Air pollution & the brain: Subchronic diesel exhaust exposure causes neuroinflammation and elevates early markers of neurodegenerative disease

Accumulating evidence points to neuroinflammation as an active participant in the progression of neurodegenerative diseases, such as Parkinson's disease (PD) and Alzheimer's disease (AD) [1‐3]. In fact, current theory holds that pro-inflammatory events in the brain very likely occur across an individual's lifespan to culminate in neuropathology [3, 4]. While environmental factors are largely implicated in the etiology of neurodegenerative disease [5, 6], at present the various sources responsible for the chronic neuroinflammation leading to central nervous system (CNS) pathology are poorly understood.

Air pollution is a mixture comprised of several components, including particulate matter (PM, the particle components of air pollution), gases, and metals, such as vanadium, nickel, and manganese [7, 8]. This toxin is readily available in the environment in many forms from multiple sources [8, 9] and exposure occurs across and individual's entire lifetime. In fact, in the US alone, millions of people are exposed to levels of air pollution above established safety standards [8, 10]. This is of significant concern, as diverse forms of air pollution have been widely implicated in inflammation and oxidative stress in humans [11].

While the majority of studies focus on the effects of air pollution in cardiovascular and pulmonary disease [12], accumulating evidence now points to a new role for air pollution in CNS disease [10]. For example, human studies have shown that living in conditions with elevated air pollution is associated with decreased cognitive function [13], AD-PD like neuropathology [14], and increased stroke incidence [15]. Even the individual air pollution components such as manganese have been linked to CNS pathology, as elevated levels of manganese in the air are linked to enhanced PD risk [16]. Consistent with human reports, recent animal studies reveal that exposure to diverse forms of air pollution by inhalation, such as urban PM [17, 18], ozone [19], DE, and manganese [20, 21] results in a common pro-inflammatory response and oxidative stress in the brain. However, given the significant expense of inhalation exposure studies, the majority of this experimental work is based on short term (one month - 10 weeks) studies, with only high exposure levels tested. While these studies are critical for understanding how air pollution affects the brain, human exposures to air pollution typically occur at lower concentrations. More specifically, PM levels in polluted US cities peak around 50 μg PM/m³ [8], near-road PM concentrations are measured around approximately 100 μg PM/m³, and occupational exposure to PM occurs around 1000 - 2000 μg PM/m³ [22, 23], where human exposure continues for years.

Diesel exhaust (DE) is a form of air pollution that has received significant attention regarding its potential effect on human health in both ambient and occupational exposure conditions [24], and several studies have documented the CNS effects of DE. For example, acute, high level DE exposure affects electroencephalogram parameters in adult human subjects [25]. Animal research has shown that the prenatal period is a critical period of vulnerability, where maternal DE exposure affects dopamine neurochemistry and causes motor deficits in offspring [26, 27]. Short term studies in young adult animals also demonstrate that DE elevates pro-inflammatory factors in the brain, using a month-long inhalation models [18, 28], intratracheal administration directly into the lung [18], and a 2 hr-long exposure by nose-only inhalation [29]. However, while air pollution exposure is known to occur across an individual's lifetime, at this time, little is known about the consequences of chronic DE exposure in the CNS.

In the current study, we begin to define the deleterious CNS effects in response to subchronic (6 month) DE exposure. More specifically, we address the minimum levels of DE necessary for neuroinflammation, and explore when these exposures are associated with early markers of pre-clinical CNS disease.

Accumulating evidence indicates that the brain detects and responds to diverse classifications of inhaled air pollution, such as metals, ozone, urban PM, and DE with a common pathway of neuroinflammation [10]. However, it is unclear whether the pro-inflammatory response in the brain is merely a marker of exposure to air pollution or whether this response is linked to more sinister consequences. Here, we begin to explore these questions using subchronic DE exposure in an effort to model the persistent nature of air pollution exposure and employ the use of lower levels that are comparable to busy road-way and occupational levels. Together, this approach allowed us to begin to address what conditions are necessary for air pollution to elicit CNS effects and assess whether markers of neurodegenerative disease pathology occur with neuroinflammation.

TNFα is a "potent" pro-inflammatory cytokine elevated in both AD and PD patients, where it is implicated to play a causal role in neurotoxicity [45]. Consistent with previous reports on short term and high exposures to air pollution [18, 28, 46] and chronic human studies [14], here we show a general pro-inflammatory response in the brain with subchronic DE exposure, which we propose may be due in large part to a systemic/peripheral effect that reaches the entire brain, rather than solely through the olfactory bulb, a favored pathway of PM entry into the brain [47, 48]. This is evidenced by the fact that the olfactory bulb showed a blunted TNFα response when compared to other regions and TNFα levels were elevated in most regions tested, with the exception of the cerebellum (Figure 1). The cerebellum contains fewer numbers of the brain's resident innate immune cell, microglia [49], and it is not traditionally involved in AD or PD pathology. Thus, consistent with prior reports [18], our current data also support that microglia may regulate the brain region-specific pro-inflammatory response to DE.

More specifically, our previous work with short term (1 month) inhalation of higher levels of DE indicated that the midbrain, which contains the substantia nigra damaged in PD, is more vulnerable to the pro-inflammatory effects of DE [18]. In particular, the midbrain produced the most robust elevation of multiple cytokines, chemokines, and nitrated protein levels when compared to other brain regions [18]. Consistent with this premise, analysis of microglial markers confirmed that the midbrain expressed highest levels of microglial markers at rest in control animals and showed the greatest elevation or microglial markers in response to short term and high DE exposure [18]. Interestingly, in response to subchronic DE in the current study, the midbrain expressed TNFα levels comparable to the other brain regions tested (Figure 1A), suggesting that perhaps the pro-inflammatory response may be tempered with longer exposures. However, the midbrain was the only region to show significantly elevated TNFα levels in response to lower levels of DE (100 μg PM/m³ ) with 6 month exposure (Figure 1A), demonstrating that the midbrain sensitivity to air pollution extends to longer and lower DE exposures.

We next sought to discern whether this enhanced sensitivity to DE in the midbrain generalized to other pro-inflammatory markers. IL-1β is another pro-inflammatory factor elevated in PD and AD that has been widely implicated in neuronal damage [50]. Here, we show that IL-1β levels are elevated in response to subchronic DE, but only at the highest concentration of 992 PM μg/m³ (Figure 2A, p < 0.05). IL-6 is both a beneficial and potentially detrimental cytokine that responds to neuronal damage and is elevated in AD and PD [51]. However, we found no significant effect of IL-6 in the midbrain in response to DE (Figure 2B, p > 0.05). MIP-1α is a chemokine important for microglial migration [52] and our current study demonstrates that subchronic DE exposure causes a reduction in MIP-1α in the midbrain at the highest concentrations tested. This decline in MIP-1α is consistent with a pattern seen in the lung of these same animals, as previously reported [30]. Thus, the enhanced sensitivity seen with TNFα in the midbrain at lower concentrations of DE is not conserved across all pro-inflammatory factors tested, which is different than what we had previously reported with one month DE exposure [18]. This suggests that perhaps compensatory mechanisms are triggered with longer exposures. Together, the data support that TNFα may be an important cytokine for the CNS effects of air pollution.

Several human studies have shown that chronic exposure to high levels of air pollution is linked to AD-like pathology, including elevation of diffuse plaques, neuroinflammation, and frontal lobe damage [13, 14, 42]. Given that neuroinflammation, particularly elevation of TNFα, has been linked to the induction of hyperphosphorylation of Tau [53], we sought to determine whether DE had an effect on this parameter in a subchronic inhalation rat model. Tau is a major component of neurofibrillary tangles found in AD and FTD patient brains where it is hyperphosphorylated at several sites, including the Ser 199 residue (Tau [pS199]) [37]. Further, hyperphosphorylation of Tau S199 has been implicated as an early marker of Tau pathology [38] We show here, that only the highest level of DE caused elevation of Tau [pS199] in the frontal lobe (Figure A, p < 0.05) and temporal lobe (Figure 3B, p < 0.05). In addition, we also show that only the highest level of DE caused elevation of Aβ42 (Figure 5, p < 0.05). These findings support that high levels of DE may be linked to neuropathology associated with pre-clinical AD and FTD markers.

Previous studies in humans from highly polluted areas show an elevation of brain α synuclein [13, 42]. However, our earlier reports employing only month-long DE exposure show robust neuroinflammation with no significant effect on α synuclein levels or evidence of neurotoxicity in the midbrain [18]. Here, we explored whether DE exposure elevated α synuclein in response to longer, subchronic DE exposure. α Synuclein is known to be elevated in the midbrain of sporadic PD patients [40], where elevation occurs early in the disease and its use has been implicated as a pre-clinical marker of PD [41]. In the current study, we show that DE increased α synuclein levels at only highest concentrations (Figure 4. p < 0.05).

Together, these results show that 6 month exposure to DE elevated TNFα in most brain regions tested, with the exception of the cerebellum. In particular, the midbrain region, which houses the substantia nigra that is selectively lost in PD, was the most sensitive to DE effects, as TNFα was elevated in response to low levels of DE (100 μg PM/m³). There was also evidence of compensatory mechanisms in the midbrain with subchronic DE exposure, as IL-6 was not significantly altered, IL-1β was only elevated at the highest concentration, and MIP-1α decreased at higher concentrations in the midbrain. Tau [pS199], a protein modification linked to both AD and FTD, was elevated at only the highest concentrations of DE in both the temporal and frontal lobes. Aβ42, a protein implicated in both AD and FTD pathology, was also increased in the frontal lobe in response to DE only at the highest concentration. Interestingly, α synuclein was elevated in the midbrain at only the highest concentration, suggesting that the TNFα increase at lower concentrations is not yet sufficient to initiate this potential marker of preclinical PD. These findings indicate that while some compensatory mechanisms may occur, the neuroinflammatory response to air pollution, particularly the TNFα response, is still present with subchronic exposure and may precede evidence of neuropathology. Future research needs to address the effects of lifetime air pollution exposure and the impact of aging on neuroinflammation and neurotoxicity.