE via inhalation promotes reactive factors in the circulation that promote BBB disruption and increased permeability through studies that utilize plasma from study animals in a BBB co-culture model (Oppenheim et al., 2013; Suwannasual et al., 2019). Alteration in BBB integrity is known to allow for increased transport of xenobiotics, inflammatory, and immunologic factors into the parenchyma that can promote detrimental outcomes in the CNS (Costa et al.,2017). However, multiple studies have also shown that PM air pollution in the ultra-fine and PM_2.5 range can enter the brain via the olfactory mucosa and afferent olfactory neurons, leading to induced inflammatory signaling (Lucchini et al., 2012; Garcia et al., 2015; Wang et al., 2017). Furthermore, air pollution exposure has been associated with olfactory bulb pathology, inflammation, and dysfunction, in youth and adult populations living in heavily polluted urban areas (Calderón-Garcidueñas et al., 2010). As such, exposure to inhaled air pollutants likely leads to pathophysiological outcomes in the CNS, including the promotion of demyelination, through both direct and indirect pathways.

While utilizing female ApoE^−/− mice for the current study allows us to compare MVE-exposure mediated outcomes in the CNS to our previous studies in male mice (Oppenheim et al., 2013; Lucero et al., 2017; Suwannasual et al., 2018), which is of great importance for understanding exposure and toxicity outcomes across sexes, it can also be viewed as a limitation of the current studies. The ApoE^−/− mice in this study were placed on a high-fat diet to initiate atherosclerosis in this mouse model in order to also characterize the effects of air pollution exposure in the progression of vascular disease in females to compare to previously reported outcomes in male mice (Lund et al., 2009; Lund et al., 2011; Oppenheim et al., 2013; Lucero et al., 2017; Suwannasual et al., 2018). However, previous studies suggest that consuming a high-fat diet can also exacerbate inflammation in the CNS (Timmermans et al., 2014); thus, the use of high-fat diet in the current study design adds an additional variable to the reported outcomes. With increased consumption of dietary fats, referred to as the “Western diet” in many populations worldwide, it is also necessary to characterize how diet may serve as an additional insult that exacerbates environmental exposures (or vice versa). The concentration of MVE used for the exposures was chosen to compare to previous studies and endpoints in the CNS from male mouse exposure studies (Oppenheim et al., 2013; Suwannasual et al., 2018; Suwannasual et al., 2019). While these concentrations are high compared to most daily human environmental exposure scenarios, they may occur near roadways during peak traffic hours, industrial or occupational settings, and highly populated urban regions (Lin et al., 2018; WHO Database, 2018; IQAir). Nevertheless, this study provides a foundation for future studies in additional animal models and exposure scenarios/time points to characterize different mechanisms involved in air pollution-mediated pathogenesis of demyelinating diseases, such as MS.

5. Conclusion.

In conclusion, inhalation exposure to MVE promoted demyelination in the cerebrum of female Apo E^−/− mice. Additionally, the degree of demyelination appears to be mediated by the presence of female sex hormones, as demyelination was exacerbated in the cerebrum of MVE-exposed ovariectomized female Apo E^−/− mice. Infiltration of CD4+ and CD8+ immune cells and increased ROS production in the CNS appeared to be primarily related to MVE exposure, as no significant differences were noted between MVE ov+ and ov− exposure groups. Measurement of estrogen receptor expression in the cerebrum revealed a decrease in ERα receptor expression with MVE-exposure; however, neither the ERβ receptor nor the PRO A/B receptor mRNA expression was statistically different across exposure groups. Collectively, our findings suggest that MVE-exposure induces T cell infiltration and ROS production, associated with increased demyelination in the cerebrum of female Apo E^−/− mice.

Highlights.

• Mixed vehicle exhaust (MVE) promotes demyelination in brains of female ApoE^−/− mice.

• MVE-exposure mediated demyelination is exacerbated in ovariectomized ApoE^−/− mice.

• Inhaled MVE is associated with increased CD4+/CD8+ cell infiltration into the CNS.

• MVE-exposure is associated with increased cerebral ROS production.

• Inhaled MVE mediates decreased cerebral estrogen receptor-α mRNA expression.

Abbreviations

Apo E^−/−

Apolipoprotein E null mouse

AD

Alzheimer’s Disease

BBB

blood-brain barrier

CNS

central nervous system

EAE

Experimental Autoimmune Encephalomyelitis

ERα

estrogen receptor alpha

ERβ

estrogen receptor beta

FA

filtered air

MOG

myelin oligodendrocyte glycoprotein

MHC

major histocompatibility complex

MS

Multiple Sclerosis

MVE

mixed vehicle exhaust

Ov+

female mice with ovaries

Ov−

ovariectomized female mice

PM

particulate matter

PRO A/B

progesterone receptors A/B

ROS

reactive oxygen species

TNF-α

tumor necrosis factor alpha

TLR

toll-like receptor