All procedures were reviewed and approved by the Institutional Animal Care and Use Committee at the University of Southern California. Male C57BL/6J mice were purchased from Jackson Laboratories and used at 8-10 weeks of age (24-29 g, at the beginning of the exposures). Animals were housed under USC Department of Animal Resources on a 12-hour light dark cycle with free access to food and water (except during the nPM/filtered air exposure periods). Mice were group housed with four mice per cage and single mice were randomized to one of four exposure groups: 1) filter 2) nPM 3) filter+BCAS 4) nPM+BCAS ( Figure 1A ).

nPM (particulate matter smaller than 0.2 μm in diameter) was collected in an urban area in central Los Angeles in direct proximity to a busy roadway, primarily impacted by traffic emissions (13, 14). Briefly, nPM was collected at 400L/min flow using a high-volume ultrafine particle sampler that removes particles larger than 0.2 μm by means of a multiple rectangular (slit) geometry jet conventional impactor. The resulting nPM was collected on pretreated Teflon filters (8x10”, PTFE, 2 μm pore) placed directly downstream of the impactor and transferred into an aqueous suspension by soaking in Milli-Q deionized water (total organic compounds <10 ppb, particle free, endotoxin <1units/mL, endotoxin-free glass vials) for 30min. Per US EPA recommended procedures, aqueous nPM suspensions were stored at -20°C (15).

The nPM suspensions were re-aerosolized by means of a HOPE jet nebulizer (Model 11310, B&B Medical Technologies, Carlsbad, CA, USA) as described in greater detail in (13). A total of 15L/min of aerosol flow was generated, with the majority (10L/min) drawn through the exposure chambers, and the remaining 5L/min diverted to Teflon and quartz filters for particle collection and characterization. The mass concentration of nPM was determined by pre- and post-weighing the Teflon filter under controlled temperature and relative humidity. Inorganic ions [ammonium (NH₄ ⁺), nitrate (NO₃ ^-), and sulfate (SO₄ ^2-)] were analyzed by ion chromatography. PM-bound metals and trace elements were analyzed by magnetic-sector inductively coupled plasma mass spectroscopy. Water-soluble organic carbon on the quartz filter was analyzed by GE-428 Sievers liquid analyzer (GE-Sievers, Boulder, CO).

Mice were exposed to either re-aerosolized nPM or filtered air for five hours per day, three days per week for 10 weeks (150 total hours) ( Figure 1A ). Exposures were conducted in temperature and air controlled sealed whole-body chambers with adequate ventilation. Different cohorts of mice were used for MRI (2 cohorts) and Immunohistochemistry (2 cohorts). Particle characteristics from all cohorts were averaged and are presented in Figure 1 . Table 1 demonstrates the particle characteristics for every nPM batch used in this study. Table 2 lists the mass fractions of organic carbon, trace elements, and metals used in each nPM batch. Histochemistry studies were performed on stored tissue samples from mice exposed to nPM in 2015-2016 using nanoparticles collected in 2015 and 2016. MRI studies were performed on mice exposed to nPM in 2018 using nanoparticles collected in 2018 (batch 6,7, Table 1 ) (18).

BCAS procedures (11, 20), were performed 30 days prior to the conclusion of nPM/filtered air exposure ( Figure 1A ). Rectal temperature was maintained at 36.5-37°C. Mice were anesthetized by intraperitoneal ketamine/xylazine and placed in the prone position. Cerebral blood flow (CBF) values were monitored with a microtip fiber probe (Probe 418-1 master probe PF 5010 Laser Doppler Perfusion Monitoring Unit, Perimed AB, Sweden) fixed to the skull 1mm posterior and 5mm left of bregma. The mouse was then rotated to the supine position and bilateral common carotid arteries were exposed through a midline cervical incision. A 0.18 mm diameter microcoil (Sawane company, Japan) was applied to one carotid artery. CBF was recorded after the first coil application. A second microcoil was applied to the other carotid artery and CBF was recorded again. The incision was closed, and the mouse recovered in a temperature-controlled environment. For details of the surgeries see Liu et al. (16). Briefly, CBF drop right after insertion of microcoils was not different between filter+BCAS (35%) and nPM+BCAS (33%) groups. Total mortality during the study was 29.2% (42 out of 144 mice) and occurred only in the BCAS groups. No mice died in the filter and nPM groups. Mortality after BCAS surgery was not different between filter+BCAS (50%) and nPM+BCAS groups (52.5%, p=0.83).

MRI scans were performed using a 7T PET-MR system (MR solutions Ltd, Guildford, UK) with a standard 20-mm internal diameter quadrature bird cage mouse head coil with a setup described in our earlier publications (21, 22). For in vivo scans the animals were anesthetized with 1.5-2.0% Isoflurane/air, cannulated for tail vein injections and monitored for respiration with an abdominal pressure-sensitive probe (80-100 BPM) and temperature with a rectal probe (36.5 ± 0.5°C) to ensure reproducible depth of anesthesia (Minerve mouse imaging cell, Esternay, France). The following sequences were used in this study: T2-w anatomical images (2D-fast spin echo (FSE), TR/TE 4,600/28 ms, 40 slices, slice thickness 500 μm, in-plane resolution 100x50 μm²), diffusion weighted echo-planar imaging (EPI, TR/TE 5,000/32, 14 slices, 1 b-value, 7 directions, slice thickness 300 μm, in-plane resolution 200×300 μm²), T1 mapping (1 slice fast low angle shot (FLASH) with variable flip angle, FA 5-45°, TR/TE = 20/4 ms, slice thickness 1 mm, in-plane resolution 60x120 μm²), T1-w dynamic contrast enhanced (DCE) MRI for BBB permeability (1 slice FLASH with identical geometry to T1 mapping, 180 images with temporal resolution of 5.1 s, FA 15°, TR/TE = 20/4ms) and T2*-w dynamic susceptibility contrast (DSC) MRI for CBF (1 slice FLASH, 80 images with temporal resolution of 1.4 s, FA 15°, TR/TE = 18/3ms, slice thickness 1 mm, in-plane resolution 120x230 μm²). Gadolinium-DTPA (Gd-DTPA) (Magnevist^®, Bayer) diluted in saline was administered with a power injector during DCE (140μl, 1:4 dilution, 600μl/min) and DSC (140μl, 1:1 dilution, 1000μl/min) scans.

Analysis of DCE-MRI and DSC-MRI data was performed using Rocketship software (https://github.com/petmri/ROCKETSHIP) (23). Analysis of angiography images was done using ImageJ and otsu-thresholded maximum intensity projection (MIP) images and presented as relative blood flow referring to the signal intensity of vessels of interest in MIP images. Diffusion tensor imaging (DTI)-EPI parametric maps were generated using DSI Studio (March 6, 2018 build; dsi-studio.labsolver.org) as previously described (24). The diffusion weighted image data was pre-processed with motion, ghosting and eddy current corrections. After the co-registration of all individual images, data means were extracted for fractional anisotropy (FA) and apparent diffusion coefficient (ADC; x10⁻³ mm²/s) through DSI studio. Regions of interest (ROIs) for corpus callosum were manually assigned by an investigator blinded to groups. To examine distinct changes, regional ROIs were drawn on ADC maps. Due to time-constraints, six animals per group underwent MRI. One animal from filter+BCAS group was excluded from the study due to gray matter changes in T2-w images that indicated development of ischemic stroke. For DSC-MRI, one animal in the nPM+BCAS group was excluded due to motion artifact.

Following the exposure period, mice were humanely euthanized with an intraperitoneal injection of ketamine and xylazine and transcardially perfused with PBS with 5 U/ml heparin followed by 4% paraformaldehyde in 0.01 mol/L PBS buffer. Brains were fixed in 4% paraformaldehyde for 24 hours at 4°C and stored in 70% ethanol. The tissue was paraffin embedded and 5 µm coronal sections were sliced from a part of the brain located from 1 mm anterior to 2 mm posterior to the bregma. Slides were deparaffined and hydrated using a series of alcohol dilutions (from 100% to 70% ethanol).

For myelin basic protein (MBP) and SMI-312 staining, antigen was retrieved with 10 mM sodium citrate and sections were blocked in 5% donkey serum in PBS with 0.3% Triton X-100 for one hour. Slides were incubated overnight at 4°C with primary antibody. Primary antibodies used were mouse anti-SMI-312 (BioLegend, 837904, 1:500) and rabbit anti-MBP (Santa Cruz Biotechnology, sc-13914-R, 1:200). Sections were then washed with PBS and incubated with secondary antibody for one hour. Secondary antibodies used were Alexa Fluor 568-donkey anti-mouse (Invitrogen; A-10037; 1:200) for SMI-312 and Alexa Fluor 568-donkey anti-rabbit (Invitrogen A; 10042; 1:200) for MBP. For IgG and Lectin double immunostaining, slides were washed in PBS and incubated with both Alexa Fluor 568-donkey anti-mouse IgG (Invitrogen; A-10037; 1:200) and Dylight 488-conjugated L. esculentum lectin (Vector Labs DL-1174; 1:200) for one hour. For all immunofluorescence, nuclei were stained with Hoechst 33342 (Life Technologies; 1:5000) for 10 minutes and sections were mounted with Agilent fluorescent mounting media. Slides were coverslipped and imaged at 400x using BZ-X810 fluorescent microscopy (Keyence, NJ). One slide per animal was assessed in each of the immunofluorescent stains (MBP, SMI312, and IgG and Lectin co-stain).

Sample sizes were n=12 per group for IgG and Lectin, n=12 for filter, nPM, and nPM+BCAS groups and n=11 for filter+BCAS group for SMI-312, and n=12 for filter, nPM, and nPM+BCAS groups and n=9 for filter+BCAS for MBP. All slides that underwent staining were analyzed and none were excluded.

Two blinded, independent observers analyzed all the sections, and their results were averaged. MBP and SMI-312 integrated fluorescence density was analyzed in both the right and left medial corpus callosum of one high-powered field (400x) and the two values were averaged. For quantification of extravascular IgG, IgG integrated density was analyzed outside of lectin-positive vascular profiles using Image J software (National Institutes of Health (NIH)) as we previously described (20). Briefly, IgG signal that colocalized with lectin-positive signal was subtracted from the total IgG immunofluorescence to obtain extravascular IgG integrated density. Extravascular IgG integrated fluorescence density was assessed in the right and left medial corpus callosum of one image (400x) and the results were subsequently averaged. All image analysis was done using Image J software (NIH) and followed the NIH Image J user guide.

Statistical analyses were carried out using GraphPad Prism (GraphPad Software LLC, version 9.0.0). The data is presented as mean ± standard error of the mean, mean ± standard deviation or violin plot with median and interquartile range as indicated in figure legends. Statistical testing was performed using 1-way or 2-way ANOVA followed by Bonferroni test. Datasets were screened for outliers using Grubb’s test and alpha = 0.05, but no outliers were detected. P-value below 0.05 was considered statistically significant.

To study the effects of combined cerebral hypoperfusion and nPM exposure, mice were exposed to nPM aerosol or filtered air for 6 weeks. Mice then underwent either bilateral carotid artery stenosis surgery or remained as controls ( Figure 1A ). After the surgeries, the nPM or filtered air exposure was continued for another 4 weeks (total exposure: ten weeks) before MRI and tissue collection ( Figure 1A ). The physiochemical properties of exposure aerosol are in Figures 1B – G . Measured average particulate matter mass concentration of the exposure aerosol was 310 µg/m³ ( Figure 1C ), total particulate number concentration 3.6 x 10⁵ particles/cm³ ( Figure 1D ), mode diameter 50 nm ( Figure 1E ), and total organic mass 43% total PM mass ( Figure 1F ). The elemental composition of the aerosol was further analyzed by inductively coupled mass spectrometry (ICP-MS), ( Figure 1G ). Overall, the exposure aerosol characteristics were typical for a sample collected from an urban area, where traffic-related emissions account for a major part of the air pollution profile, with organic matter (TOM) being the most predominant PM constituent and with an appreciable content of redox active metals and trace elements such as Fe, Mn, Ni, Cu and Zn (across nPM batches, Table 2 ) (25). The PM mass concentration, total particle number concentration, and the mode diameter were similar across nPM batches ( Table 1 ).

After the completed exposure of 10 weeks, we first measured regional CBF values in the mice using DSC-MRI ( Figures 2A, B ). As expected, both filter+BCAS and nPM+BCAS groups had lower (by 21-42%) cerebral blood flow values compared to the filtered air group ( Figure 2B ). White matter hypoperfusion was further reduced by 26% in the nPM+BCAS group compared to filter+BCAS group. No effect was seen with nPM exposure alone in controls. The 3D time-of-flight (TOF) MR angiography images were analyzed for large vessel blood flow in the neck and brain ( Figures 2C, D ). Exposure to nPM had no effect on large caliber vessel blood flow either in control (filter vs. nPM) or BCAS operated (filter+BCAS vs. nPM+BCAS) mice, illustrated for the middle cerebral artery (MCA) in Figure 2D . However, BCAS effect reduced the blood flow signal in the MCA by 17-18% (p<0.0025). Taken together, these results indicate that nPM exposure worsens white matter microvascular blood flow in hypoperfused brain.

We have previously shown that BCAS leads to BBB breakdown in mouse white matter (20), thus we used DCE-MRI to measure the BBB permeability in the corpus callosum of the mice in vivo ( Figures 3A, B ). Based on the quantification of K_trans values, the combination of nPM exposure and BCAS caused a major BBB breakdown (69% permeability increase for a <1 kDa Gd-contrast agent) in the white matter of mice ( Figure 3B ). nPM exposure alone did not affect white matter BBB permeability values measured by DCE-MRI. Postmortem studies confirmed these results by immunohistochemistry for blood-derived IgG (~150 kDa, Figures 3C, D ). IgG signal was negligible in brains of control mice with filtered air or nPM exposure ( Figure 3D ). Remarkably, quantification of IgG signal outside lectin-positive blood vessels revealed heavy deposition of IgG in the white matter of nPM exposed BCAS mice (140% increase compared to filtered air) confirming the DCE-MRI results. In addition, the IgG deposition was 44% higher in the nPM+BCAS group compared to the filter+BCAS group. These results demonstrate that nPM exposure aggravates BBB breakdown induced by hypoperfusion in the white matter of mice.

Chronic cerebral hypoperfusion induced by BCAS causes white matter damage in mice (11). We sought to examine whether diffusion-weighted MRI would reveal changes in the white matter diffusion metrics, which we showed in other models of vascular white matter injury (21). First, we measured the fractional anisotropy (FA) values in the corpus callosum ( Figures 4A, B ). We noted that both BCAS surgery alone and BCAS surgery combined with nPM exposure significantly reduced FA values in the corpus callosum (by 14 and 21%) compared to the filtered air group ( Figure 4B ). nPM exposure without BCAS did not affect FA values. Moreover, apparent diffusion coefficient values (ADC) were higher (by 10%) in the nPM+BCAS group compared to all other groups (p< 0.05) ( Figure 4C, D ). We did not detect changes in white matter ADC values in mice that underwent nPM exposure or BCAS surgery alone. Altogether, BCAS led to abnormal diffusion metrics, suggestive of white matter structural damage, with abnormalities further worsened by nPM exposure.

To study whether altered diffusion metrics in the white matter were linked to axonal damage or demyelination, we stained brain sections collected from all groups for SMI-312 (axonal marker) and myelin basic protein (MBP) ( Figures 5A – D ). As expected (26), axonal density in the corpus callosum was reduced in both BCAS groups (by 27 and 30%) compared to filtered air group, ( Figures 5A, B ). However, no nPM effect was seen either in control (filter vs. nPM) or BCAS groups (filter+BCAS vs. nPM+BCAS; Figure 5B ). Finally, myelin density in the corpus callosum was decreased in both nPM (18%) and nPM+BCAS (48%) groups compared to the filtered air group ( Figure 5D ). Demyelination was also observed in the filter+BCAS group (32% compared to filtered air group), but nPM exposure decreased the myelin density by an additional 24% when combined with BCAS (nPM+BCAS vs. filter+BCAS; Figure 5D ). These results suggest that BCAS reduces both axonal density and myelination in the white matter in mice, whereas nPM exposure primarily leads to demyelination.