Glutamatergic neurons in rodent models respond to nanoscale particulate urban air pollutants in vivo and in vitro

Abstract

Background: Inhalation of airborne particulate matter (PM) derived from urban traffic is associated with pathology in the arteries, heart, and lung; effects on brain are also indicated but are less documented.

Objective: We evaluated rodent brain responses to urban nanoscale (< 200 nm) PM (nPM).

Methods: Ambient nPM collected near an urban freeway was transferred to aqueous suspension and reaerosolized for 10-week inhalation exposure of mice or directly applied to rat brain cell cultures.

Results: Free radicals were detected by electron paramagnetic resonance in the nPM 30 days after initial collection. Chronic inhalation of reaerosolized nPM altered selected neuronal and glial activities in mice. The neuronal glutamate receptor subunit (GluA1) was decreased in hippocampus, whereas glia were activated and inflammatory cytokines were induced [interleukin-1α (IL-1α), tumor necrosis factor-α (TNFα)] in cerebral cortex. Two in vitro models showed effects of nPM suspensions within 24-48 hr of exposure that involved glutamatergic functions. In hippocampal slice cultures, nPM increased the neurotoxicity of NMDA (N-methyl-d-aspartic acid), a glutamatergic agonist, which was in turn blocked by the NMDA antagonist AP5 [(2R)-amino-5-phosphonopentanoate]. In embryonic neuron cultures, nPM impaired neurite outgrowth, also blocked by AP5. Induction of IL-1α and TNFα in mixed glia cultures required higher nPM concentrations than did neuronal effects. Because conditioned media from nPM-exposed glia also impaired outgrowth of embryonic neurites, nPM can act indirectly, as well as directly, on neurons in vitro.

Conclusions: nPM can affect embryonic and adult neurons through glutamatergic mechanisms. The interactions of nPM with glutamatergic neuronal functions suggest that cerebral ischemia, which involves glutamatergic excitotoxicity, could be exacerbated by nPM.

Figure 1

Particle size distributions for ambient and reaerosolized nPM. Abbreviations: Co, cobalt; conc, concentration; Cr, chromium; Cu, copper, GSD, geometric SD; Mn, manganese; Ni, nickel; Sc, scandium; V, vanadium. (A) Average aerosol size distributions of reaerosolized nPM collected at the downtown Los Angeles site and used in this study. (B) Average ambient and concentrated reaerosolized nPM size distributions from the same downtown Los Angeles site. (C) Ambient (left) and reaerosolized (right) nPM bulk chemical composition. The mass ratios of water-soluble organic carbon (WSOC) and inorganic ions between the two aerosols are similar, except for the lower percentage of BC in reaerosolized nPM. (D) Redox-active metals in ambient and reaerosolized nPM. (E) The EPR signal of nPM from urban Los Angeles vehicular traffic collected 30 days before.

Figure 2

Chronic exposure of mice to nPM caused a decrease of GluA1 but no change in GluA2, synaptophysin, or PSD95 in hippocampus in vivo, shown by immunoblot analysis of whole hippocampal lysates of nPM- and control air-exposed mice. (A) GluA1 levels were reduced 36% in hippocampal lysates from nPM versus control (n = 7 hippocampi per group). No significant changes were seen in GluA2 (B) or presynaptic [synaptophysin (C)], or postsynaptic [PSD95 (D)] marker proteins. *p = 0.04 by t-test.

Figure 3

In vivo exposure of mice to chronic nPM induced inflammatory responses, shown by qPCR analysis of cerebral cortex mRNA of mice exposed to reaerosolized nPM at levels equivalent to peak vehicular traffic for 10 weeks, 3 days/week, 5 hr/day, as described in “Materials and Methods.” In cerebral cortex, nPM increased mRNA levels of innate immune receptor CD14 by 75% (A); the microglial marker CD68 by 100% (B); the astrocytic marker GFAP by 50% (C); and the two proinflammatory cytokine mRNAs IL-1α (D) and TNFα (E) (n = 7 cortices per group). *p < 0.05, compared with control air-exposed mice.

Figure 4

Aqueous suspensions of nPM caused in vitro glutamatergic neurotoxicity and inflammatory responses. (A) In cultured rat hippocampal slices, nPM caused glutamate-receptor–dependent neuronal damage (LDH release into culture media), which was blocked by the NMDA receptor antagonist AP5 (1 μg/mL nPM, 48 hr, + 50 μM AP5; n = 16 hippocampal slices per treatment). (B) nPM increased NMDA-induced neurotoxicity (10 μM NMDA, 24 hr treatment with 24 hr recovery). PI uptake was measured as a percentage of NMDA-treated slices (n = 5 slices per treatment). (C) In primary E18 neuronal cultures, nPM treatment reduced neurite outgrowth; left, representative image used to measure the length of neurites (bar = 20 μm); right, exposure to nPM (2 μg/mL, 48 hr) decreased the total length of neurites by 40% (n = 50 neurons per treatment). The inhibition of neurite extension was rescued by the glutamate receptor antagonist AP5 (50 μM). (D) In primary neuron cultures, nPM caused > 50% of growth cones to collapse (n = 90 neurons per treatment). Intact growth cones retain actin-rich (red) neuritic endings (filopodia and lamellipodia (representative images; bars = 20 μm). nPM (2 μg/mL, 48 hr) decreased neuronal viability by 20% (n = 10 culture wells per treatment). (E) In mixed glial cultures, 24-hr treatment with nPM caused dose-dependent increase in mRNA levels of inflammatory cytokines IL-1α and TNFα (n = 6 glia cultures per treatment). (F) Conditioned medium (CM) from nPM-treated mixed glia (10 μg/mL nPM, 24 hr) inhibited neurite outgrowth by 30% and decreased the number of neurites by 25% (n = 80 neurons per treatment). *p < 0.05, **p < 0.01; ^#p < 0.001, by ANOVA (A–C, E) or t-test (D,F), compared with control air-exposed mice