University of Southern California Dissertations and Theses

Neuroinflammation and the behavioral consequences of air pollution over the life course

(USC Thesis Other)

Neuroinflammation and the behavioral consequences of air pollution over the life course

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content i

Neuroinflammation and the Behavioral Consequences of Air
Pollution Over the Life Course

by

Nick Woodward

A Dissertation Presented to the
FACULTY OF THE
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Gerontology)

December 2016

Copyright 2016 Nick Woodward

ii

Table of contents Page No.

Dedication ................................................................................................................................ iv
Acknowledgments .................................................................................................................... v
Abstract ................................................................................................................................... vii
Background ............................................................................................................................ viii
Chapter 1: Traffic-related air pollution accelerates myelin and neuritic aging changes in a
mouse model, with a specificity for CA1 neurons ................................................................47
Abstract ................................................................................................................................47
1. Introduction .....................................................................................................................48
2. Methods ...........................................................................................................................50
3. Results .............................................................................................................................53
4. Discussion .......................................................................................................................56
5. Conclusion .......................................................................................................................60
Tables and Figures ..............................................................................................................66
Chapter 2: TLR4 plays an integral role in glial activation by nanoscale particulate matter:
insights from microarray analysis .........................................................................................80
Abstract ................................................................................................................................80
1. Introduction .....................................................................................................................81
2. Materials and Methods: ...................................................................................................83
3. Results .............................................................................................................................88
4. Discussion .......................................................................................................................95
5. Conclusion ..................................................................................................................... 100
Tables and Figures ............................................................................................................ 106
Chapter 3: Developmental exposure to particulate air pollutants reduces cell proliferation
in the hippocampus and impairs hippocampal dependent memory ................................. 123
Abstract .............................................................................................................................. 123
1. Introduction ................................................................................................................... 124
2. Methods ......................................................................................................................... 126
3. Results ........................................................................................................................... 131
4. Discussion ..................................................................................................................... 133
5. Conclusion ..................................................................................................................... 136
Tables and Figures ............................................................................................................ 143

iii

Chapter 4: Minimizing air pollution exposure: a practical policy to protect vulnerable
older adults from death and disability ................................................................................. 149
Abstract .............................................................................................................................. 149
1. Introduction ................................................................................................................ 150
2. Materials and Methods ............................................................................................... 152
3. Results ........................................................................................................................ 155
4. Discussion .................................................................................................................. 157
5. Conclusion .................................................................................................................. 163
Tables and Figures ............................................................................................................ 167
Concluding Remarks ............................................................................................................ 172

iv

Dedication

To my family and, most especially, my mother,
for their unwavering support and unbridled enthusiasm in all aspects of life.

v

Acknowledgments

I extend the utmost thanks and gratitude to Dr. Caleb E. Finch. He leads by example in
his tireless enthusiasm for the sciences and in all intellectual inquiries. He has taught
me the value of history, and in doing so has shown me the richness that lies within
everything around us. He has continuously pushed me to look deeper, and to never be
satisfied with what you know. I would like to extend special thanks to Dr. Todd E.
Morgan, for his guidance and support in my development as a scientist. His efforts buoy
all those around him, and his presence defines the lab. I would also like to thank Dr.
Sean Curran, who has been approachable and enthusiastic, and has offered me
guidance in goals both short and far. Finally, I would like to thank Dr. Scott Kanoski,
who has opened his doors to allow us to pursue great questions together. He has taught
me to laugh in lab, and has shown me a truly incredible research group.
I would also like to thank my lab mentors David Davis and Mafalda Cacciottolo,
who put up with endless questions and taught me the nuances of benchtop science. I
extend my deepest and heartfelt thanks to Dr. Hank Cheng; my companion for many
late nights in lab, and who probed deeper, if not always related to science. I thank my
labmates Eliza Bacon, Amin Haghani, and Nikoo Haghani, and those who have worked
with me over the year, allowing me to teach them and them to teach me: Swathi Prasad,
Sam Epstein, and Richard Johnson. Thank you as well to my collaborators Morgan
Levine, and Arian Saffari, whose efforts and expertise allowed for great research.

vi

Thank you to Eileen Crimmins, whose passion for science continues to inspire
me. Her guidance and support has meant a great deal to me. Thank you also to Kate
Wilbur, for her collaborations and care; who allowed me a brief foray into being a policy
wonk. Finally, I would like to thank the Gerontology Department. It truly is a school
within the school at large, and has become a community and family to me. The efforts of
the administration, and in Maria Henke in particular, have made this school truly special.

vii

Abstract

What do we hope to achieve by researching air pollution? Similar to a clinical trial for
new drug development, any new technology or chemical used in society must be
rigorously evaluated for environmental and individual harm before being adopted.
Unfortunately, this was not the case for air pollution, specifically pollution produced by
combustion engines (traffic-related air pollution, TRAP). It has gained worldwide
dominance as a mobile energy source, while to this day not being fully understood.
Cardiovascular effects were publicly recognized as late as the mid twentieth century,
while metabolic and cognitive effects are only recently being fully recognized. The
purpose of air pollution work is two-fold. First, to fully elucidate the effects of air pollution
exposure, in order to understand the personal and societal cost of air pollution. Second,
this information must be used to mitigate harm as much as possible, whether through
galvanizing policy interventions or shifting public demand. The work presented here
addresses these aims by furthering the understanding of cognitive and metabolic effects
of pollution exposure, elucidating potential mechanisms behind these effects, and
proposing population level solutions to attenuate air pollution derived morbidity and
mortality. Chapter one addresses the knowledge gap of aging interactions with TRAP
exposure, and explores hippocampal neuronal and neuroinflammatory effects of TRAP
exposure. Chapter two begins to elucidate the pathways activated by pollution exposure
that lead to neuroinflammation. Chapter three explores the developmental role of
pollution exposure, focusing on hippocampal dependent memory impairment and
neuroinflammation. Finally, chapter four utilizes current knowledge of spatial dispersion
patterns of airborne particulate to propose policy interventions to reduce the mortality

viii

risk from pollution exposure in vulnerable older adult populations. It is the hope of this
researcher that the knowledge presented here will be valuable in furthering our
understanding of air pollution, and in reducing the current societal burden imposed by
pollution exposure.

Background

1. Introduction:
1.1.1 Epidemiological relevance of air pollution
Air pollution first gained widespread recognition in 1952 as a result of London’s
“Great Smoke.” Over the course of four days, significantly decreased air circulation,
combined with large amounts of pollution generated by coal burning plants caused an
estimated 100,000 illnesses, as well as 12,000 premature deaths (Logan et al., 1953).
This catastrophic event became a catalyst for the United States government to pass the
Clean Air Act of 1956 and the impetus for environmental movement as we know it
today.
Since the time of the Great Smoke, studies have consistently shown that
increased rates of acute myocardial infarction and cardiovascular related mortality are
strongly associated with exposure to air pollution (Pope et al., 2002, Pope et al., 2004,
Samet et al., 2000). Analysis by Pope et al. 2002 showed that the relative risk for
mortality due to living in a heavily polluted area is roughly equivalent to the relative risk
of being overweight (a BMI between 25-39.9 kg/m
2
). Although air pollution negatively

ix

affects everyone, the worst outcomes occur from exposure during development, and at
older ages.

1.1.2 Developmental exposure- Low birth weight
Because developmental cognitive defects are often associated with fetal growth
retardation, it is important that TRAP can impact fetal growth. Associations between air
pollution exposure in gestation and impaired fetal growth continue to emerge. In
particular, PM2.5 is associated with low birth weight (LBW) (< 2,500 g), preterm birth,
and small gestational size (Shah et al., 2011). However, the critical period for exposure
during pregnancy and threshold for these effects remain undefined. Discussed here are
select large-scale studies of PM exposure during gestation. Also see the
comprehensive review of Shah and Balkhair.
A large Los Angeles based study (n = 220,528) showed 5% greater risk of LBW
from PM2.5 exposure, with a range of 2.4 ug/m
3
(Wilhelm et al., 2012). Other studies
used ultrasound to determine the gestational timing of LBW association with air
pollution. The largest of these studies (17,660 pregnancies) showed the most consistent
association PM10 exposure during days 91–120 of pregnancy, where high PM10
correlated with smaller abdominal circumference, heard circumference, and femur
length (Hansen et al., 2008). Though this study did not find association with nitric oxide
(NO2) exposure, other studies associated exposures of NO2 > 38 ug/m
3
with reduced
fetal size, femur length, and biparietal distance, even when high NO2 was recorded only
for weeks 12–20 (Iniguez et al., 2012; van den Hooven et al., 2012). Other studies
associated elevated PM10 exposure with preterm birth (van den Hooven et al., 2012;

x

Ritz et al., 2000; Ritz et al., 2002). A potential mechanism underlying LBW is oxidative
stress from maternal exposure during pregnancy to TRP, e.g. increased placental DNA
adducts (Kannan et al, 2006).
Obesity is also showing association with air pollution components, which may
contribute to diabetes and the metabolic syndrome (Sun et al., 2009). Adults (n = 5,228)
exposed to NO2 showed 17% higher risk of diabetes mellitus in the top vs. lowest quintile,
differing by 4 ppb (Brook et al., 2008). There are also correlations between PM10 exposure
and the white blood cell count, a marker for systemic inflammation (Chen et al., 2008).

1.1.3 Developmental exposure- Cognition
Epidemiological studies of TRAP exposure during development show negative
associations with adult cognition (Pedersen et al., 2004; Marcelis et al., 1999; Volk et
al., 2013) and brain development (Pedersen et al., 2004; Marcelis et al., 1999; Volk et
al., 2013; Perera et al., 2012). In particular, pre- and postnatal exposure to urban TRP is
correlated with autism spectrum disorders (ASD), schizophrenia, and impaired cognitive
development. We briefly summarize these findings.
ASD was associated with local gradients in components of TRP, mainly PM2.5. Two
studies utilizing the California based CHARGE (Childhood Autism Risks from Genetics
and the Environment) database found about 2-fold higher odds ratio (OR), for
development of ASD, when living near a freeway during the 3rd trimester, and at
delivery (< 309 m defined as near, with > 1,419 m as reference group)

(Volk et al., 2013;
Volk et al., 2011). Exposure during the first postnatal year was associated with 3-fold
higher OR for ASD (Volk et al., 2013). PM2.5 had an OR of 2.08 for gestational

xi

exposure, and 2.12 for exposure in the first year of life (Volk et al., 2013; Volk et al.,
2011). Similarly, the Nurses’ Health Study, a national sample, showed an OR of 2.0 for
prenatal diesel particulate exposure, top vs. bottom quintile (PM2.5 4.40 vs. 0.60 ug/m
3
)
(Roberts et al., 2013). Contrarily, a study of Swedish twins did not find association of
TRP with ASD (PM10 3.3–4.2 ug/m
3
); however, this study measured a broader size
range of particles (Gong et al., 2014). An analysis of 35 pollution components showed
higher OR for ASD after exposure to methylene chloride, quinolone, and styrene, but
not after diesel PM, or polycyclic aromatic hydrocarbons (PAHs) (Kalkbrenner et al.,
2010). The authors noted that the control group had impairments of speech and
language, which may have biased the results towards null findings.
Schizophrenia risk is also sensitive to TRAP in top vs. bottom quintiles of urbanicity
(population density) during gestation, but not during childhood (Marcelis et al., 1999). A
study of traffic volume and urbanicity (household crowding, social stressors) concluded
that only traffic volume exposure at birth predicted schizophrenia (OR of 4.40 for the top
vs. lowest quintile of traffic exposure) (Pedersen et al., 2004). Both studies agree that
only exposure during the gestational period correlated with increased risk.
TRP exposure during development is also associated with subclinical cognitive
effects, including lower mental development, increased anxiety and depressive
behavior, and attentional problems (Perera et al., 2012; Guxens et al., 2012; Perera et
al., 2006). A Spanish national study showed decreased mental development for infants
of mothers exposed during pregnancy to elevated NO2 and benzene (Guxens et al.,
2012). Importantly for potential interventions, this association was attenuated in mothers
who self-reported a high intake of antioxidant rich foods. We also note the benchmark

xii

study of Perera et al. 2003 on PAH levels for Hispanics and African Americans in New
York City, which was the first to utilize personal monitors for PAH levels, with greater
precision than citywide measurements. Developmental measurements at birth
associated high PAH (> average 3.7ng/m
3
in maternal blood) with a 9% decrease in
birth weight, and a 2% decrease in head circumference. The OR for cognitive
developmental delay, at 36 months from PAH exposure during gestation was 2.89, for
the top vs. bottom quintile (Perera et al., 2006). By age 6–7 years, individuals in the top
exposure quintile were more anxious and depressed (OR 1.45), with more attentional
problems (OR 1.28, top vs. bottom quintile) (Perera et al., 2012). For DSM-IV oriented
anxiety problems, the OR was a striking 4.59 (Perera et al., 2012).

1.1.4 Older adult vulnerability to pollution exposure
Populations exposed to air pollution have well documented increases in the incidence of
cardiovascular disease (Chen et al., 2013). The formation of atherosclerotic plaques,
which leads to cardiovascular disease, has been shown to be exacerbated by ambient
air pollution exposure (Calderon-Garciduenas et al., 2002). Because air pollution has
the greatest effects on those with poor physiological functioning and preexisting
conditions, older adults face an increased risk for adverse health outcomes from
pollution exposure (Goldberg et al., 2001). In addition to exacerbation of chronic
conditions, older adults also show higher mortality rates as a result of pollution exposure
when compared to middle aged individuals (Katsouyanni et al., 2001). Survivors of a
previous myocardial infarction, of which older adults are more likely to be, show greater
all around mortality later in life if exposed to air pollution (Berglind et al., 2009). While

xiii

the cardiovascular effects of air pollution are well documented, the cognitive dangers of
high pollution exposure are still being fully understood.

1.2.1 Composition of traffic-related air pollution (TRAP)
We review two broad groups of TRAP: the airborne particulate matter (PM) and the
vapor (gaseous) phase, with emphasis on the particulate matter component of TRP.
Urban TRP PM is a complex and heterogeneous mixture that includes residues from
fossil fuel combustion, organic chemicals, trace metals, nitrate, and sulfate. There are
also airborne components from brake linings and the vehicular chassis, as well as
roadway components and dust. The recognized size classes of airborne PM range from
coarse PM (> 10 μm diameter) to microscopic classes with aerodynamic diameters less
than 2.5 μm (PM2.5) and 0.1 μm, (PM0.1). For each class, primary emissions are
transformed from exposure to sunlight and atmospheric ozone and nitric oxides during
diurnal and seasonal cycles. While coarse PM are largely trapped by the upper airways,
smaller PM can impact the brain directly from olfactory neurons in the nasal mucosa, as
well as by systemic effects from the lower airways (Oberdorster et al., 2004). The
smaller PM sizes are associated with many pathological effects of air pollution (Hoek et
al., 2002, Zhou et al., 2007). Although some studies lump the two smaller size classes
under PM10, all three categories have notable adverse effects, as well as different
distributions in space and dispersion characteristics.
Of the three categories of particulate matter, PM2.5 (fine PM) has received the most
attention, with current US EPA standards (EPA 2013) of 12 μg/m3. The EPA has not yet

xiv

addressed PM0.1 (UFP, or ultrafine particulates). This class of TRP warrants attention
in public health because of experimental evidence for its greater cytotoxicity (Campbell
et al., 2005; Utell et al., 2000), potentially due to the greater penetration through cell
membranes (Zhang et al., 2004). One reason UFP has not been fully appreciated is due
to monitoring technology of PM based on weight and not particle number, where it is a
large percentage of the total PM. We note alternate terminology of nanoscale PM,
PM0.2, which encompasses a larger portion of the ultrafine particles, and is considered
alongside UFP in this review. UFP is associated with numerous adverse health effects,
and comprises the majority of all PM in combustion-derived exhaust (Zhang et al.,
2004). Their small size facilitates the crossing of physiological barriers, including the
blood-brain barrier and the placenta as discussed below (Pietroiusti et al., 2013;
Oberdorster et al., 2004; Wu et al., 2011).

1.2.2 Spatial dispersion of TRAP
The large body of evidence illustrating the harm of particulate matter inhalation makes a
clear argument for reducing exposure, but what exactly is a safe distance away from
sources of air pollution, and does this vary by the type of pollution? Concentrations of
UFP show a 25-fold increase at 30 meters (100 feet) from a freeway compared to
ambient pollution within the city (Zhu et al., 2002). The dilution of UFP from the source
into background ambient levels occurs in an exponential decline with levels generally
returning to background at 300 meters away from the source (Zhu et al., 2002). This is
measured by both total particle number and particle volume. A distance of 150 meters

xv

would create a reduced concentration of approximately 20% of the original output, as
measured by total particle number.
For discussion of developmental exposure, we briefly note that adverse health effects of
air pollution exposure increase with closeness to major road (Hoek et al., 2002; Peters
et al., 2004). UFP was reduced by approximately 80% at a distance of 150 meters from
the roadway (Zhou et al., 2007). Neither PM2.5 nor PM10 decreased substantially
within 150 meters (Rooda-Knape et al., 1998), and are decreased by less than 20% at a
distance of 400 m vs. 50 m from a roadway (Janssen et al., 2001). The rate of dilution of
UFP was correlated with increased cardiopulmonary mortality, inversely with distance
from roadways (Hoek et al., 2002).
One additional factor is the assumption that being indoors will reduce exposure. The
concentration of UFP indoors is similar to outdoor measurements; meaning that older
individuals indoors are still at risk for pollution exposure (Arhami et al., 2010). In fact,
outdoor derived particulate matter has also been shown to be one of the most harmful
components of indoor air pollution (Delfino et al. 2008).

1.2.3 Physiological penetrance of TRAP
One hypothesis for the mechanism underlying air pollution's physiological effects is
unique to the nanoscale particulate matter component. Nanoscale particles can
physically translocate to the olfactory bulb, via the olfactory epithelium (Oberdorster et
al., 2004; Elder et al., 2006). For in utero exposures, it is possible that the particles
might be able to cross the placental barrier to directly interact with the fetus, which

xvi

develops a blood-brain barrier by gestational day 16 (Stolp et al., 2013). There is
experimental evidence for transplacental transfer of nano-size PM. Titanium dioxide
nanoparticles (25–70 nm dia) subcutaneously delivered to pregnant mice on GD 3, 7,
10 and 14, were detected in male brains and testes six weeks postnatally (Takeda et
al., 2009). Thus, model nanoparticles can cross both the maternal placenta and the
blood-brain barrier of the developing fetus. Ex vivo models with human placenta and
polystyrene beads show strong size dependency, with 50 and 80 nm beads rapidly
crossing the placenta, possibly by simple diffusion; larger beads > 240 nm do not cross
(Wick et al., 2010). However, this ex vivo model does not represent potential
modification of PM by proteins and lipids, which create a bio-corona, altering the
movement of the nanoparticles (Nel et al., 2009). Engineered nanoparticles of this size
show 1000-fold range of translocation to the brain (0.00006% to 0.03%) (Park et al.,
2010; Kim et al., 2010). Although particles may cross secondary barriers (placenta,
blood-brain), their inhalation or ingestion does not necessarily allow transport to these
secondary barriers. This is of relevance because most experiments inject the particles
into the animal, bypassing the lungs. The small size of nanoparticles is important, as
particles < 34 nm rapidly translocate from lung to mediastinal lymph node (Choi et al.,
2010). Notable, negatively charged particles accumulated in secondary organs more
than positively charged particles (Schleh et al., 2012).
The placenta may be more vulnerable to nanoparticle entry later in gestation, when the
placental wall has thinned and is more vascularized, but also early in gestation before
the placenta is fully formed (Ornoy et al., 2011). The period after the placenta is formed,
but before maternal-fetal circulatory systems are fully developed, could be less

xvii

vulnerable to pollution exposure, due to minimal blood flow to the fetus. Nanoparticles
may even cause fetal damage without penetrating the placenta, e.g. in vitro,
nanoparticles can cause DNA damage even when they do not cross a cell barrier (Sood
et al., 2011). We note that these are not exclusionary hypotheses, and both may
potentially be occurring.

1.2.3 Experimental models of TRAP exposure
Several labs have developed rodent models studying the developmental effects of TRP
exposure, but no single paradigm has become widely accepted. The main findings
(Tables 1–6) include effects on brain morphology, behavior, inflammatory markers, and
neurotransmitters. Four experimental paradigms are currently used (Table 1): direct
diesel exhaust inhalation, diesel exhaust particle (DEP) oropharyngeal aspiration, the
Concentrated Air Particles delivery System (CAPS), and filter-trapped nano-sized
ambient reaerosolized particles. Most studies used inhalation, while one lab used direct
oropharyngeal aspiration of DEP.

Table 1. Air pollution exposure sources.
Exposure study Methods Particulate
matter
size and
composition
Volatiles Non-
tailpipe
pollution

Secondary
aerosols
Diesel exhaust

xviii

Direct from diesel
engine
Bolton et al 2012
Auten et al
2012
18–200 µm;
extractable
organic matter,
39.8%
Yes No No
Sugamata et al 2006,
Fujimoto et al 2005,
Yokota et al 2009,
Yokota et al 2013,
Suzuki et al 2010
Umezawa
et al 2011
All size particles CO, 2.67
ppm;
NO 2, 0.23
ppm,
SO 2, < 0.01
ppm
No No
Diesel exhaust
particles
Bolton et al 2013,
2014*
Auten et al
2012
Hougaard
et al
2008**
18-200 µm

No No No
Traffic-related air pollution
Ambient Air
Wei et al, in prep.
All size particles Yes Yes Yes
Concentrated
Ambient Particle
System (CAPS)
Allen et al
2013,
2014a,
2014b
< 100 nm dia. Yes Yes Yes
Filter-trapped nano-
sized PM (nPM)
Davis et al 2013
Morgan et
al 2011
< 200 nm dia.
Filter-bound, re-
aerosolized.
No Yes Yes
Non-tailpipe pollution includes brake dust, tire erosion, and roadway dust. Secondary aerosols arise from
alterations of particulate matter by temperature, sunlight, humidity, ozone etc. *Particles delivered by
oropharyngeal introduction. **Diesel exhaust particles reaerosolized for inhalation.

Diesel exhaust: Pregnant mice were exposed to the whole exhaust stream from a diesel
engine, diluted to concentrations ranging from 0.171–3.0 mg/m3. Auten et al. 2012 and
Bolton et al. 2012 utilized a 6.4 hp direct injection single cylinder 320-cc Yanmar L70V

xix

diesel generator, operating at a constant 3600 rpm. Yokota et al. 2009 used a 2369-cc
diesel engine, operating at 1050 rpm. The exhaust includes volatile gaseous
components, notably CO, SO2, and NO2. Unlike other exposure paradigms, these
particles are not filtered by size, and retain native charges. The direct diesel exhaust
paradigm is missing other real world pollutants from vehicular traffic, e.g. rubber from
tire erosion, brake lining debris, and reaerosolized dust from roadways. Moreover,
diesel engine exhaust can represent only one type of vehicle, and the particles are
being directly emitted and thus did not undergo the secondary reactions from heat and
sun exposure, which develop as a function of time after emission.
Diesel exhaust particles (DEP)
Oropharyngeal Aspiration: (Auten et al., 2012) This is the only non-inhalation
paradigm used in the prenatal studies. Diesel exhaust particles are collected from a
single cylinder diesel engine, and then 50 μg of diesel exhaust particles (DEP) are
suspended in 50 μL of PBS with 0.05% Tween-20 and delivered by oropharyngeal
aspiration. Importantly, this delivery method bypasses nasal inhalation.
Reaersolized Inhalation: (Hougaard et al., 2008) Obtained from the National
Institute of Standards and Technology, Standard Reference Material 2975, these
particles are obtained from a diesel powered forklift, and are re-aerosolized for
inhalation delivery (Hougaard et al., 2008; Hougaard et al., 2009). Importantly, the re-
aerosolized DEP, like the resuspended DEP for oropharyngeal aspiration, lack gaseous
and volatile components. Also, because the particles are suspended in water, they are
depleted of insoluble PM. Elimination of insoluble particulate matter is of special
relevance, as this includes black carbon and polycyclic aromatic hydrocarbons (PAHs).

xx

Concentrated Ambient Particle System (CAPS): Ultrafine fractioned particulate
matter is concentrated next to a roadway for direct real time delivery at 10–20 times
ambient concentration (Demokritou et al., 2003). CAPS maintain ambient components,
including gases and volatiles, and the native charges of the particles. Importantly,
condensing the particles does not alter their natural size distributions, and does not
amplify aggregation (Demokritou et al., 2003; Campbell et al., 2005). Limitations of this
system are its dependence on the current traffic patterns, which fluctuate diurnally and
seasonally.
Filter-trapped nano-sized PM from urban TRP: This paradigm, developed by
Constantinos Sioutas at the University of Southern California (Morgan et al., 2011),
collects ambient air particulate matter, PM0.2, on the roadside next to a high traffic
source with a high-volume ultrafine particle sampler on 0.2 μm pore Teflon filters.
Collections are made for 4–6 weeks in the fall to encompass the range of temperatures
and moisture in Southern California (Daher et al., 2013; Saffari et al., 2014). The
collection is continuous and includes secondary transformations during the diurnal
cycle. Besides combustion products, the sample includes reaerosolized roadway dust,
and traces from brake lining and tire erosion that are < 0.2 μm. The filters are sonicated
in distilled water to yield a suspension, which is stored frozen until use. The
reaerosolized PM has average particle size of 60 nm at a density of 350 μg/m3, which is
about 25× greater than ambient concentrations of that size range of particle (Daher et
al., 2013; Saffari et al., 2014). The resuspension lacks ambient gases and is depleted in
water insoluble organic species including PAHs and black carbon (Morgan et al., 2011).
We designated these materials as nPM (nanoparticulate matter) in distinction from the

xxi

size class of ambient UFP in the literature. After re-aerosolization, rodents are exposed
to nPM together with ambient pollutants in the exposure room, which are 35–50% below
outdoor ambient levels, while control animals have this air filtered by HEPA filters.
In summary, each experimental paradigm represents trade-offs. While DE is the
most readily obtained, they are model emissions of one engine type and lack secondary
atmospheric transformations of ambient pollution. The CAPS fully capture TRP for PM,
gases, and volatiles, but vary diurnally and seasonally. The nPM capture diurnal
variations, but the resuspension is deficient in black carbon (BC) and PAH among other
water insoluble organic compounds.

1.3.1 Experimental effects of TRAP- Body weight
Epidemiological findings of air pollution on birth weight are corroborated in some
rodent models (Table 2). Mice exposed to 3.0 mg/m3 DEP had decreased fetal weight
as early as gestational day (GD) 18 (equivalent to the human third trimester) (Fuhimoto
et al., 2005). Several studies observed a weight decrease four weeks after birth in mice
exposed to DEP (Table 3) (Auten et al., 2012; Hougaard et al., 2008). Finally, two
studies observed a reversal later in life, with increased weight at four months of age in
mice (Bolton et al., 2012), and at eight weeks in rats (Wei et al., 2016). Decreased
weight at weaning (Hougaard et al., 2008; Auten et al., 2012), combined with increased
weight later in life (Bolton et al., 2012; Wei et al., 2016), corresponds with previously
documented rebounding in weight after a prenatal stressor (Ornoy et al., 2011). Air
pollution also has an additive effect on weight when exposure occurs in utero, followed
by a high fat diet (HFD = 45% calories from fat) after birth (Table 4) (Bolton et al., 2012;

xxii

Bolton et al., 2014). The prenatal DEP + HFD treated mice showed significant weight
gain over high fat diet alone, a compounding effect similar to what was observed for
inflammatory responses (increased cytokines, microglial activation) (Bolton et al., 2012;
Bolton et al., 2014). Males in both replicates showed greater weight gain from treatment
(Bolton et al., 2012; Bolton et al., 2014). However, females showed 4.4× greater weight
gain for the first experiment (Bolton et al., 2012), and no change in the second (Bolton
et al., 2014). Notably, the first experiment used diesel exhaust inhalation, while the
second employed oropharyngeal aspiration of DEP, which excludes gas components.
Thus, it is possible that either a species lost in the conversion to suspended diesel
particles, or the inhalation delivery route caused the weight gain. The high fat postnatal
group showed insulin resistance with elevated serum insulin in the males (Bolton et al,
2012, Bolton et al, 2014). Females showed only a change in serum leptin, while males
did not show any differences from pollution exposure.
Exposure to highly polluted ambient Beijing air (75.3 μg/m3) caused, worsened
lipid profiles and weight gain in both rat mothers and offspring (Wei et al, 2016).
Pregnant dams had higher low-density lipoprotein (LDL), total cholesterol, triglyceride,
and overall weight. Pups had increased weight at eight weeks, and worsened lipid
profiles, with increased LDL, total cholesterol, and triglyceride, and decreased in HDL.
Rodent models also corroborated mid-life and gestational weight effects (Bolton et al.,
2012; Bolton et al., 2014; Sun et al., 2005; Sun et al., 2009).
Table 2. Developmental exposure- Body weight, behavior, and cell-molecular responses.
Study Exposure
protocol
Weight, gross
changes
Behavior Cell-molecular
changes

xxiii

Diesel exhaust direct
Bolton et al 2012,
2013; GD18 (♂ + ♀)
0.5 or 2.0
mg/m
3

4 h/d; GD 7-
17
Weight at 4m:
♂ -1.1
♀ - NC
Bolton 2012:
Open field
activity at 4m:
♂ -0.7 ♀ -NC
Bolton 2013:
NC in forced
swim
CCL2/MCP-1 -
3.5
CX3CL1/
fractaline - 1.5
Auten et al 2012
GD18 (♂ + ♀)
0.5 or 2.0
mg/m
3

4 h/d; GD 9-
17
Weight at 4w:
0.5mg/m
3
(♂+♀) 0.9
2.0 mg/m
3
(♂+♀)
NC
eotaxin: 4
KC: 6
RANTES: 10 +
Fujimoto et al 2005;
GD14
0.3, 1.0, or 3.0
mg/m
3

12h/d; GD 2-
13
↑ placental weight
(♂,♀- 1.0 mg/m
3
)
↓ fetal weight (♂,♀-
3.0 mg/m
3
)

Yokota et al 2013
♂ only- behavior at
5w
1.0 mg/m
3

8 h/d; GD2-17
↓ Retention
time on
rotating rod
Cliff avoidance
latency to
jump 0.68

Yokota et al 2009
5w- behavior
1.0 mg/m
3

8 h/d; GD2-17
Spontaneous
motor activity:
♂- 0.83

Suzuki et al 2010
5w
0.171 mg/m
3
8 h/d, 5 d/w
GD 2-16
↓ Spontaneous
locomotor
activity

DEP
Hougaard et al 2008-
19w (♂+♀)
19 mg/m
3

DEP
1 h/d; GD 9-
19
Weight at 4w: 0.9 No effect in
the Morris
water maze
No indication of
any DNA
damage, nor
inflammation

xxiv

CAPS
Allen et al 2013, Allen
et al 2014a
PND60
15-240 ug/m
3

4h/d ; PND 4-
7 & 10-13
4h/d; PND56-
60
♂↓ Response
rates for FI60
(6mo)
Novel object
performance
(6mo):
♂ - 0.5, ♀ - 0.8

Allen et al 2014b
2w or 8w
200,000
particles/cm
3

96 ug/m
3
4h/d
PND 4-7 &
10-13
Lateral ventricle
size:
PND14: ♂ 3.2
PND55: ♂ 1.8
GFAP PND14:
Hipp: ♂ - 0.5, ♀ -
1.9
CC: ♂ - NC, ♀ -
1.5
IBA1 PND 55:
Hipp: ♂ - NC, ♀ -
NC
AC: ♂ - 1.3, ♀ -
NC
Filter-trapped nPM

Davis et al 2013 - 8m
350ug/m
3

5h/d
3d/w, 10 w
Tail
suspension
immobility
8m: ♂ - 2.6,
♀- NC
PD3 - JNK1 (♂,
♀) 0.7
Hipp GLU- NC
Abbreviations: AC, Anterior commissure; CAPS (Concentrated Ambient Particle System); CC, Corpus callosum;
CCL2, Chemokine (C-C motif) ligand 2; CX3CL1, Chemokine (C-X3-C motif) ligand 1; DEP, diesel exhaust
particles; FI60, Fixed interval reward 60 sec; GFAP, Glial fibrillary acidic protein; GD Gestation day; GLU,
Glutamate; Hipp, Hippocampus; IBA1, ionized calcium binding adaptor molecule 1; JNK-1, c-Jun N-terminal kinase
1; KC, keratinocyte chemoattractant; m, months; MCP-1, monocyte chemotactic protein 1; NC, no change; RANTES,
regulated on activation, normal T cell expressed and secreted; w, weeks.

xxv

1.3.2 Experimental effects of TRAP- Behavioral
Behavioral changes from pollution exposure include cognitive and locomotor
deficits (Table 2). Adult exposures show spatial memory impairments (Fonken et al,
2011). Cognitive deficits from developmental exposures include depressive symptoms,
impaired short-term memory, and decreased response rates for fixed interval sixty-
second (FI60) reward tests (Allen et al., 2014; Davis et al., 2013). Mice exposed in utero
to nPM showed increased immobility on a tail suspension test, which is a marker for
depression (Davis et al., 2013). Only males were vulnerable, with a decreased delay
until first period of immobility, and 2.6× longer immobility versus control, which
implicates activation of the amygdala (Dunlop et al., 2007; Yang et al., 2010).
Correspondingly, prenatal exposure to DEP (maternal oropharyngeal DEP aspiration)
increased adult amygdala levels of monoamine neurotransmitters (dopamine, dopamine
metabolites, and serotonin) (Yokota et al., 2013). However, the oropharyngeal DEP
model did not alter adult forced swim performance, another marker for depression
considered equivalent to tail suspension (Bolton et al., 2012). This discrepancy could be
due to the different age at assessment (8 mo vs. 4 mo), or the different pollutant
models.
Impaired short-term memory was observed in neonatal mice exposed to CAPS,
from postnatal day (PND) 4–7 and 10–13 and assessed at 2 months of age by the novel
object recognition test. In the one-hour posttest, CAPS exposed mice spent more time
with the familiar versus novel object, indicative of impaired short-term memory (Allen et
al., 2014). The FI60 test, a model of impulsivity, showed decreased response and run
rates, but only for males (Allen et al., 2014). Despite smaller overall response rates,
there was no significant difference in learning. In a separate experiment, conducted with

xxvi

the same exposure protocols, a secondary dose of pollution from PND 56–60 caused
deficits in a fixed interval waiting for reward test, a classic model of impulsivity control
(Allen et al., 2014b). These experiments included CAPS exposure on postnatal days 4–
7 and 10–13. The sensitivity to neonatal exposure is important because neonatal rodent
nervous systems are relatively less mature compared to humans (Rice et al., 2000).
Deficits in short-term memory of neonatal CAPS exposure may be mediated by
glutamatergic changes. Glutamate levels in the hippocampus, which is critical for spatial
learning and memory, were increased 1.26-fold in the male CAPS exposed mice (Allen
et al., 2014b). Though the effect returned to baseline by eight weeks of age, the
transient glutamatergic increase during development could cause persisting effects,
including excitotoxicity. Detailed studies of hippocampal circuit functions, e.g. LTP, and
synaptic density are needed. Increased inflammatory cytokine levels (Il-6, IL-1b, TNF-a)
are also relevant to behavioral deficits through their impact on synaptic plasticity. These
cytokines showed complex changes in different brain regions in mice exposed to
neonatal CAPS (Allen et al., 2014b; Block et al., 2009). For example, TNF-a modulates
glutamate, and potentiates the cell to glutamatergic excitotoxicity (Zou et al., 2005)
which could alter short-term memory later in life.
Locomotor deficits from prenatal exposure to pollutants include decreased
spontaneous motor activity and impaired balance (Table 4) (Yokota et al., 2009; Bolton
et al., 2012; Yokota et al., 2013; Suzuki et al., 2010). Intriguingly, only males have
shown decreased spontaneous motor activity in studies from several labs that include
ages from 5 weeks to 5 months (Yokota et al., 2009; Block et al., 2009; Suzuki et al.,
2010). Decreased spontaneous motor activity at age 2 months in CAPS studies was

xxvii

only observed when paired with a second treatment from PND 56–60 (Table 4) (Allen et
al., 2014). Balance was impaired on the rotating rod test in prenatally exposed male
mice at 5 weeks (Yokota et al., 2013). These mice also had decreased latency in the
cliff avoidance test (Yokota et al., 2013). The impaired performance on these two
balance tests is not attributable to differences in body weight (Yokota et al., 2013); only
Bolton et al. 2012 reported weight differences (Bolton et al., 2014). We note that a shift
from direct exposure to diesel exhaust to the oropharyngeal DEP in the same lab did
not to replicate these effects (Bolton et al., 2014).
Table 3. Developmental exposure- shared responses to traffic related air pollution across
multiple experiments.
Shared
responses
No Change
Body weight Auten 2012 ↓ 4w (DE),
Bolton 2012 ↑ 5m (DE),
Fujimoto 2005 ↓ GD14 (DE),
Hougaard 2008 ↓ 4w (DEP), Sugamata
2006 ↓ 4w (DE),
Umezawa 2011 ↓ 8w (DE),
Wei 2016↓ 2w & 3w (Bejing Air)
Allen 2014a, Bolton 2014, Davis
2013, Hougaard 2009, Suzuki
2010, Yokata 2013
Spontaneous
locomotor
activity
Bolton 2012 ↓ 5m (DE),
Hougaard 2009 ↓ 8w (DE),
Suzuki 2010 ↓ 5w (DE),
Yokata 2009 ↓ 5w (DE)
Allen 2013, Allen 2014a, Bolton
2014, Davis 2013
Cortex-
dopaminergic
Allen 2014a ↓ 9m (PND 55 CAPS),
Allen 2014b ↑ 2w & 8w (PND14
CAPS),
Suzuki 2010 ↑ 5w (DE),
Yokata 2013 ↓ 3w (DE)

Microglial
activation
Allen 2014a ↑ 9m (PND14 + 55
CAPS),
Allen 2014b ↑ 8w (CAPS),
Bolton 2013

xxviii

Bolton 2012 ↑ GD18 (DE),
Bolton 2014 ↑ 6w (DEP+HFD)
Abbreviations: CAPS, Concentrated Ambient Particle System; DE, diesel exhaust; DEP, diesel
exhaust particles; HFD, high fat diet; PND, Postnatal day

1.3.3 Experimental effects of TRAP- Gross brain morphology
Gross brain weight has not shown sensitivity to prenatal exposure to nPM (Davis
et al, 2013). However, one study of neonatal exposure to concentrated ambient TRP
reported gross enlargement of the lateral ventricles, particularly in males (Allen et al.,
2014). Conversely, nPM exposed rats show smaller lateral and third ventricles
(Woodward et al., unpublished). Cerebral vasculature has just begun to receive
attention in air pollution models. After maternal intranasal exposure to black carbon,
young adult mice had focal induction of GFAP in astrocyte endfeet on capillary
endothelia and altered arterial macrophage granules (Onoda et al., 2014). These
reports point to structural changes that could underlie cognitive dysfunctions from
prenatal exposure.
Table 4. Postnatal exposure, secondary manipulations.
Study (secondary
treatment)
Behavior Weight Serum Brain cellular Brain subcellular
Diesel exhaust- direct
Bolton et al 2012- 6m
DE+HFD
HFD- 45% fat,
beginning 4m, for 6w
Weight
♂ - NC
♀ - 4.4
Leptin:
♂ - NC
♀ - 1.7
Insulin:
♂ - 6.4
♀ - NC
IBA1:
Hypothalamus:
♂ - 1.7, ♀ - 1.3
Dentate gyrus:
♂ - 1.4, ♀ - 1.5

xxix

Amygdala:
♂ - 1.3, ♀ - NC
Hipp CA1:
♂ - 1.3, ♀ - NC
Bolton et al 2013
DEP + Nest
Restriction
DEP: 50ug DEP
every 3 d, from GD 2-
17
NR: E14-19
Contextual
fear recall:
♂ - 0.4
♀ - NC
Anxiety
behavior:
♂ - 1.6
♀ - 1.2
No
change
GD18
TLR4: ♂ - 1.3, ♀ -
NC
Casp-1:
♂ - 1.4, ♀ - NC
IL-1b: ♂ - NC ♀ -
NC
Il-10: ♂ - 0.8, ♀ -
1.4
Bolton et al 2014
DEP + High Fat Diet
DEP: 50ug DEP
every 3 days, from
GD 2-17
HFD: 45% Fat,
beginning 4m, for 9w
Anxiety:
(elevated
zero
maze)
♂ - 1.1
♀ - NC
Weight:
♂- 1.3
♀ - NC
Insulin:
♂ - 1.8
♀ - NC
↑ peripheral
macrophage
infiltration in
hypothalamus (♂
1.2)
GFAP Hipp NC
Hipp:
CD11b:
♂ - 1.6, ♀ - NC
TLR4: ♂ - 1.2, ♀ -
NC
CXC3CR1:
♂ - 1.2, ♀ - NC
CAPS
Allen et al 2013
Allen et al 2014a -9m
CAPS: PND14 and/or
PND55
Waiting
for reward
behavior:
♂ - ↓
Corpus Callosum:
IBA1:
♂ - 3.4, ♀ - NC,
C/A
♂ - 3.3, ♀ - NC,
A/C
Hipp:
GLU:
♂ - 1.6, ♀ - 0.6, C/C
DOPAC:
♂ - NC, ♀ - 1.2,
C/C
Abbreviations: A/C, Air/CAPS; C/A, CAPS/Air; CAPS, Concentrated air particle system; Casp-1, Caspase-1;
CD11b, Cluster of differentiation 11 beta; CXC3CR1, Chemokine (C-X3-C motif) receptor 1; DOPAC, 3,4-
Dihydroxyphenylacetic acid; GFAP, Glial fibrillary acidic protein; GLU, Glutamate; IBA1, Ionized calcium
binding adaptor molecule 1; IL-1b, Interleukin-1 beta; IL-10, Interleukin 10; IR, Insulin resistance; HFD, High
fat diet; Hipp CA1, Hippocampus, cornu Ammonis area 1; NC, No change; NR, Nest restriction; TLR4, Toll-
like receptor 4.

xxx

1.3.4 Experimental effects of TRAP- Neuronal changes
We note a major gap between the body of epidemiological evidence for TRP
associations with brain development and the scant information on neuronal changes in
animal models of prenatal TRP exposure. Adult TRAP exposures have shown effects
on neuronal processes, including CA1 specific decreases in neuronal processes and
spine density (Fonken et al, 2011, Woodward et al, unpublished), as well as decreased
GluA1 in the hippocampus (Morgan et al., 2011) (Table 5). Reports on neuronal
changes are scattered among different neurotransmitters, often in different brain
regions, giving little cohesion of results (Table 6). Dopamine levels in the cerebral cortex
illustrate the diversity. Mice exposed to prenatal diesel exhaust had lower cortical
dopamine for males at 3 weeks, but no change at six weeks (Yokota et al., 2013).
However, the same exposure paradigm in a different lab showed increased cortical
dopamine at 5 weeks (Suzuki et al., 2010). These studies used > 5-fold different levels
of DEP density. The turnover of dopamine, estimated by the ratio of the catabolite
DOPAC to dopamine, (DOPAC: DA) was higher in neonatally CAPS exposed male mice
at two and eight weeks (Allen et al., 2014).
Neurotransmitter changes reported for adult rodent TRP exposures have not
been borne out by prenatal exposures, potentially indicating different mechanisms. We
observed decreased glutamate receptor 1 (GLUR1) in the hippocampus in mice
exposed to nPM at age three months (Morgan et al., 2011). However, prenatal
exposure did not alter hippocampus GLUR1 at eight months (Davis et al., 2013).
Neonatal exposure to DEP transiently increased hippocampal glutamate at two weeks,
with return to baseline by eight weeks (Allen et al., 2014b).

xxxi

Cortical neurons harvested from one day old pups prenatally exposed to nPM
showed impaired differentiation and neurite initiation, with fewer stage 3 neurons,
compared to controls (Davis et al., 2013). These pilot studies give a model for linking
developmental exposure to alterations in neurons and glia.

xxxii

CA1 CA3 DG Cortex Olfactor
y Bulb
Cerebellu
m
Midbrai
n
Tempor
al lobe
Whole
brain
Citation
Neuronal
morphology
Dec-
processe
s
N/
C
Woodwar
d 2016
Dec-
processe
s
N/
C
Cacciottol
o 2016
Dec-
spine
density
+ length
Dec-
dendritic
length +
complexit
y
N/
C
Fonken
2011
MBP Dec N/
C
Woodwar
d 2016
Glial
activation
Inc-
Iba1
Inc-
Iba
1
Woodwar
d 2016
Inc-
GFAP
Kleinman
2008
Inc-
GFAP
Morgan
2011
Inc-
GFAP
Veronisi
2005
Glutamaterg
ic system
Dec- GluA1 Morgan
2011
Dopaminergi
c system
Dec- DA
neurons
(substanti
a nigra)
Veronisi
2005
BBB Inc
permeabilit
y (NaF),
MMP-9,
MMP-2
Dec- TJ
prot
Oppenhei
m 2013-
ApoE -/-
Oxidative
stress
Inc-
DHE

Inflammator
y cytokines
and cytokine
activation
factors
Inc-
TNFa
N/C-
CD68
Inc- TNFa,
CD68
Cheng
2016
(EHP)
Inc-
TNFa
Inc-
TNFa
Levesque
2011
Inc-
TNFa,
IL-1a,
Morgan
2011

xxxiii

Table 5. Adult exposure- Regional brain responses Abbreviations: Dec, decrease; Inc,
increase; N/C, no change; AP-1, activator protein 1; CA1, cornus ammonis 1; CA3, cornus ammonis 3; CCL2,
chemokine ligand 2; CCL3, chemokine ligand 3; CCL4, chemokine ligand 4; CD14, cluster of differentiation
14; CD68, cluster of differentiation 68; CXCL1, chemokine ligand 1; DA, dopamine; DG, dentate gyrus; DHE,
dihyrdroethidium; GFAP, glial fibrillary acidic protein; GluA1, glutamate receptor subunit 1a; Iba1, ionized
calcium-binding adapter molecule 1; IL-1a, interleukin 1a; IL-1b, interleukin 1b; IL-6, interleukin 6; TNFa,
tumor necrosis factor alpha.

Table 6. Developmental exposure- Neurotransmitter responses.
Experiment Yokota 2013 Suzuki 2010 Allen 2014 Allen 2014b
Exposure
Condition
Prenatal
exposure
Prenatal
exposure
Neonatal
Exposure
Neonatal Exposure
Age ♂
3w
♂
6w
♂
5w
♂
8w
♀
8w
♂
2w
♀
2w
♂
8w
♀

8w
Cortex
CD68,
CD14
Inc- TNFa,
IL-1b
N/C-GFAP
Campbell
2009
(ApoE -/-)
Inc-
activate
d NF-
kB
Inc- TNFa,
IL-1a
Campbell
2005
Inc- Ap-
1, NF-
kB
Inc-
pJNK
Kleinman
2008
Inc-
TNFa
Inc-
TNFa
N/c- TNFa Inc-
TNFa,
IL-1b,
CCL4
N/C -IL-
6
Inc-
TNFa
Levesque
2011b
N/C- IL-1b, TNFa, CCL2,
CLL3, CXCL9
Inc- IL-
1b,
TNFa,
CCL2,
CCL3,
CXCL9
Win-Shwe
2006

xxxiv

DA 0.6 NC 1.6 NC NC
DOPAC NC NC 2.0 NC NC
DOPAC:DA NC NC NC NC
HVA NC NC NC
3-MT NC NC 1.5
GABA NC NC
GLU NC NC
NE 2.0 NC NC NC NC 1.2 0.9
MHPG 1.4
NM 1.5
5-HT NC 1.3
5-HIAA NC NC
Hippocampus
DA NC
DOPAC NC NC NC
DOPAC:DA 1.4 NC
HVA 0.2 NC NC
3-MT NC
GABA NC NC NC NC
NE NC NC NC
MHPG 0.7
NM NC
5-HT NC NC NC 1.3
5-HIAA NC NC
Midbrain

xxxv

DA NC NC
DOPAC:DA NC 0.7 NC NC
HVA 0.7 NC NC NC NC
3-MT 0.7
NE NC NC NC
MHPG 0.6
NM NC
5-HIAA NC NC
Striatum
DA NC 0.8 NC
DOPAC NC NC NC
DOPAC:DA NC NC
HVA NC 0.9 NC
3-MT NC
NE NC NC NC
MHPG 0.8
5-HT NC NC NC NC
5-HIAA NC NC 0.7 NC
Cerebellum
DA NC
HVA NC
NE NC
MHPG 0.8 0.8 0.6
NM NC 0.6 NC
5-HT NC NC

xxxvi

5-HIAA NC NC
Hypothalamus
DA NC NC
DOPAC NC NC
DOPAC:DA NC NC
NE NC 1.2 0.7 NC
HVA NC NC
5-HT NC NC
5-HIAA NC 1.4
Amygdala
DA 1.5 NC
DOPAC 1.4 1.3
HVA 1.4 NC
3-MT 1.4 NC
5-HT 1.4 NC
5-HIAA 1.3 NC
Fold changes (approximated from figures when not given). Abbreviations: DA, dopamine; D OPAC, 3,4-
dihydroxyphenylacetic acid (DA catabolite) ; GABA, gamma-aminobutyric acid; GLU, glutamate; HVA,
homovanillic acid (DA catabolite); MHPG, 4-hydroxy-3-methoxyphenylglycol (NE catabolite) ; NC, No
change; NE, norepinephrine; NM, normetanephrine (NE catabolite); 3-MT, 3-methoxytyramine (NE catabolite);
5-HIAA, 5-hydroxyindoleacetic acid (serotonin catabolite); 5-HT, serotonin.

1.3.5 Experimental effects of TRAP- Inflammatory changes
Inflammation may be a major mediator of maternal systemic and placental
responses to air pollution exposure (Block et al., 2009). Adult exposure to TRAP
increases serum TNFa levels in both animal models (Levesque et al., 2011; Li et al.,
2013), and in older humans (Delfino et al., 2008). Systemic inflammation in the mother

xxxvii

increases circulating inflammatory cytokines, influencing the development of the fetus,
through methods such as the activation of microglia (Table 7) (Bolton et al., 2012;
Aaltonen et al., 2005). As noted above for neuronal changes, there is a need for similar
protocols across labs in the investigation of inflammatory effects.
Numerous cytokines are induced by TRAP exposure in adult mice, including
TNFa in the hippocampus, cortex, olfactory bulb, midbrain, and temporal lobe (Cheng et
al., 2016; Levesque et al, 2011; Levesque et al., 2014; Win-shwe et al., 2006) (Figure
5). Prenatal exposure of mice to diesel exhaust rapidly increased cytokines (IL-1b, IL-6,
IL-10, and TNF-a) on GD18 (Auten et al., 2012; Bolton et al., 2012). Microglial activation
is suggested by increased chemokines CCL2/MCP-1 and CX3CL1/Fractalkine (Bolton
et al., 2012). After CAPS neonatal exposure, proinflammatory cytokines (IL-6, IL-1b,
TNF-a) were decreased at 2 weeks in males (Allen et al, 2014b). However, by eight
weeks, a full month after the cessation of exposure, IL-1b and TNF-a rebounded to
levels 1.4-fold above control's in midbrain (Allen et al., 2014b). Females showed a
different time course, with little upregulation immediately following exposure, yet still
showing delayed increases a month later. This effect was brain region specific: unlike
the midbrain, the striatum had lower cytokines at the same times (Allen et al., 2014b).
These sex differences in cytokine responses between different brain regions clearly
show that inflammatory effects of pollution exposure must be studied in terms of brain
pathways and cannot be generalized to the entire brain.
Glial inflammatory responses are detected by the astrocyte specific GFAP
(intermediate filament glial fibrillary acidic protein) and the microglial marker of IBA1
(ionized calcium-binding adapter molecule 1). In adult exposures, microglial activation is

xxxviii

observed in the CA1 (Woodward et al, unpublished), and astrocytic activation is seen in
the cortex and midbrain (Morgan et al., 2011; Veronisi et al., 2005). These responses
were observed in neonatal exposures, and prenatal exposures compounded with a
secondary stimulus. Neonatal mouse exposures from PND 4–7 and 10–13 increased
GFAP in the hippocampus, corpus callosum, and anterior commissure in females, while
males responded with decreased GFAP (Allen et al., 2014b). These measurements
were made immediately following exposure, at two weeks. IBA1 was upregulated in the
hippocampus and corpus callosum for males, at eight weeks and 9 months, respectively
(Allen et al., 2014; Allen et al., 2014b). Females showed no change for IBA1. Adult
exposures showed both acute microglial activation (Cheng et al., 2016), and chronic
activation (Morgan et al., 2011). Adult mice of several genotypes showed brain
inflammatory responses, with induction of IL-1a and TNFa, and activation of microglia
and astrocytes (Campbell et al., 2005).
Inhalation of prenatal diesel exhaust, as well as oropharyngeal aspiration of DEP,
did not elicit any changes in GFAP or IBA1 in mice, examined at six months of age
(Bolton et al., 2012; Bolton et al., 2014). However, a high fat diet (HFD) starting at 4
months and lasting for six weeks, along with the prenatal DE exposure, increased IBA1,
but not GFAP (Bolton et al., 2012). IBA1 was increased in hypothalamus, dentate gyrus,
amygdala, and the CA1 of the hippocampus for males (Bolton et al., 2012). Females
showed changes only in the hypothalamus and dentate gyrus (Bolton et al., 2012). This
upregulation of IBA1 supports the hypothesis of air pollution exposure during gestation
as an enhancer for later life environmental insults: the DE+HFD group responded more
than either treatment alone, with changes in brain regions that did not change with only

xxxix

one of the treatments (Table 3). The two-hit hypothesis postulates that the first insult
primes the system (Bolton et al., 2012; Block et al., 2009). While inflammation may not
be activated by a single insult, the second inflammatory challenge may cause a
disproportionately larger response.
Table 7. Developmental exposure- Cytokine changes.
Allen 2014b [58]

Bolton
2012
[48]
♂
2w
♀
2w
♂
8w
♀
8w
♂+♀
GD18
Cortex
IL-6 0.5
IL-1b NC NC
Hippocampus
IL-6 NC NC
IL-1b NC NC
Midbrain
IL-6 NC 2.1
IL-1b 0.5 NC 1.4 NC
TNF-a 0.5 NC 1.5 NC
Striatum
IL-6 0.4 NC
IL-1b NC NC
TNF-a NC 0.7 NC NC

xl

Whole Brain
IL-6 3.7
IL-1b 5.5
TNF-a
IL-10 3.5
Abbreviations: GD, Gestational day; NC, No change; PND, Postnatal day.

1.3.6 Experimental effects of TRAP- Sex differences
Sex differences are apparent in rodent responses to prenatal exposure, with
greater male vulnerability observed. For open field activity, only male mice showed
deficits (Bolton et al., 2012). For the tail suspension test, a measure of depressive
behavior, only males were responsive, with no effect for females (Davis et al., 2013).
Further sex differences were shown in secondary treatments (Table 3). Release of
inflammatory cytokines is profoundly affected by sex, a trend even more pronounced
with the addition of a secondary insult. When prenatal diesel exposure is combined with
either six weeks HFD, or nest restriction from gestational day (GD) 14–19, male mice
show significant effects in numerous cytokines and chemokines, while females show no
changes (Bolton et al., 2014; Bolton et al., 2013). The combination of prenatal diesel
exposure and adult high fat diet increased serum insulin, insulin resistance, and IL-1b,
again only in male mice (Bolton et al., 2014). Brain inflammatory proteins CD11b, TLR4,
and CXC3CR1 were increased only in males (Bolton et al., 2014). Likewise, only males
had increased peripheral macrophage infiltration in the hypothalamus (Bolton et al.,
2014). Furthermore, for nest restriction paired with diesel exposure, only males showed

xli

decreased contextual fear recall, and changes in brain TLR4, caspase-1, IL-1b, and IL-
10 (Brook et al., 2008). Finally, in the combination of PND 4–7, 10–13, and 55–60
CAPS exposure, only males showed increased IBA1 staining in the corpus callosum
(Allen et al., 2014). These experimental findings demonstrate greater male vulnerability
to pollution exposure, especially when combined with a second insult.

1.4 Bibliography
1. Aaltonen, R., Heikkinen, T., Hakala, K., et al (2005). Transfer of proinflammatory
cytokines across term placenta. Obstet Gynecol, 106, 802-807.
2. Allen JL, Conrad K, Oberdörster G, et al (2013) Developmental exposure to
concentrated ambient particles and preference for immediate reward in mice.
Environ Health Perspect 121:32-8.
3. Allen JL, Liu X, Weston D, et al (2014) Developmental exposure to concentrated
ambient ultrafine particulate matter air pollution in mice results in persistent and sex-
dependent behavioral neurotoxicity and glial activation. Toxicol Sci 140:160-78.
4. Allen JL, Liu X, Pelkowski S, et al (2014). Early postnatal exposure to ultrafine
particulate matter air pollution: persistent ventriculomegaly, neurochemical
disruption, and glial activation preferentially in male mice. Environ Health Perspect.
122:939-45.
5. Auten RL, Gilmour MI, Krantz QT, et al (2012). Maternal diesel inhalation increases
airway hyperreactivity in ozone-exposed offspring. Am J Respir Cell Mol Biol.
46:454-60.
6. Berglind, N; Bellander, T; Forastiere, F; von Klot, S; Aalto, P; Elosua, R; Kulmala, M;
Lanki, T; Löwel, H; Peters, A; Picciotto, S; Salomaa, V; Stafoggia, M; Sunyer, J;
Nyberg, F (2009). Ambient Air Pollution and Daily Mortality Among Survivors of
Myocardial Infarction. Epidemiology, Vol 20; Issue 1, pp 110-118. doi:
10.1097/EDE.0b013e3181878b50.
7. Bolton JL, Smith SH, Huff NC, et al (2012). Prenatal air pollution exposure induces
neuroinflammation and predisposes offspring to weight gain in adulthood in a sex-
specific manner. FASEB J. 26:4743-54.
8. Bolton JL, Huff NC, Smith SH, et al (2013). Maternal stress and effects of prenatal
air pollution on offspring mental health outcomes in mice. Environ Health Perspect.
121:1075-82.
9. Bolton JL, Auten RL, Bilbo SD (2014). Prenatal air pollution exposure induces
sexually dimorphic fetal programming of metabolic and neuroinflammatory outcomes
in adult offspring. Brain Behav Immun.37:30-44.

xlii

10. Block, M. L., & Calderón-Garcidueñas, L. (2009). Air pollution: mechanisms of
neuroinflammation and CNS disease. Trends in neurosciences, 32, 506-516.
11. Brook, R. D., Jerrett, M., Brook, et al (2008). The relationship between diabetes
mellitus and traffic-related air pollution. Int J Occup Environ Med, 50, 32-38.
12. Campbell A, Oldham M, Becaria A, et al (2005). Particulate matter in polluted air
may increase biomarkers of inflammation in mouse brain. Neurotox. 26:133-140.
13. Chen JC, Schwartz J (2008). Metabolic syndrome and inflammatory responses to
long-term particulate air pollutants. Environ Health Perspect. 116:612-7.
14. Chen H; Goldberg MS; Burnett RT; Jerrett M; Wheeler AJ; Villeneuve PJ (2013).
Long-Term Exposure to Traffic-Related Air Pollution and Cardiovascular Mortality.
Epidemiology, 24(1):35-42. doi: 10.1097/EDE.0b013e318276c005.
15. Cheng H, Saffari A, Sioutas C, Forman HJ, Morgan TE, Finch CE. Nano-Scale
Particulate Matter from Urban Traffic Rapidly Induces Oxidative Stress and
Inflammation in Olfactory Epithelium with Concomitant Effects on Brain. Environ
Health Perspect. 2016 May 17. [Epub ahead of print] PubMed PMID: 27187980.
16. Choi HS, Ashitate Y, Lee JH, et al (2010). Rapid translocation of nanoparticles from
the lung airspaces to the body. Nat Biotechnol. 28:1300-3.
17. Daher N, Hasheminassab S, Shafer MM (2013). Seasonal and spatial variability in
chemical composition and mass closure of ambient ultrafine particles in the megacity
of Los Angeles. Environ Sci: Process Impacts 15: 283-295.
18. Davis DA, Bortolato M, Godar SC, et al (2013). Prenatal Exposure to Urban Air
Nanoparticles in Mice Causes Altered Neuronal Differentiation and Depression-Like
Responses. PLoS ONE 8: e64128.
19. Davis DA, Akopian G, Walsh JP, Sioutas C, Morgan TE, Finch CE. Urban air
pollutants reduce synaptic function of CA1 neurons via an NMDA/NȮ pathway in
vitro. J Neurochem. 2013 Nov;127(4):509-19. doi: 10.1111/jnc.12395.
20. Delfino RJ, Staimer N, Tjoa T, Gillen DL, Polidori A, Arhami M, Kleinman MT, Vaziri
ND, Longhurst J, Sioutas C. Air pollution exposures and circulating biomarkers of
effect in a susceptible population: clues to potential causal component mixtures and
mechanisms. Environ Health Perspect. 2009 Aug;117(8):1232-8. doi:
10.1289/ehp.0800194.
21. Demokritou P, Gupta T, Ferguson S, et al (2003). Development of a high-volume
concentrated ambient particle system (CAPS) for human and animal inhalation
toxicological studies. Inhal Toxicol. 15:111-29
22. Dunlop BW, Nemeroff CB (2007). The role of dopamine in the pathophysiology of
depression. Arch. Gen. Psychiatry. 64:327–337.
23. Environmental Protection Agency (2013). Federal Register. Rules and Regulations.
Vol. 78, No10
24. Fujimoto A, Tsukue N, Watanabe M, et al (2005). Diesel exhaust affects
immunological action in the placentas of mice. Environ Toxicol. 20:431-40.
25. Goldberg MS, Burnett RT, Bailar JC, Tamblyn R, Ernst P, Flegel K, Brook J,
Bonvalot Y, Singh R, Valois MF, Vincent R (2001). Identification of persons with
cardiorespiratory conditions who are at risk of dying from the acute effects of
ambient air particles. Environ Health Perspect, 109: 487–494. PMC1240570

xliii

26. Guxens M, Aguilera I, Ballester F, et al. (2012). Prenatal exposure to residential air
pollution and infant mental development: Modulation by antioxidants and
detoxification factors. Environ Health Perspect 1201:144-149
27. Hansen CA, Barnett AG, Pritchard G (2008). The effect of ambient air pollution
during early pregnancy on fetal ultrasonic measurements during mid-
pregnancy. Environ Health Perspect. 116:362-9.
28. Hoek G, Brunekreef B, Goldbohm S, et al (2002). Association between mortality and
indicators of traffic-related air pollution in the Netherlands: a cohort study. Lancet,
360:1203–1209.
29. Hougaard KS, Jensen KA, Nordly P, et al (2008). Effects of prenatal exposure to
diesel exhaust particles on postnatal development, behavior, genotoxicity and
inflammation in mice. Part Fibre Toxicol.11;5:3.
30. Hougaard, K. S., Saber, A. T., Jensen, K. A., et al (2009). Diesel exhaust particles:
effects on neurofunction in female mice. Basic Clin Pharmacol Toxicol, 105, 139-
143.
31. Iñiguez C, Ballester F, Estarlich M, et al (2012). Prenatal exposure to traffic-related
air pollution and fetal growth in a cohort of pregnant women. Occup Environ Med.
69:736-44.
32. Janssen, N. A., van Vliet, P. H., Aarts, F., (2001). Assessment of exposure to traffic
related air pollution of children attending schools near motorways. Atmospheric
environment, 35, 3875-3884.
33. Kalkbrenner AE, Daniels JL, Chen JC, et al (2010). Perinatal exposure to hazardous
air pollutants and autism spectrum disorders at age 8. Epidemiology. 21:631-41.
34. Katsouyanni K, Touloumi G, Samoli E (2001). Confounding and effect modification in
the short-term effects of ambient particles on total mortality: results from 29
European cities within the APHEA2 Project. Epidemiology,12: 521–531. Doi:
00001648-200109000-00011
35. Kannan S, Misra DP, Dvonch JT, et al (2006). Exposures to airborne particulate
matter and adverse perinatal outcomes: a biologically plausible mechanistic
framework for exploring potential effect modification by nutrition. Environ Health
Perspect. 114:1636-42.
36. Kim, Y. S., Song, M. Y., Park, J. D., et al (2010). Subchronic oral toxicity of silver
nanoparticles. Particle and fibre toxicology, 7, 20.
37. Levesque S, Taetzsch T, Lull ME, Kodavanti U, Stadler K, Wagner A, Johnson JA,
Duke L, Kodavanti P, Surace MJ, Block ML. Diesel exhaust activates and primes
microglia: air pollution, neuroinflammation, and regulation of dopaminergic
neurotoxicity. Environ Health Perspect. 2011 Aug;119(8):1149-55. doi:
10.1289/ehp.1002986. Epub 2011 May 11. PubMed PMID: 21561831; PubMed
Central PMCID: PMC3237351.
38. Levesque S, Surace MJ, McDonald J, Block ML. Air pollution & the brain:
Subchronic diesel exhaust exposure causes neuroinflammation and elevates early
markers of neurodegenerative disease. J Neuroinflammation. 2011 Aug 24;8:105.
doi: 10.1186/1742-2094-8-105. PubMed PMID: 21864400; PubMed Central PMCID:
PMC3184279.

xliv

39. Li R, Navab M, Pakbin P, Ning Z, Navab K, Hough G, Morgan TE, Finch CE, Araujo
JA, Fogelman AM, Sioutas C, Hsiai T. Ambient ultrafine particles alter lipid
metabolism and HDL anti-oxidant capacity in LDLR-null mice. J Lipid Res. 2013
Jun;54(6):1608-15. doi: 10.1194/jlr.M035014.
40. Logan WP (1953). Mortality in the London fog incident, 1952. Lancet 1:336–338.
41. Marcelis M, Takei N, van Os J (1999). Urbanization and risk for schizophrenia: does
the effect operate before or around the time of illness onset? Psychol Med. 29:1197-
203.
42. Morgan TE, Davis DA, Iwata N, et al (2011). Glutamatergic neurons in rodent
models respond to nanoscale particulate urban air pollutants in vivo and in
vitro. Environ Health Perspect. 119:1003-9.
43. Nel AE, Mädler L, Velegol D, et al (2009). Understanding biophysicochemical
interactions at the nano-bio interface. Nat Mater. 8:543-57.
44. Oberdörster G, Sharp Z, Atudorei V, et al (2004). Translocation of inhaled ultrafine
particles to the brain. Inhalation toxicology. 16:437-45.
45. Onoda A, Umezawa M, Takeda K, et al (2014) Effects of maternal exposure to
ultrafine carbon black on brain perivascular macrophages and surrounding
astrocytes in offspring mice. PLoS ONE 9: e94336.
46. Ornoy, A. (2011). Prenatal origin of obesity and their complications: gestational
diabetes, maternal overweight and the paradoxical effects of fetal growth restriction
and macrosomia. Reprod. Toxicol. 32, 205
47. Park, E. J., Bae, E., Yi, J., et al (2010). Repeated-dose toxicity and inflammatory
responses in mice by oral administration of silver nanoparticles. Environ Toxicol
Pharmacol, 30, 162-168.
48. Perera FP, Rauh V, Tsai WY, et al (2003).Effects of transplacental exposure to
environmental pollutants on birth outcomes in a multiethnic population. Environ
Health Perspect. 111:201-5.
49. Perera FP, Rauh V, Whyatt RM, et al (2006). Effect of prenatal exposure to airborne
polycyclic aromatic hydrocarbons on neurodevelopment in the first 3 years of life
among inner-city children. Environ Health Perspect. 114:1287-92.
50. Perera FP, Tang D, Wang S, et al (2012). Prenatal polycyclic aromatic hydrocarbon
(PAH) exposure and child behavior at age 6-7 years. Environ Health Perspect.
120:921-6.
51. Peters, A; von Klot, S; Heier, M; et al (2004). Exposure to Traffic and the Onset of
Myocardial Infarction. N Engl J Med, 351:1721-1730.
52. Pietroiusti, A., Campagnolo, L., Fadeel, B. (2013). Interactions of engineered
nanoparticles with organs protected by internal biological barriers. Small, 9, 1557-
1572.
53. Pope CA, Burnett RT, Thun MJ, Calle, EE; Krewski, D; Ito, K; Thurston, GD (2002).
Lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate
air pollution. JAMA, 287: 1132–1141. doi:10.1001/jama.287.9.1132.
54. Pope CA, Burnett RT, Thurston GD, Thun, MJ; Calle, EE; Krewski, D; Godleski, JJ
(2004). Cardiovascular mortality and long-term exposure to particulate air pollution:
epidemiological evidence of general pathophysiological pathways of
disease. Circulation, 109: 71–77. doi: 10.1161/01.CIR.0000108927.80044.7F

xlv

55. Rice, D., & Barone Jr, S. (2000). Critical periods of vulnerability for the developing
nervous system: evidence from humans and animal models. Environ Health
Perspec, 108, 511.
56. Ritz B, Yu F, Chapa G, et al (2000). Effect of air pollution on preterm birth among
children born in southern California between 1989 and 1993. Epidemiology; 11:502–
11.
57. Ritz B, Yu F, Fruin S, et al (2002). Ambient air pollution and risk of birth defects in
Southern California. Am J Epidemiol. 155:17-25.
58. Roberts, A. L.; Lyall, K.; Hart, et al (2013). Perinatal air pollutant exposures and
autism spectrum disorder in the children of nurses' health study II participants.
Environ Health Perspec. 121, 978-984.
59. Roorda-Knape, MC; Janssen, NA; De Hartog, JJ; et al (1998). Air pollution from
traffic in city districts near major motorways. Atmos Environ, 32-11: 1921-1930.
60. Saffari A, Daher N, Shafer MM (2014). Seasonal and spatial variation in dithiothreitol
(DTT) activity of quasi-ultrafine particles in the Los Angeles Basin and its association
with chemical species. J Envir Sci Health, Part A: Toxic/Hazardous Subst & Environ
Engin 49: 441-451.
61. Samet, JM; Dominici, F; Curriero, FC; Coursac , I; Zeger, SL (2000). Fine
particulate air pollution and mortality in 20 U.S. cities, 1987–1994. N Engl J
Med,343: 1742–1749. DOI: 10.1056/NEJM200012143432401
62. Schleh, C., Semmler-Behnke, M., Lipka, J., et al (2012). Size and surface charge of
gold nanoparticles determine absorption across intestinal barriers and accumulation
in secondary target organs after oral administration. Nanotoxicology, 6, 36-46.
63. Shah PS, Balkhair T (2011); Knowledge Synthesis Group on Determinants of
Preterm/LBW births. Air pollution and birth outcomes: a systematic review. Environ
Int; 37:498-516.
64. Sood, A., Salih, S., Roh, D., et al (2011). Signalling of DNA damage and cytokines
across cell barriers exposed to nanoparticles depends on barrier thickness. Nature
Nanotechnol. 6, 824-833.
65. Stolp, H. B., Liddelow, S. A., Sá-Pereira, I. (2013). Immune responses at brain
barriers and implications for brain development and neurological function in later life.
Front Integr Neurosci. 23;7:61
66. Sun Q, Wang A, Jin X, et al (2005). Long-term Air Pollution Exposure and
Acceleration of Atherosclerosis and Vascular Inflammation in an Animal Model.
JAMA.; 294:3003-3010.
67. Sun, Q., Yue, P., Deiuliis, J. A., et al (2009). Ambient air pollution exaggerates
adipose inflammation and insulin resistance in a mouse model of diet-induced
obesity. Circulation, 119, 538-546.
68. Suzuki T, Oshio S, Iwata M, et al (2010). In utero exposure to a low concentration of
diesel exhaust affects spontaneous locomotor activity and monoaminergic system in
male mice. Part Fibre Toxicol. 23;7:7.
69. Takeda K, Suzuki K, Ishihara A, et al (2009). Nanoparticles transferred from
pregnant mice to their offspring can damage the genital and cranial nerve
systems. J.Health Sci., 55, 95-102

xlvi

70. Umezawa M, Sakata C, Tanaka N, Tabata M, Takeda K, Ihara T, et al (2011).
Pathological study for the effects of in utero and postnatal exposure to diesel
exhaust on a rat endometriosis model. J Toxicol Sci;36:493–8.
71. Utell MJ; Frampton MW (2000). Acute Health Effects of Ambient Air Pollution: The
Ultrafine Particle Hypothesis. J of Aerosol Med. 13: 355-359.
72. van den Hooven EH, Pierik FH, de Kluizenaar Y, et al (2012). Air pollution exposure
during pregnancy, ultrasound measures of fetal growth, and adverse birth outcomes:
a prospective cohort study. Environ Health Perspect. 120:150-6.
73. Veronesi B, Makwana O, Pooler M, Chen LC. Effects of subchronic exposures to
concentrated ambient particles. VII. Degeneration of dopaminergic neurons in Apo
E-/- mice. Inhal Toxicol. 2005 Apr;17(4-5):235-41. PubMed PMID: 15804941.
74. Volk HE, Hertz-Picciotto I, Delwiche L, et al (2011). Residential proximity to freeways
and autism in the CHARGE study. Environ Health Perspect. 119:873-7.
75. Volk HE, Lurmann F, Penfold B, et al (2013). Traffic-related air pollution, particulate
matter, and autism. JAMA Psych.70:71-7.
76. Wei Y, Zhang JJ, Li Z, Gow A, Chung KF, Hu M, Sun Z, Zeng L, Zhu T, Jia G, et
al.Chronic exposure to air pollution particles increases the risk of obesity and
metabolic syndrome: findings from a natural experiment in Beijing. FASEB J. 2016
Jun;30(6):2115-22. doi: 10.1096/fj.201500142. Epub 2016 Feb 18. PubMed PMID:
26891735Wick P, Malek A, Manser P, et al (2010). Barrier capacity of human
placenta for nanosized materials. Environ Health Perspect. 118:432-6.
77. Wilhelm M, Ghosh JK, Su J, et al (2012). Traffic-related air toxics and term low birth
weight in Los Angeles County, California. Environ Health Perspect. 120:132-8.
78. Win-Shwe, Yamamoto S, Ahmed S, Kakeyama M, Kobayashi T, Fujimaki H. Brain
cytokine and chemokine mRNA expression in mice induced by intranasal instillation
with ultrafine carbon black. Toxicol Lett. 2006 May 25;163(2):153-60. Epub 2005
Nov 15. PubMed PMID: 16293374.
79. Wu J, Wang C, Sun J, et al (2011). Neurotoxicity of silica nanoparticles: brain
localization and dopaminergic neurons damage pathways. ACS Nano. 28;5:4476-89.
80. Yang TT, Simmons AN, Matthews SC, et al (2010). Adolescents with major
depression demonstrate increased amygdala activation. J Am Acad Child Adolesc
Psychiatry. 49:42-51.
81. Yokota S, Moriya N, Iwata M, et al (2013). Exposure to diesel exhaust during fetal
period affects behavior and neurotransmitters in male offspring mice. J Toxicol
Sci.38:13-23
82. Zhang, Q; Stanier, C; Canagaratna, M; et al (2004). Insights into the chemistry of
new particle formation and growth events in Pittsburgh based on aerosol mass
spectrometry. Environ Sci Tech, 38: 4797-4809.
83. Zhou Y, Levy JI (2007). Factors influencing the spatial extent of mobile source air
pollution impacts: a meta-analysis. BMC Public Health, 7:89.
84. Zou, J. Y., & Crews, F. T. (2005). TNFα potentiates glutamate neurotoxicity by
inhibiting glutamate uptake in organotypic brain slice cultures: neuroprotection by
NFκB inhibition. Brain research, 1034, 11-24.

47

Chapter 1

Traffic-related air pollution accelerates myelin and neuritic aging changes in
a mouse model, with specificity for CA1 neurons.

Woodward NC
1*
, Pakbin P
2
, Sioutas C
2
, Cacciottolo M
1
, Morgan TE
1
, Finch CE
1,3
.

1 Leonard Davis School of Gerontology, 2 Viterbi School of Engineering, 3 Dornsife College
University of Southern California, 3715 McClintock Ave Los Angeles, California, USA
*Corresponding author. Email nwoodwar@usc.edu.

Abstract:
Traffic-related air pollution (TRAP) is associated with lower cognition and
reduced total white matter volume in older adults, specifically for airborne particulate
matter <2.5 µm diameter (PM2.5). Young rodents exposed to TRAP show microglial
activation and neuronal atrophy. Older mice potentially show microglial activation,
therefore we further investigated age differences in the impact of TRAP on neurites,
white matter, and microglial activation. Young and older mice (3 and 18 month, female
C57BL/6J) were intermittently exposed during 10 weeks to nano-particulate matter
(nPM, <0.2 µm diameter), a nanoscale subset of urban PM2.5. Young exposed mice

48

showed reduction of neurites and white matter, with microglial activation in the CA1
hippocampal region, but not in the dentate gyrus. These changes approximate baseline
aging changes in older controls, and suggest that TRAP exposure in young can cause
neurodegenerative changes resembling older age. Notably, the CA1 selectivity models
the CA1 vulnerability in Alzheimer’s disease. The older exposed mice had minimal
responses to nPM suggesting a ceiling effect with age.
Keywords: aging, white matter, air pollution, particulate matter, hippocampus, CA1

1. Introduction:
Traffic-related air pollution (TRAP) is a ubiquitous environmental toxin, which is
associated with poorer cognitive performance in older populations (Power et al., 2011,
Ranft et al., 2009, Wellenius et al., 2012, Zeng et al., 2010). The lower cognitive
functioning approximates a two-year advance of normal cognitive decline from aging
(Ailshire et al., 2014a, Ailshire et al., 2014b, Weuve et al., 2012). Brain structural and
cellular changes are less documented. Small decreases of white matter (WM) volume
were detected by MRI in older U.S. women of the Women’s Health Initiative Memory
Study (WHIMS) cohort who resided in zones of fine particulate matter (PM2.5), with dose
exposure in proportion to quartile PM2.5 (Chen et al., 2015). Histochemical studies
further showed WM microglial activation and blood brain barrier leakage in a small
postmortem sample of young adults from a highly polluted Mexican city (Calderon et al.,
2008, Calderon et al., 2011). In several rodent models, TRAP exposure activated
microglia (Block et al., 2007, Block and Calderon, 2009, Cheng et al., 2016a). Effects of
aging in animal models of TRAP have received limited study and show divergent age

49

responses. Mumaw et al, 2016, reported increased cerebral cortex microglial reactivity
with aging in male C57BL/6J mice (‘B6’) exposed to TRAP model in vivo. However, in
our prior study aging B6 male mice did not induce detoxifying genes in the cerebellum
(Zhang et al., 2012).
In view of the potential importance of white matter loss from TRAP exposure in
the WHIMS cohort (Chen et al., 2015), we examined the impact of aging on responses
to TRAP in white matter and microglia of the hippocampus in female mice. Neuronal
responses were included, because PM2.5 exposure of C57BL/6J male mice caused
selective atrophy of hippocampal CA1 pyramidal neurons (Fonken et al., 2011). The
selectivity of CA1 neurons to TRAP is relevant to cognitive loss in association with
PM2.5, because the CA1 pyramidal neurons are the most vulnerable to Alzheimer’s
disease (AD) (Padurariu et al., 2012, Serrano-Pozo et al., 2011).
The present study addresses interactions of age and TRAP by chronically
exposing young (3 month) and older (18 month) mice to nPM a nano-size subfraction of
ambient PM2.5, which has greater cytotoxicity in vivo and in vitro than larger PM
(Gillespie et al., 2013, Li et al., 2003). In neonatal neuron cultures nPM inhibited neurite
outgrowth, caused growth cone collapse, and impaired pyramidal cell differentiation
(Davis et al., 2013). In vitro, nPM also activates microglia (Cheng et al., 2016a, Cheng
et al., 2016b). The age of 21 months at tissue collection demographically approximates
a paramenopausal group a decade or more earlier than in the WHIMS cohort. This age
also minimizes pathological confounds from tumors and other senescent organ
pathology which emerge after 22-24 months (Finch et al. 1969; Finch and Foster, 1971)
in association with exponential increasing mortality (Finch and Pike, 1996).

50

We predicted that older mice would be less sensitive to nPM because older male
rats had decreased excitotoxic damage to CA1 neurons from kainic acid (Kesslak et al.,
1995); fewer glutamatergic receptors in CA1 subfields and other hippocampal subfields
in mice (Magnusson and Cotman, 1993), and in rat CA1 neurons, smaller kainic
induced EPSP (Kerr et al., 2002). In vitro, nPM shows glutamatergic pathway
involvement with greater CA1 vulnerability and attenuation of nPM effects by AP5, the
NMDA receptor antagonist (Davis et al., 2013b).

2. Methods:
2.1 Animals and Ethics Statement:
C57BL/6J female mice, 3 and 18 mo were obtained from the NIA Aging Mouse
Colony. Protocols were approved by the University of Southern California Institutional
Animal Care and Use Committee, and animals were maintained under standard
conditions according to NIH guidelines.

2.2 nPM Collection and Exposure:
The nPM utilized in these studies are a subfraction of ultrafine PM (<200 nm diameter)
collected from urban air in Los Angeles, California, near the CA-110 Freeway (Misra et
al., 2002). These aerosols represent a mix of fresh ambient PM mostly from vehicular
traffic on this freeway (Ning et al., 2007). The nPM was collected on Teflon filters
resuspended in deionized water by vortexing and sonication (Morgan et al., 2011). The
nPM comprise approximately 20% by mass of ambient PM2.5 in that location (Sardar et
al., 2005) Water soluble metals and organic compounds were efficiently transferred

51

from the filter collection medium into aqueous suspension used for exposures (Morgan
et al., 2011). Relative to the total filtered-trapped ultrafines (PM0.2), the nPM subfraction
eluted into aqueous phases is depleted of black carbon and water insoluble organic
compounds. The chemical composition of nPM used in these exposures is shown in
Table S1 of the supplement. The nPM is resuspended to 340 µg/ml, with trace
endotoxin levels (measured by Limulus amebocyte assay); frozen stocks at 20 °C retain
chemical stability for >30 days, including long-lived free radical species (Morgan et al.,
2011, Li et al., 2003). Mice were exposed five hours a day, three days a week, for ten
weeks. Experimental timeline is shown in Fig.1.
2.3 Weight and Behavior:
Weight: Weight was taken prior to exposure, periodically throughout exposure,
and before tissue collection. Two-way repeated measures ANOVA and Bonferroni post-
hoc test was used to determine differences during exposure, and one-way ANOVA with
Tukey post-hoc test was used to detect differences at the end of exposure.
Novel Object Recognition (NOR): Short and long term memory was assessed
with the NOR test. Mice were tested on a three-day protocol, to assess short and long
term memory, and exploratory behavior. On day one mice were individually acclimatized
to a dimly-lit black Plexiglas cubic box (20 x 20 x 20cm) for 15 min. After 24 h mice
were placed back into the box, and exposed to two identical novel objects (3.5 x 8 cm),
which were affixed to the floor and placed symmetrically at 6 cm from the nearest walls.
Mice were placed in a corner, facing the center and at equal distances from the two
objects. Their start position was rotated and counterbalanced throughout the test.

52

Exploration, defined as sniffing or touching, of the two objects was recorded. Sitting on
the object was not considered exploration. Ninety minutes after the first trial, one object
was replaced, and the procedure was repeated; 24 h later, the novel object was
replaced with a second novel object, and the trial was repeated to assess long term
memory. The novelty exploration index was calculated by time spent exploring the novel
object, divided by time spent exploring the previous object. Statistical analysis for all
tests used ANOVA, with Tukey’s post-hoc test.
Spontaneous Alternation of Behavior (SAB): Working memory was assessed by
the spontaneous alternation of behavior (SAB) test. The apparatus consisted of three
equivalent arms (15 x 8 x 10 cm) made of black Plexiglas with equal angles between all
arms. Mice were individually placed in one arm and allowed to freely explore the
apparatus for 10 min. The sequence and entries in each arm were recorded and
percent alternation was determined from consecutive entries to the three different arms
over the total number of transitions.

2.4 Histochemistry:
Immunofluorescence: Following saline perfusion, brain hemispheres were
immersed in 4% paraformaldehyde overnight; cryoprotected in 30% sucrose; embedded
in Optimal Cutting Temperature medium and sliced sagittally in 18 µm thick sections on
a cryostat. Sections were stored at -80 °C. Tissue sections were permeabilized with 1%
NP-40 and blocked with 5% bovine serum albumin. Primary antibodies to Iba1 (ionized
calcium binding adaptor molecule 1 (1:500, 019-19741, Wako Pure Chemical Industries,

53

AB839504) or MBP (myelin basic protein, 1:1000, ab40390, Abcam, AB1141521) were
added overnight at 4 °C. Immunofluorescence was visualized by Alex Fluor 488 and
594 antibodies (1:400, goat, Molecular Probes).
Amino-Cupric-Silver Stain: Slides were defrosted, incubated in 20% silver nitrate
for 15 min, followed by 20% silver nitrate for 15 min, before developing with a solution of
formaldehyde, citric acid, nitric acid, and ammonium hydroxide (de Olmos et al., 1994).
Slides were dehydrated, coverslipped with Permount (SP15-500, Fisher Scientific), and
visualized on a Leica brightfield microscope.
Analysis: Using Image J, images were thresholded and quantified for total
integrated density. Silver-stained images were analyzed to resolve cell bodies and
processes. Results were normalized to the average of 3 mo controls. Statistical analysis
used ANOVA, with Tukey’s post-hoc test.

3. Results:
3.1 Histochemistry
The hippocampus and corpus callosum were examined for neuronal
morphological changes, white matter myelin basic protein (MBP), and microglial
activation; illustrative images, Figure 2. Baseline age changes are summarized in Table
1; nPM responses of the young, Table 2.
Neurites: Silver staining showed effects of age and nPM on total neurites, with
no change in perikarya. In controls, 18 mo vs 3 mo CA1 neurites showed 25% reduced
area (Figure 3A, B, ANOVA p<0.05). Exposure to nPM did not further decrease 18 mo

54

neurite area (Figure 3A, B; ANOVA p<0.05). After nPM exposure, young mice had 25%
fewer neuritic processes in the CA1 stratum oriens and stratum radiatum (Figure 3A, B,
ANOVA p<0.05). Dentate gyrus neurites did not change by age or nPM exposure
(Figure 3C, D). Perikaryal staining did not change with age or exposure in any region
analyzed (Supplementary Figure 2).
White Matter: Myelin basic protein (MBP) was decreased in older mice in the
stratum oriens of the CA1, and the polymorphic layer of the dentate gyrus (Figure 4A, C,
ANOVA p<0.05). nPM exposure decreased MBP by 50% in the stratum oriens of the
CA1 for young mice (Figure 4A, ANOVA p<0.05). Exposure did not alter polymorphic or
molecular layers of the dentate gyrus (Figure 4C, D), or in the corpus callosum (Figure
4E). Older nPM exposed mice showed no further decrease in MBP.
Microglial Activation: Iba1 immunostaining, a marker for microglial activation,
showed limited age related increases (Figure 5A, nPM group only, ANOVA p<0.05).
nPM exposure increased Iba1 in young mice by 50% in the stratum oriens of the CA1
(Figure 5A; ANOVA p<0.05), and the polymorphic layer of the dentate gyrus (Figure 5C,
ANOVA p<0.05). The stratum radiatum of the CA1, molecular layer of the dentate gyrus,
and the corpus callosum were unchanged (Figure 5B, D, E).

3.2 Weights and Behavior
Body weight: Weight loss occurred in all groups during the first three weeks of
exposure (two-way ANOVA, p<0.01); presumably due to handling and noise stress.
Young control and exposed mice, as well as older control mice, recovered in weight

55

over the course of the experiment, ending the 10 week exposure at the same weight as
the start. In contrast, older exposed mice did not recover in weight during the exposure
(one-way ANOVA, p<0.05; Figure 6), instead maintaining the initial weight loss. By 4
weeks after cessation of exposure the older exposed mice had recovered in weight.
Young mice, both nPM and control, who were stable in weight throughout exposure,
gained weight after conclusion of the exposure (two-way ANOVA, p<0.0001), potentially
indicating stress from the exposure paradigm.
Cognition and Activity: No memory deficits were observed in SAB or NOR tests
as a result of age or treatment (Supplementary Figures 1A, B, C). For NOR, no changes
were seen discrimination index in either the 90 m or 24 h test, with all groups spending
50% more time exploring the novel object (Supplementary Figure 1A, B). For working
memory, no changes were observed in the SAB test (Figure 1C).
nPM exposure decreased motor activity in both tests. In SAB, the total arm
entries were decreased in both young and older mice, versus age matched controls
(Figure 7A, two-way ANOVA, p<0.05). The NOR showed 30% less exploration for older
mice exposed to nPM, vs controls (Figure 7B, ANOVA p<0.01). Individual weight loss
was correlated with locomotor activity change for older exposed mice (r=0.5076,
Supplementary Figure 3), with heavier animals exploring more.

56

4. Discussion
Young female mice (3 mo) given 10 weeks of intermittent exposure to ambient
nPM from urban traffic emissions showed hippocampal CA1 subregion-specific
decrease of MBP in WM and of neuritic processes, together with microglial activation.
The observed reduction in MBP is the first experimental evidence of WM alteration by
air pollution exposure. These findings give an experimental model for correlations of
WM volume loss in WHIMS cohort (Chen et al., 2016) and of WM microglial activation in
a small postmortem sample from a highly polluted Mexican city (Calderon et al., 2008,
Calderon et al., 2011). The loss of MBP in myelin sheaths could contribute to the
slowing of neuronal connectivity, which is a factor in cognitive performance and a
hallmark of aging (Turken et al., 2008).
Responses to nPM were age specific. Young mice exposed to nPM had reduced
MBP and neurites, and increased microglial activation. The older control (non-exposed)
mice (21 months), had 25% lower CA1 neurite density and 50% less MBP, with a trend
toward microglial activation. However, the older mice did not respond to nPM with
further atrophic changes. Their lack of response to nPM is consistent with the smaller
kainate lesions of aging male rats (Kesslak et al., 1995). These findings closely match
the ‘age-ceiling effect’, which we observed in older male B6 mice (18 mo), wherein nPM
exposure did not induce phase II electrophile responses in cerebellum, as well as in
liver and lung (Zhang et al., 2012). In contrast, Mumaw et al., 2016, reported that older
male mice (18 mo) had greater inflammatory increase of TNFa mRNA and microglial
responses in cerebral cortex in response to another TRAP model (mixed vehicle
emissions).

57

The mechanisms behind the age-ceiling effect to nPM could involve an age-
related loss of glutamate receptors (Magnussen and Cotman, 1993) and age-related
insensitivity to excitotoxins in the CA1 (Kerr et al., 2002), reported for older male rats.
As noted above, nPM shows glutamatergic pathway involvement with greater CA1
vulnerability that is attenuated by the NMDA receptor antagonist AP5 (Davis et al.,
2013b).
The CA1 of young mice showed subregional differences in response to nPM. The
stratum oriens of the CA1 responded most to nPM exposure, with an 25% reduction in
neurites, 50% reduction in white matter, and 50% increase in Iba1. The stratum
radiatum showed changes only in neurites, with white matter and Iba1 remaining
unchanged. These regions, though adjacent, differ in vascularization, cell population,
and connectivity, any of which could explain the divergent responses to nPM. The
stratum oriens is more densely vascularized than the stratum radiatum (Duvernoy et al.,
2013, Grivas et al., 2003) and has different cell composition (Klausberger and Somogyi,
2008), response to LTP (Christie et al., 2000, Ouardouz et al., 1995), and connectivity.
For example, the entorhinal cortex projections to the stratum oriens are denser than to
the stratum radiatum, while CA3 sends projections to both strata, but with more
projections to the radiatum, via the Schaffer collaterals (Figure 2E). The MBP and
microglial changes in response to nPM were observed only in the stratum oriens, which
predicts LTP impairments to nPM in the oriens.
The selective atrophy of CA1 neurites in young B6 female mice confirms the
Golgi findings of Fonken et al., 2011, for young male B6 mice; in both studies the DG
neurites were unchanged. The silver stain used here did not resolve dendritic spines or

58

other neuronal subprocesses. The differential CA1 vulnerability to two models of TRAP
exposure closely matches the CA1 vulnerability in AD, wherein the CA1 neurons
undergo earlier and greater degeneration than the DG (Padurariu et al., 2012, Serrano-
Pozo et al., 2011). This regional vulnerability has important implications for the cognitive
consequences of TRAP exposure. The CA1 is integral in spatial memory (Tsien et al.,
1996), consistent with the poorer performance on the Barnes maze of PM2.5 exposed
mice (Fonken et al., 2011). The CA1 also mediates object recognition, specifically
pattern completion recall, which is based on familiar cues, and is mediated by the
CA3/CA1 pathway (Leal and Yassa, 2015).
In our study, older mice had reduced exploration in the novel object recognition
test (30% reduction), and both young and older exposed mice showed less alternations
in the SAB test (20% reduction). The novel object recognition test measures declarative
memory, but does not specifically resolve pattern completion. Memory tests that
delineate between mechanisms of recall (Fanselow 1990, Matus-Amat et al., 2004)
could be considered in future studies.
The transport of nPM and other TRAP components into the brain is unresolved
and includes at least two routes, ‘nose-to-brain’ and ‘lung-to-brain’. Other ultrafine PM
(radiolabelled carbon and manganese) can be translocated to the brain from the
olfactory neurons in the nasal epithelium into the olfactory bulb, but also to other brain
regions (Oberdorster et al., 2004, Elder et al., 2006), We recently showed rapid
inflammatory responses in the olfactory bulb to inhaled nPM (Cheng et al. 2016).
Additionally, Mumaw et al., 2016, also showed evidence for a lung-to-brain route in
responses to ozone which did not involve TNFa or other cytokines. Nonetheless, TRAP

59

exposure can increase TNFa in humans (Delfino et al., 2009) and mice (Li et al. 2013,
van Eeden et al., 2001). Elucidating the mechanisms behind the response to nPM and
other TRAP components is also complicated by chemical heterogeneity (Morgan et al.,
2011; Liu et al., 2016). Systemic effects of inhaled TRAP are also consistent with the
broad brain regional responses to TRAP inhalation, which include cerebellum (Cheng et
al., 2016; Zhang et al., 2012) and other regions that are multiple synapses away from
olfactory input. Radiolabeled ultrafine PM also rapidly reached the cerebellum at about
the same time as the olfactory bulb (Oberdorster et al., 2004).
The cellular mechanisms of nPM induced neurodegeneration could be mediated
by chronic microglial activation, which produces both extracellular reactive oxygen
species and neurotoxic factors (Block et al., 2007, Mumaw et al., 2016, Davis et al.,
2013b). LPS based microglial activation shows neuronal loss in microglial rich brain
regions (Qin et al., 2007), and causes reduced neuronal processes in the CA1
(Richwine et al., 2008). Exposure to nPM shows neuroinflammatory effects including
induced inflammatory cytokines seen in the cerebral cortex (Morgan et al., 2011), and
increased phase II response genes in the cerebellum (Zhang et al., 2012). nPM
treatment of mixed glial cultures increased IL-1a and TNFa with dose dependence
(Morgan et al., 2011). The current study extends these results of neuroinflammation to
document microglial activation in the CA1 and dentate gyrus, and subregional specificity
of the nPM response.
Lastly we note metabolic effects that differed by age. Mice experienced acute
stress induced weight loss at the beginning of the exposure. All young mice and older
control mice steadily recovered weight throughout the exposure, ending at comparable

60

weights to the start of exposure. Older exposed mice did not recover weight during the
exposure, only recovering after cessation of exposure. Moreover, individual weight loss
was correlated with locomotor activity change for older exposed mice, with heavier
animals exploring more. We suggest that the weight loss may be due to systemic
inflammatory responses, as reflected in the increased blood TNFa in mice (Li et al.,
2013) and in humans (Delfino et al., 2009, van Eeden et al., 2001). Older adults also
show increased leptin levels (Wang et al., 2014) and insulin resistance (Bolton et al.,
2012, Sun et al., 2010) in correlation with exposure. We suggest that cognitive
responses to air pollution will involve systemic interactions with metabolism, including
downstream effects from the lung-brain axis (Mumaw et al., 2016), as well as more
direct neurotoxic effects of the complex components of air pollution.

5. Conclusion:
Shown here, nPM leads to declines in neurites and white matter MBP, with
regional specificity to the CA1, with DG insensitive. These declines may result from
increased microglial activation caused by nPM exposure. This predicts a potential
degradation of CA1 functions in human populations exposed to high levels of air
pollution, which has high relevance for accelerated age impairments of spatial memory
and pattern completion recall. Future imaging studies of WHIMS and other well defined
cohorts may resolve earlier stages of neurodegenerative responses to TRAP and their
relationship to AD risk.

61

Acknowledgements: We thank Amin Haghani for his technical support; Prof. Joshua
Millstein for statistical advice; and Prof. Jiu Chiuan Chen for his careful review of the
manuscript.

Funding: This work was supported by the National Institutes of Aging (T32AG0037, R21
AG-040753, R21 AG-050201)

Bibliography:

1. Ailshire JA, Clarke P. Fine particulate matter air pollution and cognitive function
among U.S. older adults. J Gerontol B Psychol Sci Soc Sci. 2015 Mar;70(2):322-
8. doi: 10.1093/geronb/gbu064.
2. Ailshire JA, Crimmins EM. Fine particulate matter air pollution and cognitive
function among older US adults. Am J Epidemiol. 2014 Aug 15;180(4):359-66.
doi: 10.1093/aje/kwu155.
3. Block ML, Zecca L, Hong JS. Microglia-mediated neurotoxicity: uncovering the
molecular mechanisms. Nat Rev Neurosci. 2007 Jan;8(1):57-69.
4. Block ML, Calderón-Garcidueñas L. Air pollution: mechanisms of
neuroinflammation and CNS disease. Trends Neurosci. 2009 Sep;32(9):506-16.
doi: 10.1016/j.tins.2009.05.009.
5. Bolton JL, Smith SH, Huff NC, Gilmour MI, Foster WM, Auten RL, Bilbo
SD. Prenatal air pollution exposure induces neuroinflammation and predisposes
offspring to weight gain in adulthood in a sex-specific manner. FASEB J. 2012
Nov;26(11):4743-54. doi: 10.1096/fj.12-210989.
6. Calderón-Garcidueñas L, Solt AC, Henríquez-Roldán C, Torres-Jardón R, Nuse
B, Herritt L, Villarreal-Calderón R, Osnaya N, Stone I, García R, Brooks DM,
González-Maciel A, Reynoso-Robles R, Delgado-Chávez R, Reed W. Long-term
air pollution exposure is associated with neuroinflammation, an altered innate
immune response, disruption of the blood-brain barrier, ultrafine particulate
deposition, and accumulation of amyloid beta-42 and alpha-synuclein in children
and young adults. Toxicol Pathol. 2008 Feb;36(2):289-310. doi:
10.1177/0192623307313011.
7. Calderón-Garcidueñas L, Engle R, Mora-Tiscareño A, Styner M, Gómez-Garza
G, Zhu H, Jewells V, Torres-Jardón R, Romero L, Monroy-Acosta ME, Bryant C,
González-González LO, Medina-Cortina H, D'Angiulli A. Exposure to severe

62

urban air pollution influences cognitive outcomes, brain volume and systemic
inflammation in clinically healthy children. Brain Cogn. 2011 Dec;77(3):345-55.
doi: 10.1016/j.bandc.2011.09.006.
8. Chen JC, Wang X, Wellenius GA, Serre ML, Driscoll I, Casanova R, McArdle JJ,
Manson JE, Chui HC, Espeland MA. Ambient air pollution and neurotoxicity on
brain structure: Evidence from women's health initiative memory study. Ann
Neurol. 2015 Sep;78(3):466-76. doi: 10.1002/ana.24460.
9. Cheng H, Saffari A, Sioutas C, Forman HJ, Morgan TE, Finch CE. Nano-Scale
Particulate Matter from Urban Traffic Rapidly Induces Oxidative Stress and
Inflammation in Olfactory Epithelium with Concomitant Effects on Brain. Environ
Health Perspect. 2016 May 17.
10. Christie BR, Franks KM, Seamans JK, Saga K, Sejnowski TJ. Synaptic plasticity
in morphologically identified CA1 stratum radiatum interneurons and giant
projection cells. Hippocampus. 2000;10(6):673-83.
11. Conrad CD. Chronic Stress-Induced Hippocampal Vulnerability: The
Glucocorticoid Vulnerability Hypothesis. Reviews in the neurosciences.
2008;19(6):395-411.
12. Davis DA, Bortolato M, Godar SC, Sander TK, Iwata N, Pakbin P, Shih JC,
Berhane K, McConnell R, Sioutas C, Finch CE, Morgan TE. Prenatal exposure to
urban air nanoparticles in mice causes altered neuronal differentiation and
depression-like responses. PLoS One. 2013 May 29;8(5):e64128. doi:
10.1371/journal.pone.0064128.
13. Davis DA, Akopian G, Walsh JP, Sioutas C, Morgan TE, Finch CE. Urban air
pollutants reduce synaptic function of CA1 neurons via an NMDA/NȮ pathway in
vitro. J Neurochem. 2013 Nov;127(4):509-19. doi: 10.1111/jnc.12395.
14. de Olmos JS, Beltramino CA, de Olmos de Lorenzo S. Use of an amino-cupric-
silver technique for the detection of early and semiacute neuronal degeneration
caused by neurotoxicants, hypoxia, and physical trauma. Neurotoxicology and
teratology 1994; 16(6): 545-561.
15. Delfino RJ, Staimer N, Tjoa T, Gillen DL, Polidori A, Arhami M, Kleinman MT,
Vaziri ND, Longhurst J, Sioutas C. Air pollution exposures and circulating
biomarkers of effect in a susceptible population: clues to potential causal
component mixtures and mechanisms. Environ Health Perspect. 2009
Aug;117(8):1232-8. doi: 10.1289/ehp.0800194.
16. Driscoll I, Hamilton DA, Petropoulos H, Yeo RA, Brooks WM, Baumgartner RN,
Sutherland RJ. The aging hippocampus: cognitive, biochemical and structural
findings. Cereb Cortex. 2003 Dec;13(12):1344-51.
17. Duvernoy HM, Cattin F, Risold P. The Human Hippocampus- Functional
Anatomy, Vascularization, and Serial Sections with MRI. Springer. 2013
18. Elder A, Gelein R, Silva V, Feikert T, Opanashuk L, Carter J, Potter R, Maynard
A, Ito Y, Finkelstein J, Oberdörster G.Translocation of inhaled ultrafine
manganese oxide particles to the central nervous system. Environ Health
Perspect. 2006 Aug;114(8):1172-8. Erratum in: Environ Health Perspect. 2006
Aug;114(8):1178.
19. Fanselow, M. S. (1990). Factors governing one-trial contextual conditioning.
Animal Learning & Behavior, 18(3), 264–270. doi:10.3758/bf03205285

63

20. Fotenos AF, Snyder AZ, Girton LE, Morris JC, Buckner RL. Normative estimates
of cross-sectional and longitudinal brain volume decline in aging and
AD. Neurology. 2005 Mar 22;64(6):1032-9.
21. Finch CE. The neurobiology of middle-age has arrived. Neurobiol Aging. 2009
Apr;30(4):515-20; discussion 530-33. doi: 10.1016/j.neurobiolaging.2008.11.011.
22. Finch CE, Foster JR. Hematologic and serum electrolyte values of the C57BL-6J
male mouse in maturity and senescence. Lab Anim Sci. 1973 Jun;23(3):339-49.
23. Finch CE, Pike MC. Maximum life span predictions from the Gompertz mortality
model. J Gerontol A Biol Sci Med Sci. 1996; 51(3):B183–B194.
24. Fonken, L. K., Xu, X., Weil, Z. M., Chen, G., Sun, Q., Rajagopalan, S., & Nelson,
R. J. (2011). Air pollution impairs cognition, provokes depressive-like behaviors
and alters hippocampal cytokine expression and morphology. Mol Psychiatry,
16(10), 987–995. doi:10.1038/mp.2011.76
25. Gillespie P, Tajuba J, Lippmann M, Chen LC, Veronesi B. Particulate matter
neurotoxicity in culture is size-dependent. Neurotoxicology. 2013 May;36:112-7.
doi: 10.1016/j.neuro.2011.10.006.
26. Goodrick CL. Life-span and the inheritance of longevity of inbred mice. J
Gerontol. 1975 May;30(3):257-63.
27. Grivas I, Michaloudi H, Batzios Ch, Chiotelli M, Papatheodoropoulos C,
Kostopoulos G, Papadopoulos GC. Vascular network of the rat hippocampus is
not homogeneous along the septotemporal axis. Brain Res. 2003 May
9;971(2):245-9.
28. Hesp BR, Wrightson T, Mullaney I, Kerr DS. Kainate receptor agonists and
antagonists mediate tolerance to kainic acid and reduce high-affinity GTPase
activity in young, but not aged, rat hippocampus. J Neurochem. 2004
Jul;90(1):70-9.
29. Kerr DS, Razak A, Crawford N. Age-related changes in tolerance to the marine
algal excitotoxin domoic acid. Neuropharmacology. 2002 Sep;43(3):357-66.
30. Kesslak JP, Yuan D, Neeper S, Cotman CW. Vulnerability of the hippocampus to
kainate excitotoxicity in the aged, mature and young adult rat. Neurosci Lett.
1995 Mar 24;188(2):117-20.
31. Klausberger T, Somogyi P. Neuronal diversity and temporal dynamics: the unity
of hippocampal circuit operations. Science. 2008 Jul 4;321(5885):53-7. doi:
10.1126/science.1149381.
32. Kunstyr I, Leuenberger HG. Gerontological data of C57BL/6J mice. I. Sex
differences in survival curves. J Gerontol. 1975 Mar;30(2):157-62.
33. Leal, S. L., & Yassa, M. A. (2015). Neurocognitive Aging and the Hippocampus
across Species. Trends in Neurosciences, 38(12), 800–812.
doi:10.1016/j.tins.2015.10.003
34. Li N, Sioutas C, Cho A, Schmitz D, Misra C, Sempf J, Wang M, Oberley T,
Froines J, Nel A. Ultrafine particulate pollutants induce oxidative stress and
mitochondrial damage. Environ Health Perspect. 2003 Apr;111(4):455-60.
35. Li R, Navab M, Pakbin P, Ning Z, Navab K, Hough G, Morgan TE, Finch CE,
Araujo JA, Fogelman AM, Sioutas C, Hsiai T. Ambient ultrafine particles alter
lipid metabolism and HDL anti-oxidant capacity in LDLR-null mice. J Lipid Res.
2013 Jun;54(6):1608-15. doi: 10.1194/jlr.M035014.

64

36. Liu Q, Babadjouni R, Radwanski R, Cheng H, Patel A, Hodis DM, He S,
Baumbacher P, Russin JJ, Morgan TE, Sioutas C, Finch CE, Mack WJ. Stroke
Damage Is Exacerbated by Nano-Size Particulate Matter in a Mouse
Model. PLoS One. 2016 Apr 12;11(4):e0153376. doi:
10.1371/journal.pone.0153376.
37. Magnusson KR, Cotman CW. Age-related changes in excitatory amino acid
receptors in two mouse strains. Neurobiol Aging. 1993 May-Jun;14(3):197-206.
38. Matus-Amat P, Higgins EA, Barrientos RM, Rudy JW. The role of the dorsal
hippocampus in the acquisition and retrieval of context memory
representations. J Neurosci. 2004 Mar 10;24(10):2431-9.
39. Misra C, Kim S, Shen S, Sioutas C. A high flow rate, very low pressure drop
impactor for inertial separation of ultrafine from accumulation mode particles. J
Aerosol Sci. 2002; 33:735-752
40. Morgan TE, Davis DA, Iwata N, Tanner JA, Snyder D, Ning Z, Kam W, Hsu YT,
Winkler JW, Chen JC, Petasis NA, Baudry M, Sioutas C, Finch
CE. Glutamatergic neurons in rodent models respond to nanoscale particulate
urban air pollutants in vivo and in vitro. Environ Health Perspect. 2011
Jul;119(7):1003-9. doi: 10.1289/ehp.1002973.
41. Ning Z, Geller MD, Moore KF, Sheesley R, Schauer JJ, Sioutas C. Daily variation
in chemical characteristics of urban ultrafine aerosols and inference of their
sources. Environ Sci Technol. 2007 Sep 1;41(17):6000-6.
42. Oberdörster G, Sharp Z, Atudorei V. Translocation of inhaled ultrafine particles to
the brain. Inhal Toxicol. 2004; 16: 437-445.
43. Padurariu M, Ciobica A, Mavroudis I, Fotiou D, Baloyannis S. Hippocampal
neuronal loss in the CA1 and CA3 areas of Alzheimer's disease
patients. Psychiatr Danub. 2012 Jun;24(2):152-8.
44. Power MC, Weisskopf MG, Alexeeff SE, Coull BA, Spiro A 3rd, Schwartz
J. Traffic-related air pollution and cognitive function in a cohort of older
men. Environ Health Perspect. 2011 May;119(5):682-7. doi:
10.1289/ehp.1002767.
45. Qin L, Wu X, Block ML, Liu Y, Breese GR, Hong JS, Knapp DJ, Crews
FT. Systemic LPS causes chronic neuroinflammation and progressive
neurodegeneration. Glia. 2007 Apr 1;55(5):453-62.
46. Ranft U, Schikowski T, Sugiri D, Krutmann J, Krämer U. Long-term exposure to
traffic-related particulate matter impairs cognitive function in the elderly. Environ
Res. 2009 Nov;109(8):1004-11. doi: 10.1016/j.envres.2009.08.003.
47. Richwine AF, Parkin AO, Buchanan JB, Chen J, Markham JA, Juraska JM,
Johnson RW. Architectural changes to CA1 pyramidal neurons in adult and aged
mice after peripheral immune stimulation. Psychoneuroendocrinology. 2008
Nov;33(10):1369-77. doi: 10.1016/j.psyneuen.2008.08.003.
48. Serrano-Pozo A, Frosch MP, Masliah E, Hyman BT. Neuropathological
Alterations in Alzheimer Disease. Cold Spring Harbor Perspectives in
Medicine:2011;1(1):a006189. doi:10.1101/cshperspect.a006189.
49. Sun Q, Yue P, Deiuliis JA, Lumeng CN, Kampfrath T, Mikolaj MB, Cai Y,
Ostrowski MC, Lu B, Parthasarathy S, Brook RD, Moffatt-Bruce SD, Chen LC,
Rajagopalan S. Ambient air pollution exaggerates adipose inflammation and

65

insulin resistance in a mouse model of diet-induced obesity. Circulation. 2009
Feb 3;119(4):538-46. doi: 10.1161/CIRCULATIONAHA.108.799015.
50. Tsien JZ, Huerta PT, Tonegawa S. The essential role of hippocampal CA1
NMDA receptor-dependent synaptic plasticity in spatial memory. Cell. 1996 Dec
27;87(7):1327-38.
51. Turken A, Whitfield-Gabrieli S, Bammer R, Baldo JV, Dronkers NF, Gabrieli
JD. Cognitive processing speed and the structure of white matter pathways:
convergent evidence from normal variation and lesion studies. Neuroimage. 2008
Aug 15;42(2):1032-44. doi: 10.1016/j.neuroimage.2008.03.057.
52. Uz T, Pesold C, Longone P, Manev H. Aging-associated up-regulation of
neuronal 5-lipoxygenase expression: putative role in neuronal
vulnerability. FASEB J. 1998 Apr;12(6):439-49.
53. van Eeden SF, Tan WC, Suwa T, Mukae H, Terashima T, Fujii T, Qui D, Vincent
R, Hogg JC. Cytokines involved in the systemic inflammatory response induced
by exposure to particulate matter air pollutants (PM(10)). Am J Respir Crit Care
Med. 2001 Sep 1;164(5):826-30.
54. Wang Y, Eliot MN, Kuchel GA, Schwartz J, Coull BA, Mittleman MA, Lipsitz LA,
Wellenius GA. Long-term exposure to ambient air pollution and serum leptin in
older adults: results from the MOBILIZE Boston study. J Occup Environ Med.
2014 Sep;56(9):e73-7. doi: 10.1097/JOM.0000000000000253.
55. Wellenius GA, Boyle LD, Coull BA, Milberg WP, Gryparis A, Schwartz J,
Mittleman MA, Lipsitz LA. Residential proximity to nearest major roadway and
cognitive function in community-dwelling seniors: results from the MOBILIZE
Boston Study. J Am Geriatr Soc. 2012 Nov;60(11):2075-80. doi: 10.1111/j.1532-
5415.2012.04195.x.
56. Weuve J, Puett RC, Schwartz J, Yanosky JD, Laden F, Grodstein F. Exposure to
particulate air pollution and cognitive decline in older women. Arch Intern Med.
2012 Feb 13;172(3):219-27. doi: 10.1001/archinternmed.2011.683.
57. Wilson DW, Aung HH, Lame MW, Plummer L, Pinkerton KE, Ham W, Kleeman
M, Norris JW, Tablin F. Exposure of mice to concentrated ambient particulate
matter results in platelet and systemic cytokine activation. Inhal Toxicol. 2010
Mar;22(4):267-76. doi: 10.3109/08958370903278069.
58. Zeng Y, Gu D, Purser J, Hoenig H, Christakis N. Associations of environmental
factors with elderly health and mortality in China.Am J Public Health. 2010
Feb;100(2):298-305. doi: 10.2105/AJPH.2008.154971.
59. Zhang H, Liu H, Davies KJ, Sioutas C, Finch CE, Morgan TE, Forman HJ. Nrf2-
regulated phase II enzymes are induced by chronic ambient nanoparticle
exposure in young mice with age-related impairments. Free Radic Biol Med.
2012 May 1;52(9):2038-46. doi: 10.1016/j.freeradbiomed.2012.02.042.

66

Tables and Figures:

Figure 1

Figure 1: Experimental exposure schedule. Alternate day intermittent exposure schedule is
expanded for first week, with MWF exposure days indicated in bold. Age of mice at beginning
and end of experiment are labeled below the timeline.
W
Begin
M
End
F
0
3,18 mo
Tissue Collection
Behavior
14 10 5
Weeks
6,21 mo

67

Figure 2

68

Figure 2: Regions analyzed and example stains. A- Cresyl violet stain depicting hippocampus
and corpus callosum. Black box denotes region depicted in boxes B, C and D. Silver stain (B),
Myelin basic protein stain (C), and Iba1 stain (D) of cornus ammonis 1 (CA1) stratum oriens and
stratum radiatum, with edge of corpus callosum on the left. Scale bar is 100 µm for A, B, C, and
D. E- Model of hippocampal connectivity. Weight of lines represents strength of connections
between different subregions. DG-MO, dentate gyrus molecular layer; DG-PO, dentate gyrus
polymorphic layer; EC, entorhinal cortex; MF, mossy fibers; PP, perforant pathway; SC,
Schaeffer collaterals.

Table 1

Age Differences
Region Neurites Myelin Microglial
Activation
CA1

Oriens Dec Dec
~
Radiatum 0 0 0
DG

Polymorph 0 Dec 0
Molecular 0 0 0
Corpus Callosum NA 0 0
Table 1: Summary of age differences in control animals. 0 denotes no change, “Inc” denotes an
increase versus young control, “Dec”, a decrease versus young control, and NA, not measured.
In the cornus ammonis 1 (CA1) stratum oriens, older mice were borderline significant (shown by
~), with p<0.05 in a two tailed t-test. DG-dentate gyrus

69

Table 2

Treatment Differences
Region Neurites Myelin Microglial
Activation
CA1

Oriens Dec Dec Inc
Radiatum Dec 0 0
DG Polymorph 0 0 Inc
Molecular 0 0 0
Corpus Callosum NA 0 0
Table 2: Summary of treatment differences in young animals. Values are versus young control
air mice. 0 denotes no change, “Inc“ denotes increase, “Dec” shows a decrease, and NA means
not measured. CA1- cornus ammonis 1, DG- dentate gyrus.

70

Figure 3

Figure 3: Neurites, visualized by silver staining, in regions of the hippocampus and corpus
callosum (See Fig. 2). A) Stratum oriens of the cornus ammonis 1 (CA1). All groups had
decreased processes versus young control (ANOVA p<0.05), with no further decrease seen by
exposure in older mice. B) Stratum radiatum of the CA1. Young and old treated mice had
decreased processes versus young control (ANOVA p<0.05). C) Polymorphic layer of the
dentate gyrus (DG). D) Molecular layer of the DG. Mean +/- SEM; N=9

B A
C D

71

Figure 4

Figure 4: White matter staining by myelin basic protein IHC, in the hippocampus and corpus
callosum (See Fig. 2). A) Stratum oriens of the cornus ammonis 1 (CA1). All groups had
decreased processes versus young control (ANOVA p<0.05), with no further decrease seen by
exposure in older mice. B) Stratum radiatum of the CA1. C) Polymorphic layer of the dentate
gyrus (DG), with age decreases observed (ANOVA p<0.05). D) Molecular layer of the DG. E)
Forceps major of the corpus callosum. Mean +/- SEM; N=9

CA1- Stratum Oriens
0
50
100
150
3m 18m
*
*
*
% 3m control
CA1- Stratum Radiatum
0
50
100
150
3m 18m
Control
nPM
Values are %
vs 3m control
DG- Polymorph Layer
0
50
100
150
3m 18m
***
**
% 3m control
DG- Molecular Layer
0
50
100
150
3m 18m
Corpus Callosum-
Forceps Major
0
50
100
150
3m 18m
% 3m control A B
C D
E

72

Figure 5

Figure 5: Microglial activation, measured by Iba1 immunohistochemistry (IHC), in regions of the
hippocampus and corpus callosum (See Fig. 2). A) Stratum oriens of the cornus ammonis 1
(CA1). Young exposed mice had increased microglial Iba1 versus control (ANOVA p<0.05).
Older control mice showed a trend toward decrease, with no change observed by ANOVA, but
p<0.05 by two-tailed t-test. B) Stratum radiatum of the CA1. C) Polymorphic layer of the dentate
gyrus (DG), with increased staining in young exposed mice versus control (ANOVA p<0.05). D)
Molecular layer of the DG. Increased Iba1 expression was observed in young mice after nPM
exposure control (ANOVA p<0.05). E) Microglial expression in the corpus callosum, no
significant change. Mean +/- SEM; N=9

A
B
D
C
E

73

Figure 6

Figure 6: Weights throughout exposure, and one month after the end of exposure. Initial weight
loss was observed in all groups (two-way ANOVA, Bonferroni post-hoc test, p<0.01). All young
mice as well as older control mice recovered in weight during the exposure, while older exposed
mice did not, with a significant difference at the end of exposure (one-way ANOVA, Tukey post-
hoc test, p<0.05). Older exposed mice recovered in weight after the end of exposure. Mean +/-
SEM; N=9

Weights
15
20
25
30
3-CTL
3-nPM
18-CTL
18-nPM
*
Weeks
Exposure Behavior
**
Weight (Grams)

74

Figure 7

Figure 7: Activity results from behavioral tests. A) Novel object recognition test; total number of
object explorations in the initial test (one-way ANOVA, p<0.01) B) Spontaneous alternation of
behavior test; total number of alternations between the three arms. Significant difference
between control and nPM animals (two-way ANOVA, p<0.05) Mean +/- SEM; N=9

Novel Object Recognition
Total Activity
0
20
40
60
80
**
Control
nPM
3m 18m
Total Explorations
SAB
0
20
40
60
3m 18m
*
*
Total alterations
A
B

75

Supplementary:
Supplementary Figure 1

Supplementary Figure 1: Results of SAB and NOR memory tests. A) Ratio of novel object
explorations on the 90-minute trial of novel object recognition. Results are potentially
confounded by reduced initial exploration of the object B) Ratio of novel object explorations on
the 24-hour trial of novel object recognition. C) Results of spontaneous alternation of behavior.
Y-axis is the percent of correct arm choices, representative of working memory performance.

A
B
C

76

Supplementary Figure 2

Supplementary Figure 2: Neuronal perikarya, visualized by silver staining, in regions of the
hippocampus and corpus callosum. No changes observed for any region. A) Stratum oriens of
the cornus ammonis 1 (CA1). B) Stratum radiatum of the CA1. C) Polymorphic layer of the
dentate gyrus (DG). D) Molecular layer of the DG.

B A
C D

77

Supplementary Figure 3
Supplementary Figure 3: Correlation analysis of weights at end of exposure, and total number of
explorations in the initial test of novel object recognition (NOR). Significant correlation in both
treatment groups for weight and total exploration.

Supplementary Table 1.

Chemical Composition

Elements/ Species ng/µg PM
Total Organic Carbon (TOC) 586.3 ± 52.2
S 48.31±2.17
Na 31.91±2.07
Ca 27.4±1.04
K 9.29±0.76
Mg 4.4±0.31
Weights:Activity NOR
15 20 25 30 35
0
20
40
60
80
100
3c
3n
18c
18n
*
*
Weights
Total Explorations

78

Zn 2.59±0.08
Cu 1.11±0.04
Ba 0.84±0.07
Al 0.52±0.03
P 0.5±0.04
Sr 0.28±0.02
Fe 0.27±0.01
Mn 0.27±0.02
Sb 0.13±0.01
V 0.11±0.01
Ni 0.076±0.004
Mo 0.047±0.004
Li 0.043±0.003
Pb 0.025±0.001
As 0.025±0.002
Sn 0.023±0.001
Cr 0.0175±0.0012
Rb 0.0102±0.0007
Cd 0.0078±0.0003
Eu 0.0038±0.0003
Pd 0.0024±0.0002
Ti 0.0019±0.0001
Cs 0.0008±0.0001
Tl 0.00075±0.00003

79

Supplementary Table 1: Physical and Chemical Characteristics of nPM used in the exposures.
Average exposure mass concentration: 342 ± 81.2 µg/m
3
. Average exposure number
concentration: 1.4×10
5
± 9.7×10
3
particles /cm
3

80

Chapter 2

TLR4 plays an integral role in glial activation by nanoscale particulate
matter- insights from microarray analysis.

Abstract

Traffic related air pollution (TRAP) exposure is linked with poorer cognition. It is
hypothesized to be a consequence of the neuroinflammatory effects of TRAP exposure,
however, the mechanisms by which TRAP elicits neuroinflammation remain unclear.
This experiment examines the role of nanoscale particulate matter (nPM, diameter <0.2
um) in glial activation. Utilizing an in vitro model of microglia and astrocyte primary
cultures, cells are treated with nPM, with LPS as a model endotoxin response, and
analyzed by microarray. Utilizing Whole Genome Correlation Network Analysis
(WGCNA) and Transcription Factor Target (TFT) element enrichment analysis, we
demonstrate strong activation of the NF-kB and Interferon pathways. Guided by these
findings, we knockdown TLR4 to further elucidate the pathway activated by nPM. TLR4,
as activated by nPM treatment, acts in a MyD88 dependent pathway, activating the NF-
kB and JAK/STAT pathways. TLR4 activation by nPM does not elicit classic endotoxin
activation of TRIF and interferon beta (IFN-b). These results provide valuable insight

81

into glial activation by nPM, and will serve as a tool to future inquiries into cognitive
effects of air pollution.

1. Introduction

Ambient particulate matter, a ubiquitous environmental toxin largely derived from
combustion engines, is associated with poorer cognitive outcomes in both adult (Power
et al., 2011, Ranft et al., 2009, Wellenius et al., 2012, Zeng et al., 2010) and
developmental exposures (Guxens 2012, Perera 2006, Perera 2012). Though cognitive
decline from TRAP exposure has been extended to animal models (Allen 2014, Bolton
2012, Fonken 2011), the pathways through which nPM evokes cognitive impairment are
not yet fully understood.
Exposure to ambient particulate matter (PM), is linked with chronic
neuroinflammation. In a small postmortem sample from Mexico City, high PM exposure
correlated with microglial activation and increased inflammatory markers in the brain
(Calderon 2004, Calderon 2008). Animal studies extend these findings, showing
increased astrocytic activation (Kleinman 2008), increased microglial activation (Cheng
2016, Morgan 2011), and increases in brain IL-1a, TNFa, and NF-kB (Campbell 2005,
Fonken 2011, Levesque 2011, Morgan 2011). It is hypothesized that this
neuroinflammation, and its concomitant oxidative stress, is the source of the cognitive
effects of air pollution exposure (Block 2009). Indeed, ApoE knockout mice, which are
characterized by higher levels of oxidative stress, show dopaminergic neuron loss in the
substantia nigra after exposure to PM, while wildtype mice did not (Veronesi 2005).

82

Direct treatment of PM in vitro induces microglial activation (Block 2004, Cheng
2016) and TNFa production from both astrocytes and microglia (Cheng 2016). Microglial
enriched neuronal cultures show dopaminergic neurotoxicity from PM treatment, with
neuron only cultures showing no neurotoxicity (Block 2004). Conditioned media from
astrocyte, microglial, or mixed glial cultures treated with nanoscale PM (nPM, diameter
<0.2 um) all inhibit neurite outgrowth when added to neuronal cultures (Cheng 2016).
Activated microglia release numerous inflammatory cytokines, as well as reactive
oxygen species (ROS) (Lull 2010). When chronically activated this can lead to neuronal
atrophy and death (Block 2007).
PM is broken down into coarse (PM10, diameter <10 um), fine (PM2.5, diameter
<2.5 um), and ultrafine or nanoscale (PM0.1, diameter <0.1 um; nPM, diameter <0.2 um,
respectively). Though PM2.5 (diameter <2.5 um) is the most commonly measured and is
currently regulated by the EPA, it is the nanoscale PM (nPM diameter <0.2 um) that is
believed to be the most harmful to protected organs like the brain (Block 2009, Peters
2006). Nano-sized particles can physically translocate to the olfactory bulb via the
olfactory epithelium (Oberdorster 2004, Elder 2006). Acute nPM inhalation causes
microglial activation and oxidative stress in both the olfactory epithelium and olfactory
bulb, with extended exposures inducing TNFa in the cerebral cortex and cerebellum
(Cheng 2016b).
Though the ability of nPM to cause neuroinflammation is well supported, the
pathway through which nPM induces glial activation is still unknown. The current study
addresses this knowledge gap by treating mixed glial cultures (astrocytes and microglia)

83

with nPM and doing whole genome microarray analysis. LPS treatment is used as a
reference of a well characterized endotoxin response.
Whole genome correlation network analysis (WGCNA) is used to identify novel
gene modules responding to nPM stimulation, which can be further analyzed for
transcription factor target (TFT) enrichment. Showing TFT enrichment for NF-kB and
Interferon-y (IFN-y), mixed glial cultures were then treated with toll like receptor 4
(TLR4) siRNA to knock down endogenous TLR4, a receptor responsible for both NF-kB
and IFN-y activation. The effect of silenced TLR4 on cell viability and pathway activation
in response to nPM treatment was then analyzed.

2. Materials and Methods:

2.1 nPM Collection of nano-scale particulate material (nPM)
The nPM utilized in these studies are a subfraction of ultrafine PM (<200 nm diameter)
collected from urban air in Los Angeles, California, near the CA-110 Freeway (Misra et
al., 2002). These aerosols represent a mix of fresh ambient PM mostly from vehicular
traffic on this freeway (Ning et al., 2007). The nPM was collected on Teflon filters
resuspended in deionized water by vortexing and sonication (Morgan et al., 2011). The
nPM comprise approximately 20% by mass of ambient PM2.5 in that location (Sardar et
al., 2005) Water soluble metals and organic compounds were efficiently transferred
from the filter collection medium into aqueous suspension used for exposures (Morgan
et al., 2011). Relative to the total filtered-trapped ultrafines (PM0.2), the nPM subfraction
eluted into aqueous phases is depleted in black carbon and water insoluble organic
compounds. The chemical composition of nPM used in these exposures is shown in

84

Table S1 of the supplement. The nPM is resuspended to 300 µg/ml, with trace
endotoxin levels (measured by Limulus amebocyte assay); frozen stocks at 20 °C retain
chemical stability for >30 days, including long-lived free radical species (Morgan et al.,
2011, Li et al., 2003).

2.2 Cell Culture
Primary mixed glia were cultured from neonatal cerebral cortex rat (mixed sex
Fisher 344 rats; Harlan, Indianapolis, IN). Cerebral cortex was mechanically
dissociated, strained by a 70 um Millipore filter, and plated onto 75 cm
2
cell culture
flasks in Dulbecco’s modified Eagle’s medium/F12 (cellgro, Meditatech, Herndon, VA)
supplemented with 10% fetal bovine serum, 1% penicillin, and 1% L-glutamine. Cultures
were incubated at 37°C with 95%/5% mixture of air and CO2. Media was refreshed
twice during the first week, and once during the second week. For microarray
experiments, cells were trypsonized and plated on 6-well plates, then treated with nPM
(10 µg/mL, 24 h) or LPS (100 ng/mL, 24 h).

2.3 RNA analysis
Microarray analysis was performed by Affymetrix rat whole genome 230.2 array.
q-PCR was performed using Bio-Rad CFX96 Real-Time System, with Biopioneer
master mix (QPCR-10) with SYBR green detection. Primers were generated by NCBI
Primer-BLAST (Basic Local Alignment Search Tool). Data are quantified as ΔΔ CT and
normalized to GAPDH.

85

2.4.1 Microarray normalization
Raw Affymetrix data were normalized by the Robust Multi-array Averaging R
algorithm, which smooths between chip variances: xnorm = Fn-1(F1(x)), where F1 is the
average of all chips analyzed, and Fn is the nth chip. The output is then transformed
into a log2 fold change. LPS and nPM microarray were performed separately, and
normalized to controls within each experiment.

2.4.2 Significance Analysis of Microarrays
Normalized data was analyzed using the Significance Analysis of Microarrays
(SAM). This program computes significance for each RNA measured by 100
permutations of the data. The input into SAM was an individual log2 fold change score
for each microarray probe, with significance expressed as q-values, in order to account
for false discovery rate.
After completion of normalization and significance analysis, the fold change vs
control from each probe of each experiment was compared across experiments. Each
individual dataset was cut off to a 1% rate of false positives, and the resulting files were
merged using Statistical Analysis System (SAS).

2.4.3 Weighted Gene Co-expression Network Analysis
Gene expression networks were identified by Weighted Gene Co-expression
Network Analysis (WGCNA) package in R (Langfelder 2008). Adjacencies were
calculated for each set using a soft-thresholding power of 6 (β value to which gene-gene
relationships are raised to). Adjacency for each gene pair were then defined using

86

Topological Overlap Matrix (TOM) and the consensus topological overlap was calculated
using the component-wise (parallel) minimum of the individual TOMs for nPM and LPS
responses. Hierarchical clustering with TOM as the input, identified consensus modules
specified as: unsigned network, minimum module size of 30, medium sensitivity
(deepSplit=2), and associations defined using biweight midcorrelations to measure
similarity between the groups.
From the module assignment, eigengenes were estimated for each module.
Eigengenes are the first principal component of the standardized expression profiles for
each set. An eigengene represents the gene’s overall score for a given module, based
on the sample gene expression values for all genes assigned to the module. We then
examined shared responses to nPM and LPS treatment for each module using biweight
midcorrelations, with adjustment of significance for multiple comparisons.
Finally, we examined module membership and screened for hub genes. Although the
hierarchical clustering assigns each gene to one module, module membership can be
statistically examined by calculating eigengene-based connectivity (kME) as the
correlation for each gene-by-module pair. kME is calculated in each set (nPM and LPS)
and then consensus kME scores are generated as the meta-analytic scores by
Stouffer’s method (Stouffer et al., 1949). Based on kME, hubs for each module were
defined as the genes with the highest connectivity (kME ≥ 0.80).

2.4.4 Pathway Analysis
Pathway enrichment analysis was done with Gene Ontology (Gene Ontology
Consortium, geneontology.org), with a p-value cutoff of <0.0001, and with Kyoto

87

Encyclopedia of Genes and Genomes (www.genome.jp/kegg/), with a p-value cutoff of
<0.0001.
Fold change genes from each dataset, and blue and turquoise module hub
genes from each dataset were also analyzed for enriched transcription factor targets
(TFTs) using WEB-based Gene Set Analysis Toolkit (WEB-GESTALT). Significance
cutoff was set to p<0.01.

2.5 TLR4 siRNA Cultures
Glia was cultured and maintained adhering to the protocols listed above (section
2.2). For siRNA experiments, cells were trypsonized and back transfected onto 6-well or
96-well plates, along with lipofectamine (10 µL/mL media) and TLR4 siRNA (Thermo
Fisher Scientific, siRNa ID 198667) or negative control siRNa (Thermo Fisher Scientific,
AM4611). After 48 hours, cells were treated with either nPM (10 µg/mL, 24 h) or LPS
(100 ng/mL, 24 h). After treatment, cultures were harvested for RNA (TRIzol, 15596026,
GIBCO BRL, Gaithersberg MD), and protein (RIPA buffer, 89900, Thermo Fisher
Scientific).

2.5.2 Inflammatory Multiplex Assay
Protein levels of 9 inflammatory proteins were analyzed the V-PLEX
Proinflammatory Panel 2 immunoassay (K15059D-1, Mesoscale Diagnostics, Rockville,
MD). The plate was blocked with proprietary blocking solution for 1 hour then washed 3
times. Samples were added and allowed to incubate at room temperature for 2 hours.
Plate was then washed, detection antibody was added, and it was incubated for 2

88

hours. Plate was washed 3 times, read buffer was added, and the plate was read on the
MESO QuickPlex SQ 120 (Mesoscale Diagnostics, Rockville, MD).

2.5.3 Statistical Analysis
Significance for q-PCR, MTT, and multiplex assay was done with one-way
ANOVA with Tukey’s posttest. All analysis was carried out using Graphpad Prism 6.

2.5.3 Heatmap Generation
A heatmap results from TLR4 pathway analysis was generated using MATLAB
(Mathworks). First, the matrices of the average of results is calculated then imported to
MATLAB where the heatmap branching points are determined by hierarchical clustering
analysis.

3. Results

3.1 Fold-change
Genomic responses of mixed glia to nPM and to LPS treatment were evaluated
from microarray data as fold-change. For nPM treatment, 1996 RNAs responded, with
increases of 920 and decreases of 1076, shown as Venn diagrams (Figures 1A, B, C).
For LPS treatment, 1316 RNAs responded, with increases of 606 and decreases of 710.
Supplementary Table 1 shows the top 20-fold changes for gene expression increases
and decreases for nPM (Supplementary Table 1A) and LPS (Supplementary Table 1B)
treatment.

89

Between the two datasets, 530 responders shared, of which, 268 were increased
in both LPS and nPM, 223 were decreased in both (Figure1A, B, C), while 39 RNAs
were divergent between nPM and LS treatment in direction of change (not shown).
Supplementary Table 2 shows the top 20 increases and decreases shared between the
treatments. These responses include mainly chemokines, cytokines, and other
inflammatory mediators.

3.2 Pathway Analysis
RNAs altered by fold change were analyzed for pathway enrichment by GO and
KEGG. GO showed enrichment for immune and stress responses with immune system
process, immune response, response to stress, defense response, and cytokine
responses all enriched in both datasets, as well as in shared responses (Supplementary
Table 3). KEGG analysis for the top ten pathways showed toll-like receptor signaling
pathway as one of the top ten, with a ratio of enrichment of 3.42 (data not shown).

3.3 Weighted Gene Co-expression Network Analysis (WGCNA)
WGCNA identified 45 correlation modules from the microarray data. Figure 2A
gives a cluster dendrogram of the genes queried: the top of Figure 2A shows branch
points of gene to gene correlations. Multiple clusters of genes were then grouped into
modules. The multi-colored bar shows the module each gene cluster was placed into.
The bottom two bars show increase (red) or decrease (blue) of each gene for nPM or
LPS treatment, respectively.

90

The statistical significance of each module was determined for nPM, LPS, or
consensus datasets (Figure 3). The consensus dataset includes all genes that
responded to both nPM and LPS treatment. The modules are given on the left by color.
Increase (red) or decrease (blue) for each module is depicted in the colored bar to the
right of the module notation. Of these 45 modules, 38 were enriched in the nPM
dataset, and 4 in the LPS dataset (Figure 3). Two modules in Figure 2A, blue and
turquoise, were enriched in the consensus dataset (Figure 3, right column). The blue
module consists of RNAs that increased (total 2042), while the turquoise module
contains those that decreased (total 2612).
Hub genes within each module were identified by the criterion kME >0.80. In the
blue module, the nPM responders had 2-fold more hubs (1196/2042) than LPS
responders (558/2042). By consensus analysis, 497 of the 2042 are hub genes. The
turquoise module had 2-fold more hub genes for nPM (1842) than for LPS (681), with
530 shared hub genes. The top 20 hub genes, based on a cutoff of 0.8 kME, of the blue
and turquoise consensus modules are given in supplementary table 4. These two
consensus modules both showed strong inverse correlation: R= -0.994 for nPM and R=
-0.968 for LPS. Both modules showed strong correlations between kME and fold-
change (Supplementary Figure 1).

3.4 Transcription Factor Target Sequences
Results from SAM fold change analysis and WGCNA were analyzed for
transcription factor target (TFT) enrichment. For fold change, all increased and
decreased genes were used. For WGCNA, only the hub genes (kME<0.80) from the

91

blue and turquoise module of each dataset were used. Responding RNAs were
analyzed for transcription factor targets (TFTs) of their genes (Table 1). Analysis of all
fold change responders (SAM, Significance Analysis of Microarrays) did not show
enrichment of any transcription factor sites for nPM, while LPS showed 10 enriched
TFTs. Because we detected almost 2,000 responding RNAs for nPM, the lack of
enriched TFTs when measuring by fold change alone suggests a broader genomic
response to nPM than to LPS. Among hub genes from the blue and turquoise modules
in the nPM dataset, 7 TFTs were enriched, while hub genes in the LPS dataset showed
8 (Table 1). LPS showed a consistent enrichment of interferon regulatory factors (IRFs)
and NF-kB associated binding sites (Table 1). This trend continued for the shared
genes, by both fold change and WGCNA.

3.5 Verification by q-PCR
Nine responses of TNF signaling pathways were further assessed by q-PCR to
verify microarray findings, and showed the same direction of change for nPM and LPS
treatment (Supplementary Table 5). Responses to LPS and nPM treatment were
significant for BRCA, Jak2, Stat1, TNF-a, TNFrSF9, TRAFd1, while FOS, TRAF3ip, and
TRAF6 did not respond (Supplementary Table 5).

3.6.1 TLR4 siRNA
Efficacy of TLR4 siRNA treatment was measured by protein and mRNA. TLR4
mRNA was increased 150% by nPM treatment and 30% by LPS (Figure 4A). TLR4
siRNA treatment reduced mRNA levels by 70% in control groups, and 90% in nPM

92

treated groups (Figure 4A). Scrambled RNA showed no reduction in TLR4 mRNA. TLR4
protein was not altered by nPM or LPS treatment (Figure 4B). siRNA for TLR4 reduced
protein levels approximately 60% in all groups (Figure 4B). Scrambled RNA showed no
differences versus controls.

3.6.2 Nitrite Concentration
Nitrite concentration was measured by the griess assay. Both nPM and LPS
treatment showed 200% increase in absorbance, indicating increased concentration of
nitrite in the media (Figure 5). TLR4 siRNA rescued this effect, restoring nitrite to normal
levels.

3.6.3 Proinflammatory Multiplex
Protein from control and nPM, both with and without siRNA, were analyzed for a
panel of inflammatory genes. Treatment with TLR4 siRNA rescued the effect for all
proteins observed, returning them back to control levels (Figure 6). siRNA treatment
alone altered protein concentrations for only IL-4 (Figure 6E).
TNFa was increased by nPM treatment (p<0.001). siRNA treatment reduced
nPM protein concentration (p<0.001), and was not different versus controls. IFN-y was
increased by nPM treatment (p<0.001), and was rescued by TLR4 knockdown
(p<0.001). IL-5 was also increased by nPM treatment (p<0.001), and rescued by TLR4
knockdown (p<0.05). IL-1b was also increased by nPM treatment (p<0.001), and
decreased by TLR4 siRNA (p<0.001). siRNA cultures treated with nPM had increased
IL-1b versus control cultures (P<0.01), but not increased versus siRNA treated control

93

cultures. IL-4 concentration was decreased by nPM treatment (p<0.05). Notably, siRNA
alone caused similar magnitude decreases, and did not rescue nPM effects (p<0.05).
KC concentration was increased by nPM treatment (p<0.001), and rescued by siRNA
treatment (p<0.001). nPM increased IL-6 protein (p<0.0001). siRNA treatment reduced
IL-6, with siRNA cultures treated with nPM not significantly different versus controls.

3.6.4 q-PCR of TLR4 siRNA
Extent of involvement of TLR4 activation and its constituent pathways was
determined using q-PCR. All results are shown in log 2 fold change. Corroborating
TNFa protein results, TNFa mRNA was induced by nPM treatment (p<0.0001), and
siRNA treatment reduced TNFa mRNA (p<0.0001) (Figure 7A). NF-kB transcription
increase was observed in response to nPM treatment (p<0.001) and by LPS treatment
(p<0.05). TLR4 siRNA attenuated nPM response (p<0.0001) (Figure 7B). MyD88 was
increased in both LPs and nPM treatment (p<0.01), and nPM response was returned to
baseline by siRNA (p<0.001). TRAF6, a downstream component of MyD88 dependent
TLR4 activation, was not changed in any treatment (not shown). TRAFD1, a negative
feedback regulator of TRAF6, was increased by nPM and LPS treatment (p<0.0001),
and rescued by TLR4 siRNA for the nPM treatment (p<0.0001) (Figure 7D). The
JAK/STAT pathway was activated by nPM treatment, and rescued by siRNA. JAK2
mRNA was increased by nPM and by LPS treatment (p<0.0001), and rescued by siRNA
treatment (p<0.0001) (Figure 7E). STAT1 mRNA was increased by nPM and by LPS
treatment (p<0.0001), and also rescued by siRNA treatment (p<0.0001) (Figure 7F).
iNOS mRNA was induced by both LPS and nPM (P<0.0001), and attenuated by TLR4

94

knockdown (p<0.0001) (Figure 7G). Notably, though IFN-y protein was increased
(Figure 6B), IFN-y mRNA was not changed in any treatments, potentially signaling a
return to baseline. Components of the TLR4 endocytosis pathway, which responds to
endotoxin binding, were also analyzed. TRIF (Toll/interleukin-1 receptor domain-
containing adaptor protein inducing interferon beta) was decreased by LPS treatment
(p<0.05, not shown), with no response to nPM. Interferon-b (IFN-b) showed no change
by nPM or LPS (not shown). Scrambled RNA controls showed similar trends as the
media only control for all genes, with nPM induction consistent in all genes versus
media only controls, and significant differences between siRNA and scrambled RNA
cultures treated with nPM for all genes queried.

3.6.5 Heatmap
A heatmap was generated of all genes queried by q-PCR in determining TLR4
pathway activation (Figure 8). Genes and treatment groups are clustered based on
correlation, with genes clustered on the y-axis, and treatment groups clustered on the x-
axis. Most notably, the three siRNA treated groups are all clustered together, denoting
similar trends for all genes queried. Comparably, for the media only cultures, the control
group is as far removed as possible from the nPM and LPS group. The scrambled RNA
treatment groups are clustered together, adjacent to the media nPM and media only
LPS groups. NF-kB pathway genes were clustered closest to TLR4, with TNFa, NF-kB,
and JAK2 in the same branch as TLR4.

95

4. Discussion
Exposure to particulate matter is associated with numerous cognitive
consequences (Power et al., 2011, Wellenius et al., 2012, Perera et al., 2006, Perera et
al., 2012, Guxens et al., 2012). Though hypothesized to be a consequence of chronic
neuroinflammation (Block et al., 2009), the direct role and extent of PM induced
neuroinflammation remains unclear. We demonstrate here that nPM causes widespread
changes in gene expression, with TLR4 integral to the inflammatory response to nPM.
Neuroinflammation by PM exposure is characterized primarily by glial activation
markers like GFAP and Iba1, (Cheng et al., 2016, Kleinman et al., 2008, Morgan et al.,
2011) and inflammatory cytokines (Campbell et al., 2005, Fonken et al., 2011,
Levesque et al., 2011, Morgan et al., 2011). Little work has been done to delineate the
full extent of nPM activation of glia, and what intracellular processes are involved in this
activation.
Microarray analysis of nPM treated mixed glial cultures showed nPM altered
transcription of 1,996 genes, with 920 increased and 1,076 decreased at twenty-four
hours. Approximately 25% (530) of the altered transcripts in the nPM treated cultures
were also observed in LPS treatment. Gene ontology pathway analysis of the nPM
treated cultures showed enrichment in many of the main stress and immune pathways,
for example: immune system processes, immune response, response to stress, and
defense response. nPM treatment also showed strong enrichment of the Toll-like
receptor signaling pathway, as seen by KEGG pathway analysis.
Responses to nPM and LPS were also analyzed by WGCNA. WGCNA created
correlation networks using gene transcription changes from both datasets. These

96

correlation networks are then defined into clusters and grouped into 45 different
modules. Of the 45 modules, 38 were enriched by nPM treatment, while 4 were
enriched by LPS treatment. In the consensus dataset, genes altered by both nPM and
LPS treatment, two modules were enriched. These two modules, blue (2042 genes) and
turquoise (2612 genes) (color names given arbitrarily), showed strong inverse
correlations (R= -0.994 for nPM, R=-0.968 for LPS) between the modules, with blue
genes all upregulated in the microarray, and turquoise genes all downregulated. The
blue module is enriched for multiple immune and stress response pathways (GO
pathway analysis), as well as multiple response to external stimulus pathways like
response to TNFa, response to LPS, and response to IFN-y (not shown). Interestingly,
GO pathways analysis of the turquoise module, all downregulated transcription
products, yielded only a single lysosome associated pathway as enriched (not shown).
In order to determine transcriptional regulators of nPM response, the results from
fold change and WGCNA were analyzed for transcription factor target enrichment. nPM
showed no enrichment for any TFTs when analyzed by fold change, with enrichment for
interferon regulatory factor (IRF) elements when the WGCNA modules were input. LPS
showed enrichment for numerous IRFs and NF-kB TFT elements, by both fold change
and WGCNA. For shared genes, there was strong enrichment of the IRF and NF-kb
TFT elements, with 6 elements enriched for IRF and 5 for NF-kB (by fold change).
These results highlight the limitations of fold change, as well as the utility of LPS as a
reference inflammatory response. By fold change alone nPM had no TFT element
enrichment. A recurring problem in microarray datasets is that analyzing the entirety of
the response, 2,000 significant differences for nPM treatment, dilutes any responses,

97

and creates difficulties in determining biological significance. When nPM response was
then narrowed down by what was shared with LPS, effectively reducing the input gene
list by 75%, strong enrichment for IRF and NF-kB became apparent.
TLR4 activation can induce numerous IRFs, as well as NF-kB (Pasparakis et al.,
2009). As a probable upstream regulator of the glial response to nPM, we then knocked
down TLR4 expression and repeated the experiment. Though traditionally thought of as
an endotoxin response, TLR4 is increasingly understood to respond to a broad range of
ligands, including damage associated molecular patterns (DAMPs) (Molteni et al., 2016,
Ooboshi et al., 2016). These DAMPs could potentially by produced by initial nPM
treatment, with TLR4 as the most upstream cellular response to nPM. TLR4 is critical to
immune activation, and its activation by nPM and the downstream effectors stemming
from its activation could be responsible for the detrimental effects observed by nPM
exposure (Block et al., 2007). TLR4 could be involved in air pollution related systemic
inflammation, as TLR4
-/-
mice showing a diminished ozone mediated airway neutrophilia
(Gabehart et al., 2015). In vitro treatment of TRL4
-/-
macrophages showed a reduction in
IL-6 and TNFa, although curiously only for PM10 and not PM2.5 (Shoenfelt et al., 2009).
TLR4 knockdown reduced the expression of numerous inflammatory cytokines,
many of which are downstream of NF-kB and IFN (Pasparakis et al., 2009). Specifically,
protein concentrations of cytokines IL-1B, IL-4, IL-5, IL-6, TNFa were reduced, as well
as IFN-y and CXCL1. TNFa and IL-1b, have both been shown to increase as a result of
in vivo TRAP exposure (Morgan et al., 2011, Win-Shwe et al., 2006, Levesque et al.,
2011, Campbell et al., 2005), and are relevant to the cognitive consequences of
exposure (Cite). TLR4 knockdown also rescues nitrate induction, a metabolite of nitric

98

oxide, and a marker of nitrosylative stress. Overall, TLR4 silenced mixed glial cultures
demonstrate that TLR4 activation is necessary for the inflammatory response to nPM.
TLR4 activation can function through MyD88 dependent or independent signaling
pathways. MyD88 dependent pathway activation results in NF-kB and AP-1 activation
(cite), while MyD88 independent pathway results in TLR4 endocytosis and increased in
interferon response elements (Cite, potential oversimplification). Following TLR4
activation by nPM, MyD88 dependent pathway components were the main affected
signals at 24-hour post treatment. NF-kB is activated by TLR4 through MyD88
dependent signaling. NF-kB mRNA was increased by nPM treatment, with TLR4
knockdown abolishing this increase. In contrast to NF- B signal, JNK1/2 and c-Jun,
another MyD88 dependent pathway in TLR4 signaling, did not change during nPM
exposure, demonstrating that not all MyD88 dependent pathways were affected by
nPM. Analysis of TRAF6, a kinase downstream of MyD88 and responsible for JNK1/2
and NF-kB activation, showed no change in mRNA by nPM treatment. However,
TRAFD1, a negative feedback regulator of TRAF6 (Sanada et al., 2008) showed
increased transcription by nPM treatment, with effects reversed by TLR4 knockdown.
This presents the possibility that at the 24-hour time point inflammatory responses are
beginning to recover and return to baseline. This highlights the need for future time
course studies to further elucidate the response. The lack of response by TRAF6,
combined with the induction of TRAFD1, suggests that NF-kB is regulated by another
pathway as well. TNFa is increased in both protein and mRNA by nPM treatment, and
TNFR1 can activate NF B through tumor necrosis factor receptor type-1 associated
death domain protein (TRADD) and recruitment of the ubiquitin ligase TNF receptor

99

associated factor 2 (TRAF2) and ubiquitination of the adaptor receptor-interacting
protein 1 (RIP1) (Sanada et al., 2008). Previous evidence from our lab demonstrated
the importance of theTNFR1 pathway during nPM toxicity (Cheng et al., 2016). Since
silencing of TNFR1 protected mixed glial cultures against nPM toxicity (Cheng et al.,
2016), activation of this pathway might be the cause of excessive NF B activation and
induction of inflammatory reactions.
The JAK/STAT pathway also showed induction by nPM treatment, seen by JAK2
and STAT1 mRNA increase, and was also mediated by TLR4. JAK/STAT activation can
stem from TNFa signaling, as TNFR1 receptor activation can activate the JAK/STAT
pathway (Guo et al., 1998). JAK/STAT is a component of the innate immune response,
and its activation could contribute to nPM toxicity. iNOS expression, part of the innate
immune response, was induced by nPM exposure, dependent on TLR4. It is possible
that JAK/STAT activation mediates this expression of iNOS (Dell’Albani et al., 2001).
Graphical representation of the pathway analysis is seen in the heatmap (Figure
8). TLR4 knockdown by siRNA treatment clustered control, nPM, and LPS groups
together, adjacent to media only control; signifying that decreasing TLR4 eliminated
much of the activation seen in both nPM and LPS treatment. Notably, the NF-kB
pathway genes, NF-kB and TNFa, were most closely clustered with TLR4, denoting
highest correlation in response.
TLR4 activation by LPS leads to endocytosis of the TLR4 receptor, and the
activation of endotoxin specific immune pathways (Tan et al., 2015). TRIF is a TLR4
adaptor protein activated in this endocytosis pathway. Downstream effects of this

100

pathway include IRF3 activation and IFN-b transcription (Rajaiah et al., 2015).
Transcription of TRIF and IFN-b are unchanged by nPM treatment, suggesting nPM
treatment does not elicit endotoxin specific pathways.
A hypothesized pathway of nPM activation of TLR4 pathways is given in figure
9. Briefly, nPM activates TLR4 via MyD88 dependent pathways, which could lead to
increased NF-kB transcription, and the concomitant increase in inflammatory cytokine
production observed here. The JAK/STAT pathway is also induced by nPM treatment, in
a TLR4 dependent manor; potentially being activated by TNFR1 activation. TLR4
receptor activation by endocytosis, done in response to CD14 and LPS binding, was not
changed by nPM.

5. Conclusion
Evidence presented here demonstrates that TLR4 plays an integral role in glial
response to nPM. Knocking down TLR4 reduces numerous inflammatory cytokines, and
attenuates activation of NF-kB and JAK/STAT pathways. These results further the
current understanding of nPM mechanisms in the brain, and could serve as valuable
tools to further understanding the mechanisms by which nPM elicits neuroinflammation,
and how this inflammation leads to neuron damage and cognitive effects.

Bibliography
1. Allen JL, Liu X, Pelkowski S, Palmer B, Conrad K, Oberdörster G, Weston D,
Mayer-Pröschel M, Cory-Slechta DA. Early postnatal exposure to ultrafine

101

particulate matter air pollution: persistent ventriculomegaly, neurochemical
disruption, and glial activation preferentially in male mice. Environ Health
Perspect. 2014 Sep;122(9):939-45. doi: 10.1289/ehp.1307984. Epub 2014 Jun 5.
PubMed PMID: 24901756; PubMed Central PMCID: PMC4154219.
2. Block ML, Calderón-Garcidueñas L. Air pollution: mechanisms of
neuroinflammation and CNS disease. Trends Neurosci. 2009 Sep;32(9):506-16.
doi: 10.1016/j.tins.2009.05.009. Epub 2009 Aug 26. Review. PubMed PMID:
19716187; PubMed Central PMCID: PMC2743793.
3. Block ML, Zecca L, Hong JS. Microglia-mediated neurotoxicity: uncovering the
molecular mechanisms. Nat Rev Neurosci. 2007 Jan;8(1):57-69. Review.
PubMed PMID: 17180163.
4. Bolton JL, Smith SH, Huff NC, Gilmour MI, Foster WM, Auten RL, Bilbo
SD. Prenatal air pollution exposure induces neuroinflammation and predisposes
offspring to weight gain in adulthood in a sex-specific manner. FASEB J. 2012
Nov;26(11):4743-54. doi: 10.1096/fj.12-210989. Epub 2012 Jul 19. Erratum in:
FASEB J. 2014 Jun;28(6):2737-8. PubMed PMID: 22815382.
5. Calderón-Garcidueñas L, Solt AC, Henríquez-Roldán C, Torres-Jardón R, Nuse
B, Herritt L, Villarreal-Calderón R, Osnaya N, Stone I, García R, Brooks DM,
González-Maciel A, Reynoso-Robles R, Delgado-Chávez R, Reed W. Long-term
air pollution exposure is associated with neuroinflammation, an altered innate
immune response, disruption of the blood-brain barrier, ultrafine particulate
deposition, and accumulation of amyloid beta-42 and alpha-synuclein in children
and young adults. Toxicol Pathol. 2008 Feb;36(2):289-310. doi:
10.1177/0192623307313011. Epub 2008 Mar 18. PubMed PMID: 18349428.
6. Calderón-Garcidueñas L, Reed W, Maronpot RR, Henríquez-Roldán C, Delgado-
Chavez R, Calderón-Garcidueñas A, Dragustinovis I, Franco-Lira M, Aragón-
Flores M, Solt AC, Altenburg M, Torres-Jardón R, Swenberg JA. Brain
inflammation and Alzheimer's-like pathology in individuals exposed to severe air
pollution. Toxicol Pathol. 2004 Nov-Dec;32(6):650-8. PubMed PMID: 15513908.
7. Campbell A, Oldham M, Becaria A, Bondy SC, Meacher D, Sioutas C, Misra C,
Mendez LB, Kleinman M. Particulate matter in polluted air may increase
biomarkers of inflammation in mouse brain. Neurotoxicology. 2005
Jan;26(1):133-40. PubMed PMID: 15527881.
8. Cheng H, Davis DA, Hasheminassab S, Sioutas C, Morgan TE, Finch CE. Urban
traffic-derived nanoparticulate matter reduces neurite outgrowth via TNFα in
vitro. J Neuroinflammation. 2016 Jan 26;13:19. doi: 10.1186/s12974-016-0480-3.
PubMed PMID: 26810976; PubMed Central PMCID: PMC4727336.
9. Cheng H, Saffari A, Sioutas C, Forman HJ, Morgan TE, Finch CE. Nano-Scale
Particulate Matter from Urban Traffic Rapidly Induces Oxidative Stress and

102

Inflammation in Olfactory Epithelium with Concomitant Effects on Brain. Environ
Health Perspect. 2016 May 17. [Epub ahead of print] PubMed PMID: 27187980.
10. Dell'Albani P, Santangelo R, Torrisi L, Nicoletti VG, de Vellis J, Giuffrida Stella
AM. JAK/STAT signaling pathway mediates cytokine-induced iNOS expression in
primary astroglial cell cultures. J Neurosci Res. 2001 Sep 1;65(5):417-24.
PubMed PMID: 11536325.
11. Elder A, Gelein R, Silva V, Feikert T, Opanashuk L, Carter J, Potter R, Maynard
A, Ito Y, Finkelstein J, Oberdörster G.Translocation of inhaled ultrafine
manganese oxide particles to the central nervous system. Environ Health
Perspect. 2006 Aug;114(8):1172-8. Erratum in: Environ Health Perspect. 2006
Aug;114(8):1178. PubMed PMID: 16882521; PubMed Central PMCID:
PMC1552007.
12. Fonken LK, Xu X, Weil ZM, Chen G, Sun Q, Rajagopalan S, Nelson RJ. Air
pollution impairs cognition, provokes depressive-like behaviors and alters
hippocampal cytokine expression and morphology. Mol Psychiatry. 2011
Oct;16(10):987-95, 973. doi: 10.1038/mp.2011.76. Epub 2011 Jul 5. PubMed
PMID: 21727897; PubMed Central PMCID: PMC3270364.
13. Guo, D., Dunbar, J.D., Yang, C.H., Pfeffer, L.M., Donner, D.B., 1998. Induction of
Jak/STAT signaling by activation of the type 1 TNF receptor. J Immunol 160(6),
2742-2750.
14. Guxens M, Aguilera I, Ballester F, et al. (2012) Prenatal exposure to residential
air pollution and infant mental development: Modulation by antioxidants and
detoxification factors. Environ Health Perspect 1201: 144-149
15. Kleinman MT, Araujo JA, Nel A, Sioutas C, Campbell A, Cong PQ, Li H, Bondy
SC. Inhaled ultrafine particulate matter affects CNS inflammatory processes and
may act via MAP kinase signaling pathways. Toxicol Lett. 2008 May
5;178(2):127-30. doi: 10.1016/j.toxlet.2008.03.001. Epub 2008 Mar 18. PubMed
PMID: 18420360; PubMed Central PMCID: PMC2397447.
16. Levesque S, Taetzsch T, Lull ME, Kodavanti U, Stadler K, Wagner A, Johnson
JA, Duke L, Kodavanti P, Surace MJ, Block ML. Diesel exhaust activates and
primes microglia: air pollution, neuroinflammation, and regulation of
dopaminergic neurotoxicity. Environ Health Perspect. 2011 Aug;119(8):1149-55.
doi: 10.1289/ehp.1002986. Epub 2011 May 11. PubMed PMID: 21561831;
PubMed Central PMCID: PMC3237351.
17. Li N, Sioutas C, Cho A, Schmitz D, Misra C, Sempf J, Wang M, Oberley T,
Froines J, Nel A. Ultrafine particulate pollutants induce oxidative stress and
mitochondrial damage. Environ Health Perspect. 2003 Apr;111(4):455-60.
18. Misra C, Kim S, Shen S, Sioutas C. A high flow rate, very low pressure drop
impactor for inertial separation of ultrafine from accumulation mode particles. J
Aerosol Sci. 2002; 33:735-752

103

19. Molteni M, Gemma S, Rossetti C. The Role of Toll-Like Receptor 4 in Infectious
and Noninfectious Inflammation. Mediators Inflamm. 2016;2016:6978936. doi:
10.1155/2016/6978936. Epub 2016 May 18. Review. PubMed PMID: 27293318;
PubMed Central PMCID: PMC4887650.
20. Morgan TE, Davis DA, Iwata N, Tanner JA, Snyder D, Ning Z, Kam W, Hsu YT,
Winkler JW, Chen JC, Petasis NA, Baudry M, Sioutas C, Finch
CE. Glutamatergic neurons in rodent models respond to nanoscale particulate
urban air pollutants in vivo and in vitro. Environ Health Perspect. 2011
Jul;119(7):1003-9. doi: 10.1289/ehp.1002973. PubMed PMID: 21724521;
PubMed Central PMCID: PMC3222976.
21. Ning Z, Geller MD, Moore KF, Sheesley R, Schauer JJ, Sioutas C. Daily variation
in chemical characteristics of urban ultrafine aerosols and inference of their
sources. Environ Sci Technol. 2007 Sep 1;41(17):6000-6.
22. Oberdörster G, Sharp Z, Atudorei V, Elder A, Gelein R, Kreyling W, Cox
C. Translocation of inhaled ultrafine particles to the brain.Inhal Toxicol. 2004
Jun;16(6-7):437-45. PubMed PMID: 15204759.
23. Ooboshi H, Shichita T. [DAMPs (damage-associated molecular patterns) and
inflammation]. Nihon Rinsho. 2016 Apr;74(4):573-8. Review. Japanese. PubMed
PMID: 27333742.
24. Pasparakis, M., 2009. Regulation of tissue homeostasis by NF-|[kappa]|B
signalling: implications for inflammatory diseases. Nature Reviews Immunology
9(11), 778-788.
25. Perera FP, Rauh V, Whyatt RM, Tsai WY, Tang D, Diaz D, Hoepner L, Barr D,
Tu YH, Camann D, Kinney P. Effect of prenatal exposure to airborne polycyclic
aromatic hydrocarbons on neurodevelopment in the first 3 years of life among
inner-city children.Environ Health Perspect. 2006 Aug;114(8):1287-92. PubMed
PMID: 16882541; PubMed Central PMCID: PMC1551985.
26. Perera FP, Tang D, Wang S, Vishnevetsky J, Zhang B, Diaz D, Camann D, Rauh
V. Prenatal polycyclic aromatic hydrocarbon (PAH) exposure and child behavior
at age 6-7 years. Environ Health Perspect. 2012 Jun;120(6):921-6. doi:
10.1289/ehp.1104315. Epub 2012 Mar 14. PubMed PMID: 22440811; PubMed
Central PMCID: PMC3385432.
27. Power, M. C., Weisskopf, M. G., Alexeeff, S. E., Coull, B. A., Spin III, A., &
Schwartz, J. (2011). Traffic-related air pollution and cognitive function in a cohort
of older men. Environmental health perspectives, 119(5), 682.
28. Ranft U, Schikowski T, Sugiri D, Krutmann J, Krämer U. Long-term exposure to
traffic-related particulate matter impairs cognitive function in the elderly. Environ
Res. 2009 Nov;109(8):1004-11. doi: 10.1016/j.envres.2009.08.003. Epub 2009
Sep 4. PubMed PMID: 19733348.

104

29. Sanada, T., Takaesu, G., Mashima, R., Yoshida, R., Kobayashi, T., Yoshimura,
A., 2008. FLN29 deficiency reveals its negative regulatory role in the Toll-like
receptor (TLR) and retinoic acid-inducible gene I (RIG-I)-like helicase signaling
pathway. J Biol Chem 283(49), 33858-33864.
30. Sardar SB, Fine PM, Mayo PR, Sioutas C. Size-fractionated measurements of
ambient ultrafine particle chemical composition in Los Angeles using the
NanoMOUDI. Environ Sci Technol. 2005 Feb 15;39(4):932-44. PubMed PMID:
15773464.
31. Shoenfelt J, Mitkus RJ, Zeisler R, Spatz RO, Powell J, Fenton MJ, Squibb KA,
Medvedev AE. Involvement of TLR2 and TLR4 in inflammatory immune
responses induced by fine and coarse ambient air particulate matter. J Leukoc
Biol. 2009 Aug;86(2):303-12. doi: 10.1189/jlb.1008587. PubMed PMID:
19406832; PubMed Central PMCID: PMC2726765.
32. Tan Y, Zanoni I, Cullen TW, Goodman AL, Kagan JC. Mechanisms of Toll-like
Receptor 4 Endocytosis Reveal a Common Immune-Evasion Strategy Used by
Pathogenic and Commensal Bacteria. Immunity. 2015 Nov 17;43(5):909-22. doi:
10.1016/j.immuni.2015.10.008. Epub 2015 Nov 3. PubMed PMID: 26546281;
PubMed Central PMCID: PMC4685471.
33. Tin-Tin-Win-Shwe, Yamamoto S, Ahmed S, Kakeyama M, Kobayashi T, Fujimaki
H. Brain cytokine and chemokine mRNA expression in mice induced by
intranasal instillation with ultrafine carbon black. Toxicol Lett. 2006 May
25;163(2):153-60. Epub 2005 Nov 15. PubMed PMID: 16293374.
34. Rajaiah R, Perkins DJ, Ireland DD, Vogel SN. CD14 dependence of TLR4
endocytosis and TRIF signaling displays ligand specificity and is dissociable in
endotoxin tolerance. Proc Natl Acad Sci U S A. 2015 Jul 7;112(27):8391-6. doi:
10.1073/pnas.1424980112. Epub 2015 Jun 23. PubMed PMID: 26106158;
PubMed Central PMCID: PMC4500272.
35. Veronesi B, Makwana O, Pooler M, Chen LC. Effects of subchronic exposures to
concentrated ambient particles. VII. Degeneration of dopaminergic neurons in
Apo E-/- mice. Inhal Toxicol. 2005 Apr;17(4-5):235-41. PubMed PMID:
15804941.
36. Wellenius GA, Boyle LD, Coull BA, Milberg WP, Gryparis A, Schwartz J,
Mittleman MA, Lipsitz LA. Residential proximity to nearest major roadway and
cognitive function in community-dwelling seniors: results from the MOBILIZE
Boston Study. J Am Geriatr Soc. 2012 Nov;60(11):2075-80. doi: 10.1111/j.1532-
5415.2012.04195.x. Epub 2012 Nov 5. PubMed PMID: 23126566; PubMed
Central PMCID: PMC3498530.
37. Zeng Y, Gu D, Purser J, Hoenig H, Christakis N. Associations of environmental
factors with elderly health and mortality in China.Am J Public Health. 2010

105

Feb;100(2):298-305. doi: 10.2105/AJPH.2008.154971. Epub 2009 Dec 17.
PubMed PMID: 20019314; PubMed Central PMCID: PMC2804639.

106

Tables and Figures

Figure 1

Figure 1: Venn diagram of increased and decreased RNAs, by fold change. A- total
RNAs altered by either treatment. B- Increased RNAs by either treatment. C- decreased
RNAs by either treatment.

107

Figure 2
Figure 2: WGCNA analysis of nPM and LPS datasets, grouping genes into modules
and showing fold change of RNA. A: Dendrogram using hierarchical clustering to group
genes by lowest values in a dissimilarity matrix (not shown). B: The module, labeled by
color, for each gene. Module placement is determined by highest similarity between
clusters of genes, and multiple clusters can be placed into a single module. C: Fold
change of RNA for each gene from nPM and LPS datasets, respectively. Red
represents increased fold change, and blue represents decreased fold change. A
portion of B and C has been expanded for clarity.

108

Figure 3

Figure 3: Significance of each module generated by the WGCNA, for nPM, LPS, and
consensus datasets. Module colors are given in the leftmost column, and are the same
modules from Figure 2B. Colors are named to modules arbitrarily. Significant modules

109

are denoted by the * (p<0.01). Of these 45 modules, 38 were enriched in nPM, 4 in
LPS, and 2 in the consensus dataset. Increased modules are signified by red, and
decreased modules signified by blue.

Table 1

Table 1: Transcription factor target analysis. The number given is the number of
different TFTs enriched for each dataset queried. Only the blue and turquoise modules
were used for WGCNA results. IRFs represents binding sites for multiple different
regulatory factors. Abbreviations: AP-1, activator protein 1; BACH1, BTB and CNC
homology 1; FC, fold change IPF1, insulin promoter factor 1; IRF, interferon regulatory
factor; NFAT, nuclear factor of activated T-cells; NF-kB, nuclear factor kappa-light-
chain-enhancer of activated B cells; NF1, neurofibromin 1; SRF, serum response factor;
STAT1, signal transducers and activator 1.

Transcription
factor
nPM

LPS

Shared

Fold
Change
WGCNA Fold
Change
WGCNA Fold
Change
WGCNA
AP-1 - - 1 - - -
BACH1 - 1 - - - -
IPF1 - 1 - - - -
IRFs - 3 6 4 6 4
NFAT - 1 1 1 - 1
NF-kB - - 2 2 5 2
NF1 - 1 - - - -
SRF - - - - 2 1
STAT1 - - - 1 1 -

110

Figure 4

Figure 4: mRNA and protein levels of TLR4. A- TLR4 was induced in mRNA but not
protein by nPM treatment. TLR4 siRNA reduced mRNA levels by 90% in nPM (ANOVA,
p<0.0001) and 70% in control groups (p<0.0001). LPS treatment reduced TLR4 mRNA
(p<0.05). B- TLR4 protein did not change by nPM or LPS treatment. TLR4 siRNA
reduced protein level by approximately 60% in all groups (ANOVA, p<0.05). Only select
significance is given in graphs.

A
B

111

Figure 5
Figure 5: Nitrite levels were measured by griess assay. Treatment by nPM and LPS
increased nitrate by 200% relative to controls (ANOVA, p<0.01). TLR4 siRNA rescued
this effect, with no difference between siRNA cultures treated with nPM or LPS versus
control groups.

0
2
4
6
8
G r ie s s a s s a y
C o n c e n tra tio n (  M )
C o n tro l
T LR 4
siR N A
S c ra m b le d
siR N A
C o n tro l
L P S
n P M
* * * *
* * * *
* * * *

112

Figure 6

B
C
D
E
F
G
A

113

Figure 6: Protein was measured by multiplex immunoassay. Only nPM and siRNA
groups were measured by multiplex. A- TNFa was increased by nPM treatment
(p<0.001). siRNA treatment reduced nPM protein concentration (p<0.001), and was not
different versus controls. B- IFN-y was increased by nPM treatment (p<0.001), and was
rescued by TLR4 knockdown (p<0.001). C- IL-5 was increased by nPM treatment
(p<0.001), and rescued by TLR4 knockdown (p<0.05). D- IL-1b was also increased by
nPM treatment (p<0.001), and decreased by TLR4 siRNA (p<0.001). siRNA cultures
treated with nPM had increased IL-1b versus control cultures (P<0.01), but not
increased versus siRNA treated control cultures. E- IL-4 concentration was decreased
by nPM treatment (p<0.05). Notably, siRNA alone caused similar magnitude decreases,
and did not rescue nPM effects (p<0.05). F- KC concentration was increased by nPM
treatment (p<0.001), and rescued by siRNA treatment (p<0.001). G- nPM increased IL-
6 protein (p<0.0001). siRNA treatment reduced IL-6, with siRNA cultures treated with
nPM not significantly different versus controls.

114

Figure 7: mRNA analysis of components of the TLR4 pathway. All values are given as
log 2 fold change. A- TNFa mRNA was induced by nPM and by LPS treatment
(p<0.0001, p<0.001, respectively). siRNA treatment reduced TNFa mRNA (p<0.0001).
B- NF-kB mRNA was increased by nPM treatment (p<0.001) and LPS treatment
T N F  m R N A
0 .1 2 5
0 .2 5
0 .5
1
2
4
8
1 6
E x p re s s io n re la tiv e to c o n tro l
C o n tro l
T LR 4
siR N A
C o n tro l
L P S
n P M
* * *
* * * *
* * * *

N F k B m R N A
0 .2 5
0 .5
1
2
4
E x p re s s io n re la tiv e to c o n tro l
C o n tro l
L P S
n P M
* * *
* * * *
*
C o n tro l
T LR 4
siR N A

M y D 8 8 m R N A
0 .5
1
2
E x p re s s io n re la tiv e to c o n tro l
C o n tro l
L P S
n P M
* *
* * *
* *
C o n tro l
T LR 4
siR N A

T R A F D 1 m R N A
0 .2 5
0 .5
1
2
4
E x p re s s io n re la tiv e to c o n tro l
C o n tro l
L P S
n P M
* * * *
* * * *
* * * *
C o n tro l
T LR 4
siR N A

J A K 2 m R N A
0 .5
1
2
4
8
E x p re s s io n re la tiv e to c o n tro l
C o n tro l
L P S
n P M
* * * *
* * * *
* * * *
C o n tro l
T LR 4
siR N A

S T A T 1 m R N A
0 .5
1
2
4
8
1 6
E x p re s s io n re la tiv e to c o n tro l
C o n tro l
L P S
n P M
* * * *
* * * *
* * * *
C o n tro l
T LR 4
siR N A

iN O S m R N A
1
3 2
1 0 2 4
E x p re s s io n re la tiv e to c o n tro l
C o n tro l
L P S
n P M
* * * *
* * * *
* * * *
C o n tro l
T LR 4
siR N A

A
B
C
D
E
F
G

115

(p<0.05), and rescued by siRNA treatment (p<0.0001). C- MyD88 was increased by
both nPM and LPS (p<0.01), and rescued by siRNA (p<0.001) D- TRAFD1 mRNA was
increased by nPM treatment (p<0.0001) and by LPS treatment (p<0.0001), and rescued
by siRNA treatment (p<0.0001). E- JAK2 mRNA was increased by nPM and LPS
treatment (p<0.0001), and rescued by siRNA treatment (p<0.0001). F- STAT1 mRNA
was increased by nPM and LPS treatment (p<0.0001), and rescued by siRNA treatment
(p<0.0001). G- iNOS mRNA was increased by nPM and LPS treatment (p<0.0001), and
rescued by siRNA treatment (p<0.0001).

Figure 8

Figure 8: Heatmap of q-PCR results of TLR4 pathway analysis. Fold change values
were standardized, and are shown from red (increased) to blue (decreased). Genes
queried are clustered on the y-axis, with genes clustered based on correlation.
Treatment groups are clustered on the x-axis, with groups showing similar responses
clustered closer together. Group nomenclature is labeled as first letter is RNA
transfection, and second letter is treatment condition. C, control; S, siRNA TLR4; R,
scrambled RNA; N, nPM; L, LPS.

116

Figure 9

Figure 9: Proposed pathway of TLR4 activation following nPM treatment. Yellow
proteins had increased mRNA or protein (cytokines) and were attenuated by TLR4
knockdown. Dark blue proteins were unresponsive to nPM. Green proteins were not
examined, and are proposed intermediates. nPM treatment activated MyD88 dependent
pathways, increasing NF-kB mRNA and increasing downstream cytokine productions of
NF-kB activation. JAK/STAT pathway was also activated by nPM exposure in a TLR4
dependent manor. LPS mediated TLR4 receptor activation by endocytosis was not
altered by nPM treatment.

117

Supplementary information:

Supplementary Table 1

Supplementary Table 1: Top 20 increased and decreased RNAs, by fold change, for
each treatment. Bold names are those included in the ‘response to LPS’ pathway in
Gene Ontology.

Gene
Symbol
Fold
Change
Atp1b1 0.08
C1qb 0.08
Ms4a6b 0.08
Arl11 0.08
Clec4a3 0.10
Gpnmb 0.11
C1qa 0.12
Csf1r 0.13
Lpl 0.13
C1qc 0.13
Rgs10 0.13
Cndp1 0.15
Egr2 0.15
Sepp1 0.15
Laptm5 0.15
Cd68 0.15
Folr2 0.16
Cxcr4 0.16
Acta1 0.17
Fos 0.17

Gene
Symbol
Fold
Change
Cxcl2 160
Cxcl9 136
Cxcl10 126
Cxcl11 120
Ifit3 110
Il1a 108
Ccl5 107
Cxcl3 104
Ifit2 86.2
RT1-Ba 60.8
Ccl20 59.9
Rsad2 59.8
Cfb 56.4
Mx1 55.5
RT1-Da 47.9
Cmpk2 45.5
Il10 43.3
Ccl4 40.5
Herc6 39.9
F10 39.4
Gene
Symbol
Fold
Change
Actg2 0.03
Cnn1 0.05
Acta1 0.06
Rasgrp3 0.11
Pcp4l1 0.13
Tnnt2 0.14
Tspan8 0.15
Omd 0.15
Hmcn1 0.16
Tpm2 0.16
Gpr34 0.16
Tinagl1 0.16
Calb1 0.17
Fbln5 0.17
Ogn 0.17
Cdh1 0.18
Myl9 0.18
Synm 0.19
Tnmd 0.19
Heph 0.19

Gene
Symbol
Fold
Change
Mmp3 165
Ccl20 72.9
Cxcl2 61.8
Il1b 49.3
Pla2g2a 49.2
Mmp9 46.1
Ccl5 42.4
Cxcl3 39.4
Cxcl1 34.2
Cfb 34.0
Il1a 32.3
Ptges 25.4
Il6 24.8
Cxcl6 24.3
Rsad2 20.7
Mmp13 20.6
Ifit3 19.7
Cxcl10 19.2
Slpi 18.0
Orm1 15.7
LPS Top 20 RNA Responses nPM Top 20 RNA Responses

118

Supplementary Table 2

Supplementary Table 2: Top 20 increased and decreased genes that were shared in
both nPM and LPS datasets. Values given are fold change, and bold gene names
signifies their inclusion in the “response to LPS” gene ontology pathway.
Gene
Symbol
nPM
Fold
Change
LPS
Fold
Change
Cxcl2 160 61.8
Ccl20 59.9 72.9
Ccl5 107 42.4
Cxcl3 104 39.4
Il1a 108 25.3
Cfb 56.4 34
Ifit3 110 19.7
Cxcl10 126 19.2
Rsad2 59.8 20.7
Il1b 25.3 49.3
Il6 35.3 24.8
Mx1 55.5 15.3
Mmp3 16.4 165
Ptges 21.7 25.4
H2-Ea 47.9 14.3
Cxcl1 19.7 34.2
Pla2g2a 16.6 49.2
Mmp13 21.3 20.6
Cd74 35.7 11.7
Csf3 19.2 15.7

Gene
Symbol
nPM
Fold
Change
LPS
Fold
Change
Acta1 0.06 0.17
Tgfbr1 0.28 0.23
Mfap4 0.30 0.39
Igf1 0.22 0.27
Ogn 0.23 0.41
Plbd1 0.44 0.35
Cndp1 0.36 0.15
Gmpr 0.34 0.32
Lilrb4 0.38 0.25
Fam107a 0.20 0.41
Nkain4 0.24 0.31
Car3 0.23 0.41
Oplah 0.28 0.30
Krt19 0.31 0.37
MGC72614 0.23 0.41
Sdc2 0.39 0.43
Cnn1 0.05 0.45
Lpl 0.28 0.13
Klf2 0.30 0.41
Angpt2 0.36 0.37

119

Supplementary Table 3

Pathway nPM LPS Shared
Immune system
process
261 212 124
Immune
response
156 131 80
Response to
stress
414 333 176
Defense
response
154 134 83
Cytokine
receptor binding
48 34 25
Cytokine activity 42 28 24

Supplementary Table 3: Select pathways from Gene Ontology analysis. p value cutoff
set at <0.0001

120

Supplementary Table 4

Table 4: Top 20 consensus hub genes, as measured by kME, for the turquoise (A) and
blue (B) modules. Values given are kME, and bold gene names signifies their inclusion
in the “response to LPS” Gene Ontology pathway.

Gene Symbol kME
Slc44a2 0.98
Atp1b1 0.98
Ctcf 0.98
Sesn1 0.98
Ttyh3 0.98
Cln8 0.98
Clec4a3 0.97
Nrp1 0.97
Fry 0.97
Lpl 0.97
Pink1 0.97
Slc37a2 0.97
Ifi30 0.97
Stmn1 0.97
Eif1b 0.97
Gusb 0.97
Ms4a6b 0.97
Klf2 0.97
Timp3 0.97
Map3k1 0.96
Gene Symbol kME
Prdx5 0.99
RT1-EC2 0.99
Ccl4 0.99
Il1b 0.99
Akr1b8 0.99
Cfb 0.99
Adprm 0.99
Mmp9 0.99
Cxcl2 0.99
Cxcl3 0.99
Cd74 0.99
Fbxl5 0.99
Trib3 0.99
RGD1561113 0.99
Ccl5 0.98
Mmp3 0.98
Usp18 0.98
Irf7 0.98
Usp18 0.98
Angptl4 0.98
A B

121

Supplementary figure 1

Supplementary figure 1: Correlation of the consensus hub genes kME from the blue and
turquoise modules, and fold change.

122

Supplementary Table 5

Supplementary Table 5: Genes queried by q-PCR, to verify microarray analysis. Trend of
change was consistent for all genes analyzed. Significant increase in JAK2, STAT1, TNFa,
TNFRsf9, and TRAFD1. BRCA was decreased, while Fos and TRAF3IP and TRAF6 were
unchanged. Abbreviations: BRCA, breat cancer, Fos, FBJ murine osteosarcoma viral oncogene
homolog; JAK2, janus kinase 2; STAT1, signal transducer and activator of transcription 1;
TNFa, tumor necrosis factor alpha; TNFRSF9, TNF receptor super family 9; TRAFD1, TNF
receptor associated factor zinc finger domain containing 1; TRAF3IP, TNF receptor associated
factor 3 interacting protein; TRAF6, TNF receptor associated factor 6.

Gene
name
Microarray
fold change
nPM
q-PCR
nPM
Microarray
fold change
LPS
q-PCR
LPS
Increased
JAK2 2.7 5.1 4.2 3.3
STAT1 4.5 3.5 2.0 2.5
TNFa 2.2 19 3.3 9.9
TNFRsf9 5.9 26 2.3 22
TRAFD1 2.7 3.1 1.6 4.7
Decreased
BRCA 0.8 0.4 1.0 0.6
Unchanged
Fos 0.6 0.9 0.2 1.1
TRAF3ip 0.8 1.1 1.5 1.1
TRAF6 1.2 1.5 N/A 1.4

123

Chapter 3
Developmental exposure to particulate air pollutants reduces cell
proliferation in the hippocampus and impairs hippocampal dependent
memory
Abstract
Exposure to particulate matter from ambient air pollution is correlated with poorer
cognition, with evidence indicating memory impairments in exposed adult populations,
and laboratory models. Gestation and early development are of critical importance, as
adverse environments during development can cause long term cognitive deficits. The
current experiment explores this critical window, exposing rats from gestation to
adulthood (28 weeks) to nanoscale particulate matter (nPM, diameter <0.2 um).
Developmental exposure to nPM causes hippocampal dependent impairments in
contextual memory, as well as depressive behaviors. Analysis of cell proliferation in the
adult hippocampus showed reduced cellular proliferation in the dentate gyrus and CA1.
These results highlight a hippocampal vulnerability to developmental pollution exposure,
and are of relevance to future epidemiological studies of developmental vulnerabilities
to air pollution exposure.

124

1. Introduction:
Traffic related air pollution (TRAP) affects cognition at all stages of life. It is linked
with poorer cognitive outcomes in adults (Gatto et al., 2014; Power et al., 2011; Weuve
et al., 2012), including impaired memory (Ailshire et al., 2014a; Ailshire et al., 2014b;
Schikowski et al., 2015; Wellenius et al., 2012), increased depressive symptoms (Lim
2012), and increased anxiety (Power et al., 2015). Developmental TRAP exposure is
also linked to subclinical cognitive effects, including lower mental development,
increased anxiety and depressive behavior (Guxens et al., 2012; Perera et al., 2006;
Perera et al., 2012).
Current work in animal models also demonstrates memory deficits following
pollution exposure (Allen et al., 2014a; Fonken et al., 2011). PM2.5 exposure mice
show deficits in spatial memory as tested by barnes maze (Fonken et al., 2011), and
TRAP exposed mice show object recognition deficits by the novel object recognition
(NOR) test (Allen et al., 2014a). The hippocampus is integral for memory, and TRAP
exposure reduce neuronal processes in the hippocampus, with CA1 subregion specific
vulnerabilities (Davis et al., 2013; Fonken et al., 2011).
It is hypothesized that cognitive effects of TRAP exposure derive from
neuroinflammation, caused by increased circulating cytokines, as well as direct
translocation of particles through the olfactory system (Block et al., 2007; Block et al.,
2009; Oberdoerster et al., 2004). Chronic neuroinflammation, and the concomitant
increase in inflammatory cytokines and reactive oxygen species production, can in turn
lead to neuronal damage and reduced neuroinflammation (Block et al., 2009). TRAP
exposure increases circulating TNFa in both humans (Delfino et al., 2008) and animal
models (Levesque et al., 2011; Li et al., 2013). This increase in proinflammatory

125

cytokines from TRAP exposure can lead to blood brain barrier (BBB) disruption
(Calderon et al., 2008; Hartz et al., 2008), and peripheral cytokines can increase BBB
permeability to other circulating factors (Blamire et al., 2000). A small postmortem
sample of young adults from Mexico City show endothelial disruption and microglial
activation centered around the BBB, correlated with ambient PM2.5 exposure (Calderon
et al., 2008). Animal models exposed to diesel exhaust particles (DEP) showed
increased oxidative stress localized around the BBB (Hartz et al., 2008). This alters the
BBB, allowing for increased penetrance of inflammatory factors and leading to
neuroinflammation (Argaw et al., 2006). Neuroinflammation from TRAP exposure is well
documented, with whole brain increased inflammatory cytokines (Bolton et al., 2012;
Levesque et al., 2011), and microglial activation seen in numerous model systems
(Allen et al., 2014a; Allen et al., 2014b; Bolton et al., 2012; Levesque et al., 2011).
Though dependent on the type and duration of neuroinflammation, the chronic TRAP
induced neuroinflammation seen here is hypothesized to decrease adult neurogenesis
(Fuster-Matanzo et al., 2013; Russo et al., 2011). Preliminary evidence suggests
exactly this, though more work is need to prove this effect (Costa et al., 2015). The
current experiment will coalesce these observations, in an attempt to show a pathway of
action from TRAP inhalation to memory impairment.
Serum circulating cytokines will be measured via multiplex immunoassays, and
neurogenesis by 5-ethynyl-2’-deoxyuridine (EdU) immunohistochemistry. We aim to
further prove hippocampal involvement in TRAP induced memory impairment by
utilizing the novel object in context recognition test (NOIC).

126

2. Methods:

2.1 Animals and Ethics Statement
Sprague Dawley G0 pregnant rats were obtained from Envigo (Hayward, CA).
Protocols were approved by the University of Southern California Institutional Animal
Care and Use Committee, and animals were maintained under standard rodent chow
with ad lib water according to NIH guidelines. Prior to DRL behavior testing animals
were transitioned to reverse cycle housing for the duration of the experiment. Animals
were euthanized by ketamine (90mg/kg), xylazine (2.8 mg/kg), and acepromazine (0.72
mg/kg) cocktail.

2.2 nPM Collection and Exposure
The nPM utilized in these studies are a subfraction of ultrafine PM (<200 nm
diameter) collected from urban air in Los Angeles, California, near the CA-110 Freeway
(Misra et al., 2002). These aerosols represent a mix of fresh ambient PM mostly from
vehicular traffic on this freeway (Ning et al., 2007). The nPM was collected on Teflon
filters resuspended in deionized water by vortexing and sonication (Morgan et al.,
2011). The nPM comprise approximately 20% by mass of ambient PM2.5 in that location
(Sardar et al., 2005) Water soluble metals and organic compounds were efficiently
transferred from the filter collection medium into aqueous suspension used for
exposures (Morgan et al., 2011). Relative to the total filtered-trapped ultrafines (PM0.2),
the nPM subfraction eluted into aqueous phases is depleted in black carbon and water
insoluble organic compounds. The chemical composition of nPM used in these
exposures is shown in Table S1 of the supplement. The nPM is resuspended to 340

127

µg/ml, with trace endotoxin levels (measured by Limulus amebocyte assay); frozen
stocks at 20 °C retain chemical stability for >30 days, including long-lived free radical
species (Morgan et al., 2011, Li et al., 2003). Rats were exposed five hours a day, three
days a week. Exposure began on gestational day 2, and lasted for the duration of
gestation and through weaning. At weaning the male offspring were housed and
exposed for 21 more weeks. Total exposed time was 28 weeks, including 3 weeks of
gestation.

2.3 Behavior
Offspring began behavioral testing at ten weeks of age.
Novel Object in Context Recognition (NOIC): NOIC was used to assess hippocampal
dependent object and context recognition (Balderas et al., 2008). Animals were
habituated to novel object recognition chambers with no objects, in two separate
locations, one days one and two, for five minutes each day. Chambers were 31.75 cm x
48.3 cm x 51.75 cm, with dim lighting. On day three animals were exposed to two
distinct objects (A and B) for five minutes in the first location. Objects were cylindrical
jars filled with blue water, and square transparent glass containers, approximately 7 cm
x 8.3 cm x 10.16 cm. Animals showed no preference between objects on day three, and
objects A and B were counterbalanced between trials. 24 hr later, the animals were
placed in location two, with duplicates of object A, again for ten minutes. The following
day animals were again placed in location two, this time with objects A and B, again for
five minutes. Time spent investigating the object was recorded, with blinded
experiments evaluating behavior on a live video feed. Discrimination index was

128

calculated by exploration time of object B on the final day, divided by total time spent
exploring both objects. A discrimination index of 0.5 shows no preference for novel
object, with higher than 0.5 indicating novel object preference. Shift from baseline was
time spent exploring object B on the final day versus the first encounter with object B.
Forced Swim: Rats were placed in a cylindrical water bath, 41cm height, 26 cm
wide. Water depth was 18cm, and temperature was maintained at 24-25 degrees
Celsius. Experiment was conducted in red lighting. Trials were video recorded for five
minutes with latency to first period of immobility and total time immobile recorded.
Elevated Zero Maze: To test for anxiety behavior, rats were tested using an
elevated zero maze (63.5 cm fall height, 116.8 cm outside diameter). This maze has a
circular track, divided into four section. Two sections were open, with 3 cm high curbs,
and two sections were closed, with 17.5 cm high walls. Animals were placed in the
maze for 5 min while the experimenter records the total time spent in open sections
(defined as the head and front two paws in open arms), and total open arm entries.
Differential Reinforcement of Low Rates of Response (DRL): Animals were kept
on reverse cycle housing, with testing performed at the beginning of the dark cycle.
Operant conditioning was performed in Med Associates (Fairfax, VT). Rats were then
placed on a DRL paradigm where lever presses are only rewarded if the required time
has elapsed since the last lever press. Lever presses before the time-out period has
elapsed resets the time-out period. Training occurred over 15 days in 1 hr sessions.
Initially rats were placed on a DRL 0 (zero second timeout) for 5 one hour sessions to
train lever pressing for palatable food (45 mg pellet, 35% kcal far enriched with sucrose,
F05989, Bio-serv) paired with 1 second light stimulus. Rats pressed on average 110

129

times during the one-hour period by the end of DRL 0. Reinforcement time outs then
progressed to DRL 5 for 5 sessions, then DRL 10 for 5 sessions. Animals were then
placed on DRL 20 for 10 sessions to test impulsivity (Simon et al., 2013; Sokolowski et
al., 1994). Next, animals DRL 72 was conducted for 10 sessions. DRL 72 is
hypothesized to measure depressive behavior, and is rescued by numerous
antidepressants (O’Donnell et al., 2005). Finally, an extinction trial was performed.
Impulsive behavior is represented as the efficiency to obtain food rewards (rewards
obtained divided by lever presses).

2.4 Body Composition Analysis
Body composition of the rats was measured at 14 weeks and 28 weeks of age.
Composition was measured using nuclear magnetic resonance spectroscopy (NMR),
which measures fat mass, lean mass, and fluid mass.

2.5 Histochemistry
Immunofluorescence: Rats were perfused with 0.9% saline, followed by 4%
paraformaldehyde in 0.1M borate buffer. Following perfusion, brains were cryoprotected
in 12% sucrose and frozen in isopentane. Brains were then stored at -80 °C until use.
Brains were coronally sliced, with 30 µm thick sections, using a frozen microtome.
Floating sections were permeabilized in triton-100 for 30 minutes, blocked in 3% NDS
for 30 minutes, then incubated in primary antibody at 4 °C overnight. Primary antibodies
used were Neun (ab104224, Abcam), glial fibrillary acidic protein (GFAP, ab7260,
Abcam), ionized calcium binding adaptor molecule (Iba1, ab5076, Abcam)), myelin

130

oligodendrocyte glycoprotein (MOG, ab115597, Abcam), and doublecortin (ab18723,
Abcam). Slides were then washed and incubated with Alexa Fluor secondary antibodies
for four hours (1:500, Thermo Scientific).
5-ethynyl- 2’-deoxyuridine (EdU): Rats were intraperitoneally injected with 41.1
mg/kg EdU 7 times over three days, ending 18 days before tissue collection. After
primary staining, slides were stained for EdU using the Click-iT Plus EdU Alexa Fluor
555 Imaging Kit (C10638, Thermo Fisher Scientific). Cell counts were made using a
fluorescent microscope. Sections were taken centered around -3.80 mm from bregma.
For each animal three sections were analyzed, with all six hemispheres averaged.
Ventricle measurement: Three sequential coronal sections were plated per
animal. Lateral ventricles were measured at bregma -0.26 mm to -0.40 mm, while the
third ventricle was measured at bregma -1.30 mm to -1.50 mm. Sections were imaged
using a Leica DMLB brightfield microscope, and total area was analyzed with Image J.

2.6 Inflammatory Multiplex Assay
Protein levels of 9 inflammatory proteins were analyzed the V-PLEX
Proinflammatory Panel 2 immunoassay (K15059D-1, Mesoscale Diagnostics, Rockville,
MD). Serum was collected by cardiac puncture prior to perfusion, and blood was
centrifuged for 20 min at 4 °C. Serum was removed and stored at -80 °C until use. The
plate was blocked with proprietary blocking solution for 1 hour then washed 3 times.
Samples were added and allowed to incubate at room temperature for 2 hours. Plate
was then washed, detection antibody was added, and it was incubated for 2 hours.

131

Plate was washed 3 times, read buffer was added, and the plate was read on the
MESO QuickPlex SQ 120 (Mesoscale Diagnostics, Rockville, MD).

3. Results:
3.1 Novel Object in Context Recognition (NOIC)
NOIC showed hippocampal dependent memory deficits as a result of nPM
treatment during development, observed by impaired contextual memory for an object.
nPM exposed rats had decreased discrimination index by approximately 15% (t-test,
p<0.01, Figure 1A), and a 60 smaller shift from baseline (t-test, p<0.01, Figure 1B).

3.2 Forced Swim
In the forced swim test, nPM exposed rats showed depressive symptoms, as
quantified by more and quicker onset of immobile behavior. nPM exposure reduced the
time until first period of immobility by 30% (t-test, p<0.01, Figure 2A). Rats exposed to
nPM also had an 80% increased total time spent immobile (t-test, p<0.001, Figure 2B)
during the five-minute trial.

3.3 EdU analysis
Newly divided cells were quantified using EdU staining, with cells counted in the
dentate gyrus and CA1. In the CA1, stratum oriens, radiatum, and lacunosum were
analyzed. EdU positive cells in the stratum oriens and radiatum were decreased
approximately 60% (Figure 3A, B) (t-test, p<0.01, p<0.0001, respectively). Stratum

132

lacunosum was unaffected by treatment (data not shown), potentially due to a floor
effect of newly divided cells. The combined CA1 was decreased by 60% (Figure, 3C, t-
test, p<0.0001). All three layers of the DG, the polymorphic, granule, and molecular
layers, were reduced 40% by nPM exposure (Figure 4A, B, C, t-test, p< 0.05 for all
regions). Total cells in the dentate was reduced by approximately 40% (Figure 4D, t-
test, p<0.01).

3.4 Ventricular Area
Developmental nPM exposure lead to decreased ventricle size for all ventricles
measure. Area of the lateral ventricle, both left and right hemisphere, and third ventricle
were quantified. nPM reduced the area of both the lateral ventricles (t-test, p<0.05) and
third ventricle by approximately 25% (t-test, p<0.01) (Figure 5A, B).

3.5 Circulating Cytokines
The levels of circulating cytokines in the serum was measured by multiplex
immunoassay. nPM exposure decreased levels of anti-inflammatory cytokines, while
inflammatory cytokines remained unchanged. IL-10 was reduced by 2 pg/uL (25%, t-
test, p<0.01) by nPM exposure (Figure 6A). IL-13 was also reduced, by 0.6 pg/uL (20%,
t-test, p<0.001), by nPM exposure (Figure 6B). Finally, nPM reduced IL-4 by 0.3 pg/uL
(25%, t-test, p<0.01) (Figure 6C). Also queried by the assay were IFN-y, IL-1b, IL-5, IL-
6, CXCL1, and TNFa, all of which were not altered by nPM exposure (not listed).

133

3.6 Body Composition Analysis
Body composition was measured twice by NMR spectroscopy, at 11 and 25
weeks of age. nPM shifted body composition towards reduced fat mass. Total fat mass
was decreased by approximately 20% at both time points (t-test, p<0.05, Figure 7) by
nPM exposure. Decreased fat mass was compensated by a slight, but insignificant,
increase in lean mass. Also measured was fluid mass and total weight, neither of which
were altered by nPM exposure.

4. Discussion
Developmental exposure to nPM resulted in behavioral deficits, and decreased
cell proliferation in the hippocampus. Hippocampal dependent contextual memory was
impaired, and an increase of depressive behaviors was observed. Cell proliferation was
reduced in both the dentate gyrus and CA1 of the hippocampus. nPM exposure also
resulted in change in circulating anti-inflammatory cytokines, as seen by decreased IL-
4, IL-10, and IL-13 in serum.
The NOIC test used here provides evidence for hippocampal specific deficits
caused by nPM exposure. NOIC is hippocampal dependent, seen by injections of
anisomycin (protein synthesis inhibitor) into the hippocampus inhibiting NOIC
performance, while injections into the amygdala, insular cortex, and perirhinal cortex did
not (Balderas et al., 2008). This extends previous evidence of NOR deficits following
developmental exposure (Allen 2014), and corroborates spatial memory deficits
following adult exposure (Fonken et al., 2011).
Previous work demonstrates hippocampal sensitivity to TRAP exposure, with
subregional specific vulnerabilities. The CA1 shows special vulnerability, with decreased

134

spine densities and reduced neurite length after a 10 month TRAP exposure; while the
dentate gyrus was unchanged for any measure (Fonken et al., 2011). Mice exposed to
nPM for 10 weeks also show CA1 specificity, with neurites and myelin basic protein,
and increased microglial activation, with the DG again unresponsive (Woodward et al.,
unpublished). Hippocampal slice cultures treated with nPM also showed CA1 specific
vulnerability, with increased glutamate receptors, PSD-95, and synaptophysin in the
CA1 but not DG (Davis et al., 2013).
We hypothesized that the increased neuroinflammation seen by microglial
activation (Woodward et al., unpublished) and increased inflammatory cytokines (Win-
Shwe et al., 2006) in the hippocampus could affect cell proliferation and neurogenesis,
as chronic neuroinflammation is can reduce adult neurogenesis (Fuster-Matanzo et al.,
2013; Russo et al., 2011). Indeed, nPM exposure decreased total cell proliferation in
both the DG and CA1. Adult neurogenesis is integral to learning and memory (Costa et
al., 2015b), and it is possible that the decreased cell proliferation observed by nPM
exposure is responsible for the deficit in contextual memory seen in the NOIC test. It is
important to note that total newly proliferated cells are measured, and that no
colocalization was done. Therefore, the decreased cell number cannot be ascribed to
any individual cell population.
Developmentally exposed rats also had impaired performance in the forced swim
test, indicative of depressive behaviors. Similar effects have previously been observed
in prenatally exposed mice (Davis et al., 2013), indicating a potential prenatal
vulnerability to induction of depressive symptoms. Prenatal environmental stressors can
have exaggerated impacts on cognitive development (Kinsella et al., 2009). More so,

135

environmental tobacco smoke inhalation is shown to have exaggerated effects when
paired with maternal hardships (Perera et al., 2006). It is possible that stress from the
exposure paradigm from noise and handling could compound nPM effects.
There are two proposed mechanisms of nPM derived neuroinflammation. First,
nPM can translocate directly from the olfactory epithelium to the olfactory bulb, and into
the brain (Oberdorster et al., 2004; Elder et al., 2006). Recent evidence from our lab
showed rapid inflammatory responses in the olfactory bulb after acute nPM exposure
(Cheng et al., 2016). However, the occurrence of inflammation in regions like the cortex
(Cheng et al., 2016; Levesque et al., 2011; Morgan et al., 2011), cerebellum (Cheng et
al., 2016; Zhang et al., 2012), and midbrain (Levesque et al., 2011b), which are multiple
synaptic connections away from the olfactory bulb, suggestions the occurrence of a
second method as well. This second source of neuroinflammation could be an indirect
effect of nPM derived peripheral inflammation increasing systemic cytokines like TNFa
(Levesque et al., 2011; Li et al., 2013). These inflammatory cytokines can cross the
BBB, as well as increase BBB permeability to other circulating factors (Argaw et al.,
2006). There were no increases of inflammatory cytokines in serum of nPM exposed
rats, however all anti-inflammatory cytokines queried (IL-4, IL-10, and IL-13) were
decreased as a result of exposure, indicating a shift towards inflammatory states. This
shift toward chronic inflammation could lead to basal neuroinflammation, and impaired
cell proliferation in the brain (Fuster-Matanzo et al., 2013).
The decreased ventricle size observed in nPM exposed rats is unexpected, and
quite interesting. Mouse neonatal exposure to TRAP has been observed to increase
ventricle size in the lateral ventricles (Allen et al., 2014). Enlarged ventricles is

136

associated with multiple inflammatory disorders of the CNS (Lepore et al., 2013), and
ventricular dilation is seen following chronic infection (Hermes et al., 2008). Conversely,
there is little precedence for decreased ventricle size following environmental stressors
or disease. The most plausible cause for reduced ventricular size is cerebral swelling,
causing idiopathic intracranial hypertension (IIH) (Reid et al., 1980). While the cause of
IIH is still uncertain, evidence suggests that it is due to venous outflow obstruction
(Skau et al., 2005) which in turn increases cranial pressure. TRAP exposure has well
documented cardiovascular effects (Chen et al., 2009; Pope et al., 2002), and acute
exposures can increase global blood coagulability (Baccarelli et al., 2007). Though the
exact causes of venous outflow obstruction are not yet fully understood, it is possible
that nPM exposure can increase cerebral blood pressure, in turn reducing ventricular
size.

5. Conclusion
Developmental TRAP exposure causes hippocampal dependent memory
impairment and depressive symptoms, with memory impairments potentially stemming
from reduced cell proliferation in the DG and CA1. This hippocampal vulnerability is of
importance considering the lifelong nature of TRAP exposure, and the potential of
increased rate of hippocampal decline seen in normal aging.

137

Bibliography

1. Ailshire JA, Clarke P. Fine particulate matter air pollution and cognitive function
among U.S. older adults. J Gerontol B Psychol Sci Soc Sci. 2015 Mar;70(2):322-
8. doi: 10.1093/geronb/gbu064. Epub 2014 Jun 6. PubMed PMID: 24906394;
PubMed Central PMCID: PMC4351385.
2. Ailshire JA, Crimmins EM. Fine particulate matter air pollution and cognitive
function among older US adults. Am J Epidemiol. 2014 Aug 15;180(4):359-66.
doi: 10.1093/aje/kwu155. Epub 2014 Jun 24. PubMed PMID: 24966214; PubMed
Central PMCID: PMC4128773.
3. Allen JL, Liu X, Pelkowski S, Palmer B, Conrad K, Oberdörster G, Weston D,
Mayer-Pröschel M, Cory-Slechta DA. Early postnatal exposure to ultrafine
particulate matter air pollution: persistent ventriculomegaly, neurochemical
disruption, and glial activation preferentially in male mice. Environ Health
Perspect. 2014 Sep;122(9):939-45. doi: 10.1289/ehp.1307984. Epub 2014 Jun 5.
PubMed PMID: 24901756; PubMed Central PMCID: PMC4154219.
4. Argaw AT, Zhang Y, Snyder BJ, Zhao ML, Kopp N, Lee SC, Raine CS, Brosnan
CF, John GR. IL-1beta regulates blood-brain barrier permeability via reactivation
of the hypoxia-angiogenesis program. J Immunol. 2006 Oct 15;177(8):5574-84.
PubMed PMID: 17015745.
5. Baccarelli A, Zanobetti A, Martinelli I, Grillo P, Hou L, Giacomini S, Bonzini M,
Lanzani G, Mannucci PM, Bertazzi PA, Schwartz J.Effects of exposure to air
pollution on blood coagulation. J Thromb Haemost. 2007 Feb;5(2):252-60. Epub
2006 Nov 1. PubMed PMID: 17083648.
6. Balderas I, Rodriguez-Ortiz CJ, Salgado-Tonda P, Chavez-Hurtado J, McGaugh
JL, Bermudez-Rattoni F. The consolidation of object and context recognition
memory involve different regions of the temporal lobe. Learn Mem. 2008 Aug
21;15(9):618-24. doi: 10.1101/lm.1028008. Print 2008 Sep. PubMed PMID:
18723431; PubMed Central PMCID: PMC2632790.
7. Blamire AM, Anthony DC, Rajagopalan B, Sibson NR, Perry VH, Styles
P. Interleukin-1beta -induced changes in blood-brain barrier permeability,
apparent diffusion coefficient, and cerebral blood volume in the rat brain: a
magnetic resonance study. J Neurosci. 2000 Nov 1;20(21):8153-9. PubMed
PMID: 11050138.
8. Block ML, Calderón-Garcidueñas L. Air pollution: mechanisms of
neuroinflammation and CNS disease. Trends Neurosci. 2009 Sep;32(9):506-16.
doi: 10.1016/j.tins.2009.05.009. Epub 2009 Aug 26. Review. PubMed PMID:
19716187; PubMed Central PMCID: PMC2743793.
9. Block ML, Zecca L, Hong JS. Microglia-mediated neurotoxicity: uncovering the
molecular mechanisms. Nat Rev Neurosci. 2007 Jan;8(1):57-69. Review.
PubMed PMID: 17180163.

138

10. Bolton JL, Smith SH, Huff NC, Gilmour MI, Foster WM, Auten RL, Bilbo
SD. Prenatal air pollution exposure induces neuroinflammation and predisposes
offspring to weight gain in adulthood in a sex-specific manner. FASEB J. 2012
Nov;26(11):4743-54. doi: 10.1096/fj.12-210989. Epub 2012 Jul 19. Erratum in:
FASEB J. 2014 Jun;28(6):2737-8. PubMed PMID: 22815382.
11. Calderón-Garcidueñas L, Solt AC, Henríquez-Roldán C, Torres-Jardón R, Nuse
B, Herritt L, Villarreal-Calderón R, Osnaya N, Stone I, García R, Brooks DM,
González-Maciel A, Reynoso-Robles R, Delgado-Chávez R, Reed W. Long-term
air pollution exposure is associated with neuroinflammation, an altered innate
immune response, disruption of the blood-brain barrier, ultrafine particulate
deposition, and accumulation of amyloid beta-42 and alpha-synuclein in children
and young adults. Toxicol Pathol. 2008 Feb;36(2):289-310. doi:
10.1177/0192623307313011. Epub 2008 Mar 18. PubMed PMID: 18349428.
12. Calderón-Garcidueñas L, Reed W, Maronpot RR, Henríquez-Roldán C, Delgado-
Chavez R, Calderón-Garcidueñas A, Dragustinovis I, Franco-Lira M, Aragón-
Flores M, Solt AC, Altenburg M, Torres-Jardón R, Swenberg JA. Brain
inflammation and Alzheimer's-like pathology in individuals exposed to severe air
pollution. Toxicol Pathol. 2004 Nov-Dec;32(6):650-8. PubMed PMID: 15513908.
13. Chen JC, Schwartz J. Neurobehavioral effects of ambient air pollution on
cognitive performance in US adults. Neurotoxicology. 2009 Mar;30(2):231-9. doi:
10.1016/j.neuro.2008.12.011. Epub 2008 Dec 30. PubMed PMID: 19150462.
14. Cheng H, Saffari A, Sioutas C, Forman HJ, Morgan TE, Finch CE. Nano-Scale
Particulate Matter from Urban Traffic Rapidly Induces Oxidative Stress and
Inflammation in Olfactory Epithelium with Concomitant Effects on Brain. Environ
Health Perspect. 2016 May 17. [Epub ahead of print] PubMed PMID: 27187980.
15. Costa LG, Cole TB, Coburn J, Chang YC, Dao K, Roqué PJ. Neurotoxicity of
traffic-related air pollution. Neurotoxicology. 2015 Nov 21. pii: S0161-
813X(15)30024-3. doi: 10.1016/j.neuro.2015.11.008. [Epub ahead of print]
PubMed PMID: 26610921; PubMed Central PMCID: PMC4875879.
16. Costa V, Lugert S, Jagasia R. Role of adult hippocampal neurogenesis in
cognition in physiology and disease: pharmacological targets and
biomarkers. Handb Exp Pharmacol. 2015;228:99-155. doi: 10.1007/978-3-319-
16522-6_4. Review. PubMed PMID: 25977081.
17. Davis DA, Bortolato M, Godar SC, Sander TK, Iwata N, Pakbin P, Shih JC,
Berhane K, McConnell R, Sioutas C, Finch CE, Morgan TE. Prenatal exposure to
urban air nanoparticles in mice causes altered neuronal differentiation and
depression-like responses. PLoS One. 2013 May 29;8(5):e64128. doi:
10.1371/journal.pone.0064128. Print 2013. PubMed PMID: 23734187; PubMed
Central PMCID: PMC3667185.

139

18. Davis DA, Akopian G, Walsh JP, Sioutas C, Morgan TE, Finch CE. Urban air
pollutants reduce synaptic function of CA1 neurons via an NMDA/NȮ pathway in
vitro. J Neurochem. 2013 Nov;127(4):509-19. doi: 10.1111/jnc.12395. Epub 2013
Aug 26. PubMed PMID: 23927064; PubMed Central PMCID: PMC3818296
19. Delfino RJ, Staimer N, Tjoa T, Gillen DL, Polidori A, Arhami M, Kleinman MT,
Vaziri ND, Longhurst J, Sioutas C. Air pollution exposures and circulating
biomarkers of effect in a susceptible population: clues to potential causal
component mixtures and mechanisms. Environ Health Perspect. 2009
Aug;117(8):1232-8. doi: 10.1289/ehp.0800194. Epub 2009 Apr 29. PubMed
PMID: 19672402; PubMed Central PMCID: PMC2721866.
20. Elder A, Gelein R, Silva V, Feikert T, Opanashuk L, Carter J, Potter R, Maynard
A, Ito Y, Finkelstein J, Oberdörster G.Translocation of inhaled ultrafine
manganese oxide particles to the central nervous system. Environ Health
Perspect. 2006 Aug;114(8):1172-8. Erratum in: Environ Health Perspect. 2006
Aug;114(8):1178. PubMed PMID: 16882521; PubMed Central PMCID:
PMC1552007.
21. Fonken LK, Xu X, Weil ZM, Chen G, Sun Q, Rajagopalan S, Nelson RJ. Air
pollution impairs cognition, provokes depressive-like behaviors and alters
hippocampal cytokine expression and morphology. Mol Psychiatry. 2011
Oct;16(10):987-95, 973. doi: 10.1038/mp.2011.76. Epub 2011 Jul 5. PubMed
PMID: 21727897; PubMed Central PMCID: PMC3270364.
22. Fuster-Matanzo A, Llorens-Martín M, Hernández F, Avila J. Role of
neuroinflammation in adult neurogenesis and Alzheimer disease: therapeutic
approaches. Mediators Inflamm. 2013;2013:260925. doi: 10.1155/2013/260925.
Epub 2013 Apr 3. Review. PubMed PMID: 23690659; PubMed Central PMCID:
PMC3649701.
23. Gatto NM, Henderson VW, Hodis HN, St John JA, Lurmann F, Chen JC, Mack
WJ. Components of air pollution and cognitive function in middle-aged and older
adults in Los Angeles. Neurotoxicology. 2014 Jan;40:1-7. doi:
10.1016/j.neuro.2013.09.004. Epub 2013 Oct 19. PubMed PMID: 24148924;
PubMed Central PMCID: PMC3946571.
24. Guxens M, Aguilera I, Ballester F, et al. (2012) Prenatal exposure to residential
air pollution and infant mental development: Modulation by antioxidants and
detoxification factors. Environ Health Perspect 1201: 144-149
25. Hartz AM, Bauer B, Block ML, Hong JS, Miller DS. Diesel exhaust particles
induce oxidative stress, proinflammatory signaling, and P-glycoprotein up-
regulation at the blood-brain barrier. FASEB J. 2008 Aug;22(8):2723-33. doi:
10.1096/fj.08-106997. Epub 2008 May 12. PubMed PMID: 18474546; PubMed
Central PMCID: PMC2493447.

140

26. Hermes G, Ajioka JW, Kelly KA, Mui E, Roberts F, Kasza K, Mayr T, Kirisits MJ,
Wollmann R, Ferguson DJ, Roberts CW, Hwang JH, Trendler T, Kennan RP,
Suzuki Y, Reardon C, Hickey WF, Chen L, McLeod R. Neurological and
behavioral abnormalities, ventricular dilatation, altered cellular functions,
inflammation, and neuronal injury in brains of mice due to common, persistent,
parasitic infection. J Neuroinflammation. 2008 Oct 23;5:48. doi: 10.1186/1742-
2094-5-48. PubMed PMID: 18947414; PubMed Central PMCID: PMC2588578.
27. Kinsella MT, Monk C. Impact of maternal stress, depression and anxiety on fetal
neurobehavioral development. Clin Obstet Gynecol. 2009 Sep;52(3):425-40. doi:
10.1097/GRF.0b013e3181b52df1. Review. PubMed PMID: 19661759; PubMed
Central PMCID: PMC3710585.
28. Lepore S, Waiczies H, Hentschel J, Ji Y, Skodowski J, Pohlmann A, Millward JM,
Paul F, Wuerfel J, Niendorf T, Waiczies S.Enlargement of cerebral ventricles as
an early indicator of encephalomyelitis. PLoS One. 2013 Aug 22;8(8):e72841.
doi: 10.1371/journal.pone.0072841. eCollection 2013. PubMed PMID: 23991157;
PubMed Central PMCID: PMC3750011
29. Levesque S, Taetzsch T, Lull ME, Kodavanti U, Stadler K, Wagner A, Johnson
JA, Duke L, Kodavanti P, Surace MJ, Block ML. Diesel exhaust activates and
primes microglia: air pollution, neuroinflammation, and regulation of
dopaminergic neurotoxicity. Environ Health Perspect. 2011 Aug;119(8):1149-55.
doi: 10.1289/ehp.1002986. Epub 2011 May 11. PubMed PMID: 21561831;
PubMed Central PMCID: PMC3237351.
30. Levesque S, Surace MJ, McDonald J, Block ML. Air pollution & the brain:
Subchronic diesel exhaust exposure causes neuroinflammation and elevates
early markers of neurodegenerative disease. J Neuroinflammation. 2011 Aug
24;8:105. doi: 10.1186/1742-2094-8-105. PubMed PMID: 21864400; PubMed
Central PMCID: PMC3184279.
31. Li R, Navab M, Pakbin P, Ning Z, Navab K, Hough G, Morgan TE, Finch CE,
Araujo JA, Fogelman AM, Sioutas C, Hsiai T.Ambient ultrafine particles alter lipid
metabolism and HDL anti-oxidant capacity in LDLR-null mice. J Lipid Res. 2013
Jun;54(6):1608-15. doi: 10.1194/jlr.M035014. Epub 2013 Apr 6. PubMed PMID:
23564731; PubMed Central PMCID: PMC3646462.
32. Morgan TE, Davis DA, Iwata N, Tanner JA, Snyder D, Ning Z, Kam W, Hsu YT,
Winkler JW, Chen JC, Petasis NA, Baudry M, Sioutas C, Finch
CE. Glutamatergic neurons in rodent models respond to nanoscale particulate
urban air pollutants in vivo and in vitro. Environ Health Perspect. 2011
Jul;119(7):1003-9. doi: 10.1289/ehp.1002973. PubMed PMID: 21724521;
PubMed Central PMCID: PMC3222976.
33. Ning Z, Geller MD, Moore KF, Sheesley R, Schauer JJ, Sioutas C. Daily variation
in chemical characteristics of urban ultrafine aerosols and inference of their

141

sources. Environ Sci Technol. 2007 Sep 1;41(17):6000-6. PubMed PMID:
17937273.
34. Oberdörster G, Sharp Z, Atudorei V, Elder A, Gelein R, Kreyling W, Cox
C. Translocation of inhaled ultrafine particles to the brain.Inhal Toxicol. 2004
Jun;16(6-7):437-45. PubMed PMID: 15204759.
35. Perera FP, Rauh V, Whyatt RM, Tsai WY, Tang D, Diaz D, Hoepner L, Barr D,
Tu YH, Camann D, Kinney P. Effect of prenatal exposure to airborne polycyclic
aromatic hydrocarbons on neurodevelopment in the first 3 years of life among
inner-city children.Environ Health Perspect. 2006 Aug;114(8):1287-92. PubMed
PMID: 16882541; PubMed Central PMCID: PMC1551985.
36. Perera FP, Tang D, Wang S, Vishnevetsky J, Zhang B, Diaz D, Camann D, Rauh
V. Prenatal polycyclic aromatic hydrocarbon (PAH) exposure and child behavior
at age 6-7 years. Environ Health Perspect. 2012 Jun;120(6):921-6. doi:
10.1289/ehp.1104315. Epub 2012 Mar 14. PubMed PMID: 22440811; PubMed
Central PMCID: PMC3385432.
37. Pope CA 3rd, Burnett RT, Thun MJ, Calle EE, Krewski D, Ito K, Thurston
GD. Lung cancer, cardiopulmonary mortality, and long-term exposure to fine
particulate air pollution. JAMA. 2002 Mar 6;287(9):1132-41. PubMed PMID:
11879110; PubMed Central PMCID: PMC4037163.
38. Power, M. C., Weisskopf, M. G., Alexeeff, S. E., Coull, B. A., Spin III, A., &
Schwartz, J. (2011). Traffic-related air pollution and cognitive function in a cohort
of older men. Environmental health perspectives, 119(5), 682.
39. Power MC, Kioumourtzoglou MA, Hart JE, Okereke OI, Laden F, Weisskopf
MG. The relation between past exposure to fine particulate air pollution and
prevalent anxiety: observational cohort study. BMJ. 2015 Mar 24;350:h1111. doi:
10.1136/bmj.h1111. PubMed PMID: 25810495; PubMed Central PMCID:
PMC4373600.
40. Reid AC, Matheson MS, Teasdale G. Volume of the ventricles in benign
intracranial hypertension. Lancet. 1980 Jul 5;2(8184):7-8. PubMed PMID:
6104264.
41. Russo I, Barlati S, Bosetti F. Effects of neuroinflammation on the regenerative
capacity of brain stem cells. J Neurochem. 2011 Mar;116(6):947-56. doi:
10.1111/j.1471-4159.2010.07168.x. Epub 2011 Jan 19. Review. PubMed PMID:
21198642; PubMed Central PMCID: PMC3382108.
42. Skau M, Brennum J, Gjerris F, Jensen R. What is new about idiopathic
intracranial hypertension? An updated review of mechanism and
treatment. Cephalalgia. 2006 Apr;26(4):384-99. Review. PubMed PMID:
16556239.
43. Wellenius GA, Boyle LD, Coull BA, Milberg WP, Gryparis A, Schwartz J,
Mittleman MA, Lipsitz LA. Residential proximity to nearest major roadway and

142

cognitive function in community-dwelling seniors: results from the MOBILIZE
Boston Study. J Am Geriatr Soc. 2012 Nov;60(11):2075-80. doi: 10.1111/j.1532-
5415.2012.04195.x. Epub 2012 Nov 5. PubMed PMID: 23126566; PubMed
Central PMCID: PMC3498530.
44. Weuve J, Puett RC, Schwartz J, Yanosky JD, Laden F, Grodstein F. Exposure to
particulate air pollution and cognitive decline in older women. Arch Intern Med.
2012 Feb 13;172(3):219-27. doi: 10.1001/archinternmed.2011.683. PubMed
PMID: 22332151; PubMed Central PMCID: PMC3622279.
45. Win-Shwe, Yamamoto S, Ahmed S, Kakeyama M, Kobayashi T, Fujimaki
H. Brain cytokine and chemokine mRNA expression in mice induced by
intranasal instillation with ultrafine carbon black. Toxicol Lett. 2006 May
25;163(2):153-60. Epub 2005 Nov 15. PubMed PMID: 16293374.
46. Zhang H, Liu H, Davies KJ, Sioutas C, Finch CE, Morgan TE, Forman HJ. Nrf2-
regulated phase II enzymes are induced by chronic ambient nanoparticle
exposure in young mice with age-related impairments. Free Radic Biol Med.
2012 May 1;52(9):2038-46. doi: 10.1016/j.freeradbiomed.2012.02.042. Epub
2012 Mar 6. Erratum in: Free Radic Biol Med. 2014 Dec;77:388. PubMed PMID:
22401859; PubMed Central PMCID: PMC3342863.

143

Tables and Figures

Figure 1
Figure 1: Novel object in context recognition (NOIC). A- Discrimination index for
exploration of novel object, calculated by investigation of novel object divided by total
time spent investigating. Exposed mice had reduced exploration of novel object (t-test,
p<0.01). B- Shift from baseline, defined as difference in discrimination index between
first trial and test trial. nPM exposed mice had less increase between trials (t-test,
p<0.01).

Figure 2:

A B

A B

144

Figure 2: Forced swim test. Depressive symptoms were measured by the forced swim
test. A- Time until first period of immobility. nPM exposed mice showed lower latency (t-
test, p<0.01). B- Total time spent immobile during the trial. nPM exposed mice showed
increased total time spent immobile (t-test, p<0.001).
Figure 3
Figure 3: EdU quantification of the CA1. Cell counts are the average of six
hemispheres, measured from three slices separated by 150 um. A- nPM treatment
reduced EdU positive cells in the stratum oriens (t-test, p<0.01). B- Radiatum layer of
the CA1, 60% reduction (t-test, p<0.0001). C- Stratum lacunosum, no effect seen by
nPM treatment. Potentially confounded by a floor effect. D- Combined results from the
three analyzed regions. nPM treatment reduced EdU positive cells by 60% (t-test,
p<0.0001).

A
B
D
C

145

Figure 4
Figure 4: EdU quantification of the dentate gyrus (DG). Cell counts are the average of
six hemispheres, measured from three slices separated by 150 um. A- nPM treatment
reduced EdU positive cells in the polymorphic layer (t-test, p<0.05). B- Granule cell
layer, 40% reduction (t-test, p<0.05). C- Molecular layer was decreased by nPM
treatment (t-test, p<0.05). D- Combined results from the three analyzed regions. nPM
treatment reduced EdU positive cells by 40% (t-test, p<0.01).

A
B
D
C

146

Figure 5

Figure 5: Quantification of ventricle area. Three coronal sections per animal, were
measured and averaged. A- Lateral ventricle area, average of both hemispheres. nPM
exposed reduced ventricle size (t-test, p<0.05). B- Sample image of lateral ventricle. C-
Third ventricle area. nPM exposure decreased third ventricle size (t-test, p<0.01). D-
Representative image of the third ventricle.

A
B
D
C

147

Figure 6
Figure 6: Serum cytokines protein levels, measured by multiplex immunoassay. A- IL-
10 was decreased by nPM exposure (t-test, p<0.01). B- nPM exposure decreased IL-13
(t-test, p<0.001). C- IL-4 was decreased by nPM exposure (t-test, p<0.01).

A B
C

148

Figure 7
Figure 7: Body composition analysis by NMR. A- Total fat mass was decreased by nPM
exposure at 11 weeks of age (t-test, p<0.05). B- Total far mass was decreased by nPM
exposure after conclusion of exposure, at 25 weeks of age (t-test, p<0.05).

A B

149

Chapter 4

Minimizing Air Pollution Exposure: A Practical Policy to Protect Vulnerable
Older Adults from Death and Disability

Abstract:
Air pollution causes an estimated 200,000 deaths per year in the United States
alone. Older adults are at greater risk of mortality caused by air pollution. Here we quantify
the number of older adult facilities in Los Angeles County that are exposed to high levels
of traffic derived air pollution, and propose policy solutions to reduce pollution exposure
to this vulnerable subgroup. Distances between 20,362 intersections and 858 elder care
facilities were estimated, and roads or highways within 500 of facilities were used to
estimate traffic volume exposure. Of the 858 facilities, 54 were located near at least one
major roadway, defined as a traffic volume over 100,000 cars/day. These 54 facilities
house approximately 6,000 older adults. Following standards established for schools, we
recommend legislation mandating the placement of new elder care facilities a minimum
of 500 feet from major roadways in order to reduce unnecessary mortality risk from
pollution exposure.

150

1. Introduction
Today it is estimated that 200,000 people die per year in the United States alone
due to inhalation of air pollution (Caiazzo et al., 2005). The relative risk for mortality due
to living in a heavily polluted area is roughly equivalent to the relative risk of being
overweight (a BMI between 25-39.9 kg/m
2
) (Pope et al., 2002). Study of the ACS
Cancer Prevention II population showed that for every 10 ug/m
3
increase in fine
particulate matter (PM2.5) concentration, an important component of air pollution, there
was a concurrent increase of 6% in mortality due to cardiopulmonary conditions, an 8%
increase in mortality from lung cancer, and a 4% increase in total all-cause mortality
(Pope et al., 2002). National guidelines are 35 ug/m
3
daily maximum and 12.0 ug/m
3
annual maximum for PM2.5.
While air pollution negatively affects everyone, children and older adults are
especially vulnerable to adverse health effects. Air pollution exposure at a young age
can cause cognitive impairments and asthma (Perera et al., 2006, Morgenstern et al
2008), while pollution exposure at older ages causes a disproportionate increase in
mortality, when compared to middle aged individuals (Hoek et al., 2002; Katsouyanni et
al., 2001). Mortality from air pollution exposure is mainly due to cardiovascular and
cardiopulmonary effects (Brook et al., 2004; Chen et al., 2013; and Pope et al., 2002).
Typical age related declines in the cardiovascular system, such as decreased
reserve capacity, decreased elasticity of the arterial wall, and decreased ability to
respond to norepinephrine signals to adjust blood pressure, make the older adult
population extremely vulnerable to cardiovascular and cardiopulmonary disease, and
exposure to air pollution amplifies these risks. Although individuals aged 65 and over

151

only represent 13.3% of the population, they account for 42.8% of all cases of heart
disease, and 52.1% of coronary disease (CDC 2010).

Exposure to high levels of PM2.5
is associated with an increased intima-medial thickness, a common measure of the
progression of atherosclerosis (Adar et al., 2013). Individuals exposed to PM2.5 also
showed a decrease in heart rate variability resulting in less adaptability to changes in
cardiovascular demands, increasing susceptibility to myocardial infarction (Adar et al.,
2007). Particulate matter also causes inflammation of the alveolar cells in the lung,
which then releases signaling molecules that increase blood coaguability, raising the
chances of clot formation (Ruckerl et al., 2006).
Older adults, especially those in poor health with diminished cardiovascular
function, are not as adept at handling these added stressors, therefore they have a
higher risk of mortality as a result of the exposure. Individuals with preexisting
conditions, especially cardiovascular or cardiopulmonary conditions, are more
vulnerable to air pollution (Goldberg et al., 2001). Also, individuals who sustained a
myocardial infarction during heavy pollutant exposure show an increased 10-year future
mortality risk, and survivors of a previous myocardial infarction show greater all around
mortality later in life if exposed to air pollution (Rosenbloom et al., 2012; Berglind et al.,
2009). Exposure to PM2.5 has also been associated with increased mortality to
individuals with type-2 diabetes (Peters et al., 2012; Katsouyanni et al., 2001), where
older adults are again overrepresented, comprising 39.4% of the diabetic population.
Increasing awareness of the harmful effects of air pollution has led to the
development of guidelines to prevent excess exposure to these toxicants. The EPA has
been successful in monitoring and reducing air pollution across cities in the United

152

States, however, its measurement methods are coarse and very poorly measure the
variability within the city. Two locations within the same city often have greater
differences in pollution concentration than the difference between two cities, and the
difference in risk can also be larger within a city than between two cities (Jerrett et al.,
2005, Miller et al., 2007). In one study, the range of exposure to particulate matter within
Los Angeles was 20ug/m
3
, versus a range of 16ug/m
3
between 116 other cities studied
(Jerrett et al., 2005). Colloquially know as hot spots, these are areas within a city with
much higher pollution concentration than background, often due to higher traffic volume.
Identifying these areas of greater pollution concentration, and minimizing exposure to
sensitive populations in these areas is a critical step to minimize adverse health effects
from pollution exposure.
In this study we quantify the number of older adult facilities, specifically nursing
homes, assisted living facilities, and adult day healthcare centers, in Los Angeles
County that are currently being exposed to unnecessarily high levels of traffic derived
air pollution. Methods for the reduction in pollution exposure through the strategic
placement of facilities are proposed.
2. Materials and Methods
2.1 Data Acquisition
Data were compiled from six sources, which provided information for line
coordinates of roads and highways in Los Angeles, traffic counts at one mile
intersections or freeway exits, and addresses and occupancies for facilities throughout
Los Angeles which cater to the older adult population. Data on road and highway
coordinates came from a 2010 TIGER road file (Topologically Integrate Geographic
Encoding and Referencing) of the county of Los Angeles. The TIGER file contained

153

geographical coordinates in GCS_NORTH_AMERICAN_1983 for 2,366,677 nodes on
the centerline networks of roads used by the US Census Bureau. Additionally, the
TIGER road file also contained MAF/TIGER feature classification codes (MTCC),
depicting the type of road on which each node was located. Node coordinates were
converted to latitude and longitudes using Global Mapper 15 software so that they could
be more easily incorporated with traffic and facility location data. Traffic data is a
combination of the Los Angeles Department of Transit traffic survey section 10 year
2001-2010 summary, and the State of California 2012 Annual Average Daily Traffic
Report. The first file had eastbound, westbound, southbound, and northbound traffic
counts for 20,362 intersections in Los Angeles. The second file contained traffic count
data for one mile increments on major highways throughout the Los Angeles area.
Information on freeway name, exit names, and average monthly and daily traffic counts
were available for 763 points.
Data on assisted-living facilities (ALF), adult day health care (ADHC), and skilled
nursing facilities (SNF) were available for download from the CA.gov website.
Information on ADHC and ALF was provided through the department of social services
link (https://secure.dss.cahwnet.gov/ccld/securenet/ccld_search/ccld_search.aspx),
while information for SNF was provided through the health facilities section
(https://hfcis.cdph.ca.gov/search.aspx). These files contained facility addresses and
capacity. Very small facilities—those with less than 6 beds—and those without current
licenses were excluded from our analysis. Overall, our analytical sample consisted of
858 facilities (192 ADHC, 400 SNF, and 266 ALF), see Figure 2.

154

2.2 Analysis
The high performance computing cluster from the University of Southern
California was used for analyses. From available information on city addresses, latitude
and longitude were estimated for the facilities and intersections using the STATA
module, Geocode3 (Bernhard et al., 2013). Next, the Haversine distances—a measure
of the distance between two points on a sphere—was estimated between the 858
facilities and 2,366,677 geographical points in the TIGER road file using the STATA
module Vincentry (Nichols et al., 2003). Given that node coordinates for roads were
calculated for the midpoint of the road, distances to facilities were adjusted based on
road type to reflect a more accurate measure of the distance between the facility and
the nearest side of a road. We subtracted 66 feet from distances between facilities and
major freeways, based on the assumption that most major freeways in Los Angeles
have eight lanes and two shoulders, leading to an overall diameter of 132 feet. Similarly,
52 feet was subtracted from distances between facilities and major freeways, given that
the average diameter for a Secondary Highway Class II is 104 feet. Finally, for
residential roads, whose diameters are estimated to be approximately 60 feet, we
subtracted 30 feet from their distances to nearby facilities.
Once the distance was estimated for every facility by road point combination,
only road points that were less than or equal to 500 feet away from any given facility
were kept (n=33,064). We choose to calculate distance between facilities and road
points rather than between facilities and intersections or highway exits given that the
later may not actually represent the closest road point to a facility. However, given that
traffic counts are only provided for intersections and freeway exits, traffic counts for road

155

points within 500 feet of a facility were estimated by matching them with the nearest
intersection or freeway exit on the same street and then assigning that traffic count to
them. This was done by calculating the Haversine distance between each road point
and each intersection or freeway exit which shared either a primary or cross street.
Based on these distances, the closest location with measured traffic count was selected
and that traffic count was assigned to the road point, assuming they were no more than
one mile apart. From this we were able to match 18,048 pairs (measured traffic points
and TIGER file road points). Those that were not matched mainly consisted of local
neighborhood roads or city streets which had relatively low levels of use; therefore,
traffic counts were not measured.
3. Results

Of the 858 elder care facilities we identified in the Los Angeles County area
(Figure 1), nearly half (n=392) were located within 500 feet of at least one road or
freeway for which we had measured traffic counts (Table 1). The 392 facilities included
187 skilled nursing facilities (SNF), 131 assisted living facilities (ALF), and 74 adult day
health care centers (ADHC). The majority of the SNF were relatively large facilities,
having on average 187 (s.d.=98.6) beds each—with 65 of the 187 facilities housing over
100 beds a piece. The ALF had approximately 108 beds on average—and over half
(n=70) of the 131 facilities had 100 or more beds. Finally, capacities for the ADHC
averaged about 60, with only 10 of the 74 facilities reaching capacities of 100 residents
or more. Just over 150 of the facilities were located in the city of Los Angeles, with the
next most represented cities being Long Beach and Van Nuys, which both housed
around 4% (n=16 and n=15, respectively) of the 392 facilities.

156

Overall, there were 1,013 roads or freeways within very close proximity (500 feet
or less) of one of the 392 elder care facility. On average the traffic counts for these
roads were approximately 43,000 cars per day, with some roads reaching average
traffic counts of over 500,000 cars per day. Approximately 5% of the roads in our
analytic sample were major freeways, 16% were Secondary Highway Class II, and 79%
were residential roads. On average, these roads were 332 feet from at least one facility.
Overall, highways had the closest proximity to facilities ( =318 feet, s.d.=126), followed
by residential roads ( =330 feet, s.d.=120), and finally freeways ( =398 feet, s.d.=86).
When definitions of high risk (traffic ≥100,000 cars per day) and moderate risk
(traffic 50,000-99,999 cars per day) were considered, we found that 54 of the 392
facilities were located within high pollution concentration areas—less than 500 feet from
a high risk roadway (Figure 2), whereas 51 facilities were located in close proximity to
roads with moderate risk traffic counts. The facilities located in these dangerous areas
also tended to be significantly larger (p=.008) than those not located in high pollution
areas. On average the facilities in close proximity to busy roadways had 115 beds,
whereas those in lower pollution locations only had 92 beds, on average.
Breakdown of facility type and size of the 54 facilities with 500 feet of a high
traffic roadway are given in table 2. Of the 54 facilities, 24 were SNF’s, 26 ALF’s, and 4
ADHC’s, with an average capacity of 106, 133, and 49, respectively. There were 51
facilities located within 500 feet of a moderate risk roadway. Of these 51 facilities, 18
were SNF’s, 19 ALF’s, and 14 ADHC’s, with an average capacity of 97, 125, and 65,
respectively. In total the facilities next to high risk roadways had a maximum capacity of

157

6,198, while the facilities located next to a moderate risk roadway could house up to
5,031 individuals.
A few facilities were located next to multiple major roads, exacerbating poolution
exposure (Table 3). Nine of the facilities were within 500 feet of two high risk roads
(traffic ≥100,000 cars per day), while three facilities were within 500 feet of one
moderate and two high risk traffic count roads. Additionally, one facility sat between two
moderate and one high risk road; three facilities sat between one moderate and one
high risk road, and 38 were located next to a single high risk road.
4. Discussion

The large body of evidence illustrating the harm of particulate matter inhalation
makes a clear argument for reducing exposure, but what exactly is a safe distance
away from sources of air pollution? In order to understand what is a safe distance away
from emission sources it is necessary to look at both epidemiological evidence of near
roadway exposure, and the spatial dispersion characteristics of the components of air
pollution.
There is a clear increase of mortality for older adults when residing near a major
roadway (defined as 100,000 cars per day). A Dutch study on individuals, aged 55-69,
recorded a relative risk of 1.41 for all-cause mortality when living within 100 meters (330
feet) of a freeway, or 50 meters (165 feet) of a major urban roadway (Hoek et al., 2002).
Another study, monitoring individuals in Ontario, Canada, with a median age of 63
observed a relative risk of 1.18 for mortality when living near a major roadway
(Finkelstein et al., 2004). These findings have been corroborated in the Nurses’ Health

158

Study, which found women residing near a major roadway to have a relative risk of 1.11
for myocardial infarction, and 1.05 for all-cause mortality (Hart et al., 2013). Aside from
the immediate risks of myocardial infarction, being in a high pollution environment at the
time of a myocardial infarction increases 10-year mortality rates as well (Rosenbloom et
al., 2013). In the current study, the 54 facilities located close to major roadways housed
approximately 6,000 individuals. Considering the minimal distance needed to drastically
reduce pollution concentration and lower mortality risk, this level of exposure is truly
unnecessary.
Understanding the spatial dispersion characteristics of the relevant components
of air pollution is also critical in determining appropriate safe distances, and in creating
informed policy solutions. There are three classes of particulate matter: PM10, PM2.5,
and PM0.1, known as coarse, fine and ultrafine particulate matter, respectively. Coarse
particulate matter (PM10) is any particle between 2.5um and 10um, whereas fine
particulate matter (PM2.5) are particles with a diameter between 2.5um and 0.1um, and
PM0.1 is any particle with a diameter less than 0.1um (1x10
-6
meters). For comparison
human hair is around 70um in diameter—seven hundred times larger than the largest
PM0.1. PM2.5 has been the most widely measured and recorded, however PM0.1 has
become increasingly recognized as being especially harmful (Schulz et al., 2005; Utell
et al., 2000; Weichenthal et al., 2012). Due to their small size, this component of
particulate pollution has gone relatively unnoticed, however their small size also makes
this component of air pollution especially dangerous (Schulz et al., 2005; Utell et al.,
2000; Weichenthal et al., 2012). It is this small size class, PM0.1, which is of special
importance when considering near roadway pollution.

159

Although adverse health effects of air pollution exposure diminish the farther an
individual is from a major roadway (Hoek et al., 2002; Gauderman et al., 2007; Peters et
al., 2004), distribution patterns show that neither PM2.5 nor PM10 decreases substantially
in that distance (Roorda-Knape et al., 1998; Zhou et al., 2007). Meanwhile, PM0.1
quickly returns to background levels, with pollution concentration closely resembling
observed health risks (Hoek et al., 2002; Zhu et al., 2002; Zhou et al., 2007).
Concentrations of PM0.1 show a 25-fold increase at 100 feet (30 meters) from a freeway
compared to ambient levels within the city (Zhu et al., 2002). The dilution of PM0.1 from
the source into background ambient levels occurs in an exponential decline with levels
generally returning to background at 1000 feet (300 meters) away from the source (Zhu
et al., 2002). A distance of 500 feet (150 meters) would reduce concentration by
approximately 80% (Zhu et al., 2002). Therefore 500 feet, which is the end of the
exponential decline for PM0.1 concentration, is recommended for proposed guidelines,
as it is an acceptable distance to balance functionality and safety. The spatial dispersion
characteristics of different classes of PM highlight the importance of a more precise
analysis of pollution exposure within a city.
Policy Responses to Protect Older Adults from Air Pollution:
The National Ambient Air Qualities Standards (NAAQS) has developed special
rules for air pollution exposure of certain vulnerable segments of the population. This
special attention is based upon observations that sensitivity to particulate matter can be
much higher in vulnerable populations compared to healthy individuals. The NAAQS
has developed non-binding recommendations for reducing exposure to pollution among
these groups. For example, the EPA drafted national guidelines dictating steps that

160

should be undertaken in the “school sitting process” (EPA 2007). The guide offers
recommendations for accurately measuring air quality in a new location, deciphering the
sources of the pollution, and choosing an appropriate site, all in an effort to reduce
pollution exposure to children. Although currently there are no national requirements,
several states have taken steps to reduce pollution exposure for children. California,
Indiana, and New Mexico all have minimum safe distance requirements, and many
other states have recommendations for placing schools away from busy roadways. One
example, California Senate Bill 352 (Chapter 668, statutes of 2003, effective January
2004) outlines regulations against school sitting within 500 feet (150 meters) of a
heavily traffic roadway, as defined by 50,000 cars/day in rural areas, and 100,000
cars/day in urban environments. This policy is based on studies showing a 70% drop-off
of particulate matter with a distance of 500 feet (150 meters) from a freeway (SCAQMD
2005). School sitting laws offer a valuable template to reduce pollution exposure by
enacting regulations that ensure the safe placement of facilities for older adults.
Findings from this paper suggest that similar regulations be extended to elder care
facilities, mandating a minimum of 500 feet away from major roadways for the
placement of any new facilities. Currently in Los Angeles County alone there are
approximately 6,000 individuals unnecessarily being exposed to these dangerous levels
of pollution, and the proposed policy offers a simple and affordable method to attenuate
this significant risk.
While enforced regulation offers one remedy, intermediate steps should also be
considered. Currently, public awareness of the effects of pollution exposure, and
especially its variance within a city, is minimal. Through educating the public, consumer

161

demand for safer, less polluted sites, could drive supply. Rating elder care facilities
based on air quality, either relative to city average or on an absolute scale, would
provide the public with an opportunity to make an educated decision. The rating system
could be modeled after the Department of Health’s ratings of A, B, and C for restaurant
sanitation standards. Facilities could be given a score based on the traffic count of the
largest roadway within 500 feet of the facility. The score could be in increments of
25,000, with three categories. These categories would be: 50,000 cars per day and
below being low risk, 50,000 to 100,000 being moderate risk, and over 100,000 cars per
day being high risk. This alternative to a mandatory requirement would be an extremely
low cost and readily available solution. Indeed, the data generated here is sufficient to
rank all the facilities within Los Angeles County. This method could rely upon consumer
demand to drive lower pollution exposure to future facilities. Combining the two
proposed policy solutions would create a mix of enforced legislation at the higher end,
and consumer demand at the lower end to drive reductions in pollution exposure.

Limitations
The patterns of spatial distribution of air pollution are not as clear cut in very
dense urban areas. In a study modeling spatial distribution of air pollution in Brooklyn,
New York, PM0.1 showed only a 15-20% decrease in concentration after the first 330
feet (100 meters) from the roadway (Zwack et al., 2011). This study illustrates a primary
problem for dense urban areas, where there is such a plethora of pollution coming from
multiple sources that there may be no “safe” distance. Thus, while locating facilities a
“safe” distance away from a major pollution source (approximately 500 feet) may reduce

162

negative health outcomes, the effects will not be as beneficial in very densely packed
urban areas compared to smaller or more diffuse cities.
Also, the proposed policy does not take into account the possibility of multiple
heavily trafficked roadways being located within close proximity of the facility, such as
being located on a busy street corner. This would create especially high pollution
conditions. The decision for the single roadway perspective was made in light of ease of
implementation of the proposed policy. It is believed that the relative advantage gained
by multiple roadway analysis would be relatively few, however the cost of added
complexity for such a policy to be implemented would be much greater. Analysis of the
current dataset showed only a small subset fell within this category of being located
within 500 feet of two moderately high traffic roads.
The present analysis provides only an estimate of individuals that are exposed to
high levels of air pollution. We calculate this estimate based on the maximum
occupancy of SNF, ALF, and ADHC. These facilities, while often quite full, are not
always at maximum capacity. For example, SNF’s have an average occupancy rate of
87% in California (OSHPD 2013). However, these facilities also experience significant
turnover, increasing the number of individuals that will be housed in a high risk facility
over the course of a year.
One additional factor related to placement is the assumption that being indoors
will reduce exposure. However, residing indoors does not confer protection from PM0.1
generated outdoors, because of its small size. The concentration of PM0.1 indoors is
similar to outdoor measurements; meaning that older individuals indoors are still at risk
for pollution exposure (Arhami et al., 2010). In fact, outdoor derived particulate matter

163

has been shown to be one of the most harmful components of indoor air pollution
(Delfino et al., 2008).
5. Conclusion
Air pollution is a ubiquitous environmental toxin that has been documented to
have numerous adverse health effects on all members of the population. However, this
burden of risk is shared unequally across the population, with children and older
individuals more prone to adverse health effects than young and middle aged adults. In
some areas, this problem has begun to be addressed in children, with state legislatures
enforcing a mandatory 500 feet away from a major roadway. In light of the available
scientific evidence, we propose that the same regulations be considered for facilities
and community services for older adults, including services such as skilled nursing
facilities, assisted living facilities, and adult day health care centers. Regulations
ensuring that these facilities are required to be a safe distance from the highest levels of
air pollution offers a low cost preventative approach to reduce morbidity and mortality
associated with pollution exposure in older individuals.
Funding T32AG0037

164

References
1. Adar, S., Gold, D., Coull, B., Schwartz, J., Stone, P., & Suh, H. (2007). Focused
Exposures to Airborne Traffic Particles and Heart Rate Variability in the
Elderly. Epidemiology. 18:95-103
2. Adar, SD, Sheppard, L, Vedal, S, Polak, JF, & Sampson, PD. (2013). Fine
particulate air pollution and the progression of carotid intima-medial thickness: a
prospective cohort study from the multi-ethnic study of atherosclerosis. PLoS med.
10:e1001430
3. Arhami, M., Minguillón, M. C., Polidori, A., Schauer, J. J., Delfino, R. J., & Sioutas,
C. (2010). Organic compound characterization and source apportionment of indoor
and outdoor quasi-ultrafine particulate matter in retirement homes of the Los
Angeles Basin. Indoor air. 20:17-30
4. Berglind, N., Bellander, T., Forastiere, F., Klot, S., Aalto, P., Elosua, R., … group,
for. (2009). Ambient Air Pollution and Daily Mortality Among Survivors of
Myocardial Infarction. Epidemiology. 20:110-8
5. Bernhard, S (2013). "GEOCODE3: Stata module to retrieve coordinates or
addresses from Google Geocoding API Version 3," Statistical Software
Components. S457650, Boston College Department of Economics.
6. Brook, R. D., Franklin, B., Cascio, W., Hong, Y., Howard, G., Lipsett, M., … Smith,
S. C. (2004). Air pollution and cardiovascular disease A statement for healthcare
professionals from the expert panel on population and prevention science of the
American Heart Association. Circulation, 109, 2655–2671.
7. Caiazzo, F., Ashok, A., Waitz, I., Yim, S., & Barrett, S. (2013). Air pollution and
early deaths in the United States. Part I: Quantifying the impact of major sectors in
2005. Atmos Environ, 79, 198-208
8. Center for Disease Control (2010). Summary Health Statistics for U.S. Adults:
National Health Interview Survey, Series 10, Number 252
9. Chen, H., Goldberg, M. S., Burnett, R. T., Jerrett, M., Wheeler, A. J., & Villeneuve,
P. J. (2013). Long-term exposure to traffic-related air pollution and cardiovascular
mortality. Epidemiology. 24:35-43
10. Delfino, R., Staimer, N., Tjoa, T., Polidori, A., Arhami, M., Gillen, D., … Sioutas, C.
(2008). Circulating Biomarkers of Inflammation, Antioxidant Activity, and Platelet
Activation Are Associated with Primary Combustion Aerosols in Subjects with
Coronary Artery Disease. Environ Health Perspec. 116:898-906
11. Environmental Protection Agency. Clean Air Fine Particle Implementation (2007).
Federal Register, 72-79
12. Finkelstein, M., Jerrett, M., & Sears, M. (2004). Traffic Air Pollution and Mortality
Rate Advancement Periods. Am J Epidemiol. 160:173-7
13. Gauderman WJ, Vora H, McConnell R, Berhane K, Gilliland F, Thomas D,
Lurmann F, Avol E, Kunzil N, Jerrett M, Peters J (2007). Effect of exposure to
traffic on lung development from 10 to 18 years of age: a cohort study. Lancet.
369:571-17
14. Goldberg, M., Burnett, R., & 3rd, B. (2001). Identification of persons with
cardiorespiratory conditions who are at risk of dying from the acute effects of
ambient air particles. Environ Health Perspect. 4:487-94

165

15. Hart, J., Rimm, E., Rexrode, K., & Laden, F. (2013). Changes in Traffic Exposure
and the Risk of Incident Myocardial Infarction and All-Cause
Mortality. Epidemiology. 24:734-42
16. Hoek, G., Brunekreef, B., Goldbohm, S., Fischer, P., & van den Brandt, P. A.
(2002). Association between mortality and indicators of traffic-related air pollution
in the Netherlands: a cohort study. Lancet,360, 1203–9.
17. Jerrett, M., Burnett, R., Ma, R., Pope, C., Krewski, D., Newbold, K., … Thun, M.
(2005). Spatial Analysis of Air Pollution and Mortality in Los
Angeles. Epidemiology. 16:727-36
18. Katsouyanni, K., Touloumi, G., Samoli, E., Gryparis, A., Tertre, A., Monopolis, Y.,
… Schwartz, J. (2001). Confounding and Effect Modification in the Short-Term
Effects of Ambient Particles on Total Mortality: Results from 29 European Cities
within the APHEA2 Project. Epidemiology. 12:521-31
19. Miller, K., Siscovick, D., Sheppard, L., Shepherd, K., Sullivan, J., Anderson, G., &
Kaufman, J. (2007). Long-Term Exposure to Air Pollution and Incidence of
Cardiovascular Events in Women. N Engl J Med. 356:447-58
20. Morgenstern, V., Zutavern, A., Cyrys, J., Brockow, I., Koletzko, S., Krämer, U., …
Group, L. (2008). Atopic diseases, allergic sensitization, and exposure to traffic-
related air pollution in children. Am J Respir Crit Care Med. 177:1331-7
21. Nichols A (2003). "VINCENTY: Stata module to calculate distances on the Earth's
surface," Statistical Software Components. S456815, Boston College Department
of Economics, revised 16 Feb 2007.
22. Office of Statewide Health Planning and Development (2013). Long-term care
facility annual utilization data. www.oshpd.ca.gov.
23. O'Neill MS, Veves A, Sarnat JA, Zanobetti A, Gold DR, Economides PA, Horton
ES, Schwartz J (2007). Air pollution and inflammation in type 2 diabetes: a
mechanism for susceptibility. Occup Environ Med, 64: 373-379.
24. Perera, F., Rauh, V., Whyatt, R., Tsai, W.-Y., Tang, D., Diaz, D., … Kinney, P.
(2006). Effect of Prenatal Exposure to Airborne Polycyclic Aromatic Hydrocarbons
on Neurodevelopment in the First 3 Years of Life among Inner-City
Children. Environ Health Perspec. 114:1287-92
25. Peters, A; von Klot, S; Heier, M; Trentinaglia, I; Hörmann, A; Wichmann, HE and
Löwel H (2004). Exposure to Traffic and the Onset of Myocardial Infarction. N Engl
J Med, 351:1721-1730.
26. Peters, A. (2012). Epidemiology: air pollution and mortality from diabetes
mellitus. Nature reviews. Endocrinology, 8, 706–7.
27. Pope, C., Burnett, R., Thun, M., Calle, E., Krewski, D., Ito, K., & Thurston, G.
(2002). Lung cancer, cardiopulmonary mortality, and long-term exposure to fine
particulate air pollution. JAMA, 287, 1132–41.
28. Roorda-Knape, MC; Janssen, NA; De Hartog, JJ; Van Vliet, PH; Harssema, H;
Brunekreef, B (1998). Air pollution from traffic in city districts near major
motorways. Atmos Environ. 32-11: 1921-1930.
29. Rosenbloom, J., Wilker, E., Mukamal, K., Schwartz, J., & Mittleman, M. (2012).
Residential Proximity to Major Roadway and 10-Year All-Cause Mortality After
Myocardial Infarction. Circulation. 125:2197-203

166

30. Ruckerl, R., Ibald-Mulli, A., & Koenig, W. (2006). Air pollution and markers of
inflammation and coagulation in patients with coronary heart disease. Am J Respir
Crit Care Med. 173:432-41
31. Schulz, H., Harder, V., Ibald-Mulli, A., Khandoga, A., Koenig, W., Krombach, F.,
Radykewicz, R., Stampfl, A., Thorand, B., Peters, A. (2005). Cardiovascular effects
of fine and ultrafine particles. J of Aerosol Med. 18: 1-22.
32. South Coast Air Quality Management District (2005). Air Quality Issues in School
Site Selection. Guidance Document.
33. Utell, M. J., & Frampton, M. W. (2000). Acute health effects of ambient air
pollution: the ultrafine particle hypothesis. J Aerosol Med. 13, 355–59.
34. Weichenthal, S (2012). Selected physiological effects of ultrafine particles in acute
cardiovascular morbidity. Environ Research. 115: 26-36
35. Zhou, Y., & Levy, J. I. (2007). Factors influencing the spatial extent of mobile
source air pollution impacts: a meta-analysis. BMC public health,7, 89.
36. Zhu, Y., Hinds, W., Kim, S., & Sioutas, C. (2002). Concentration and Size
Distribution of Ultrafine Particles Near a Major Highway. J Air Waste Manag Assoc.
52, 1032-42.
37. Zwack, L., Paciorek, C., Spengler, J., & Levy, J. (2011). Modeling spatial patterns
of traffic-related air pollutants in complex urban terrain. Environ Health Perspect.
119:852-9

167

Tables and Figures
Figure 1: Older Adult Care Facilities in Los Angeles County

Figure 1: A descriptive map of major roadways in Los Angeles combined with older
adult care facilities. Gray lines denote primary roads/highways, black lines are interstate
freeways, and black squares are older adult facilities.

168

Table 1: Descriptive statistics of all the facilities mapped in the study.
Descriptive characteristics for facilities (n=392) within 500 feet of a major road or
freeway with measured traffic
Characteristic Statistic
Facility Type (N)
ADHC 74
ALF 131
SNF 187
Capacity, mean (SD) 94.82 (58.95)
Distance from major road/freeway, mean feet (SD) 332.32 (120.69)
Traffic Count for major road/freeway, mean (SD) 43,459.2 (102,127.5)
Road type for major road/freeway (N)
Primary Road (Interstate) 5.0%
Secondary Road (Major/US Highway) 16.1%
Local Road (Residential Road) 78.87%
ADHC (Adult day health care), ALF (Assisted Living Facility), SNF (Skilled Nursing Facility)

169

Table 2: Number and capacity of facilities next to major roadways
Facility Type Number next to high
risk road (capacity)
Number next to
moderate risk road
(capacity)
Skilled nursing facility 24 (106) 18 (97)
Assisted living facility 26 (133) 19 (125)
Adult day health care
center
4 (49) 14 (65)
Total 6,198 5,031

Table 2: Depicts the number and average capacity (given in parenthesis) of facilities
located next to high risk and moderate risk roadways. Totals given is the maximum
number of people housed at one time in these 54 high risk, and 51 moderate risk
facilities.

170

Table 3: The number of Facilities with Combinations of High and Moderate Risk
Roads

Number of High Risk Roads

0 1 2 Total
Number of Moderate Risk
Roads

0 294 38 9 341
1 42 3 3 48
2 2 1 0 3
Total 338 42 12 392

Figure 2: High Risk Facilities in Los Angeles County

0
200
400
100 200 300 400 500 600
Distance to Nearest
Major Road (Feet)
Traffic for Nearest Major Road (1,000 Cars/Day)
ADHC ALF SNF

171

Figure 2: A plot of older adult facilities, measuring distance to a major roadway on the y-
axis and average traffic count per day on the x-axis. The size of the circle indicates size
of the facility, and color indicates type.

172

Concluding Remarks
Regardless of the ubiquity of environmental air pollution, much of its effects are
still unknown. Demonstrated here we have seen that nPM exposure causes changes in
the adult mouse hippocampus that mimic aging. Neurites and myelin basic protein are
decreased, while microglial activation is increased. Interestingly, instead of aggravating
age effects, it seems that older mice are less sensitive to air pollution exposure. Utilizing
a model system, we have demonstrated that glial activation from nPM is largely due to
TLR4 activation, and that a large portion of the response to nPM is shared with LPS
response. Finally, we saw a developmental vulnerability of the hippocampus to nPM
exposure, with nPM exposure causing hippocampal dependent memory deficits, and
reduced cell proliferation in the adult brain following developmental exposure.
What does this mean in light of the policy proposals of the final chapter?
Regardless of cognitive and inflammatory insensitivity to air pollution exposure, older
adults still face aggravated risk of morbidity and mortality following both acute and
chronic exposures. We demonstrated developmental vulnerability, but how exactly is a
pregnant mother supposed to integrate these findings into her daily routine? While we
are unfortunate enough to work with such a deeply entrenched pollutant as fossil fuels
and combustion engines, we are lucky enough because the findings of our field have
given us very real opportunities to reduce the societal burden of pollution exposure.
nPM, what we believe to be the most critical component of air pollution, is a fickle
environmental toxin. In as little as 500 feet from the source it is reduced by 80%. By
understanding the mechanisms of air pollution, and critical windows of exposure, it is
possible to greatly reduce the impact on society.

Abstract (if available)

Abstract What do we hope to achieve by researching air pollution? Similar to a clinical trial for new drug development, any new technology or chemical used in society must be rigorously evaluated for environmental and individual harm before being adopted. Unfortunately, this was not the case for air pollution, specifically pollution produced by combustion engines (traffic-related air pollution, TRAP). It has gained worldwide dominance as a mobile energy source, while to this day not being fully understood. Cardiovascular effects were publicly recognized as late as the mid twentieth century, while metabolic and cognitive effects are only recently being fully recognized. The purpose of air pollution work is two-fold. First, to fully elucidate the effects of air pollution exposure, in order to understand the personal and societal cost of air pollution. Second, this information must be used to mitigate harm as much as possible, whether through galvanizing policy interventions or shifting public demand. The work presented here addresses these aims by furthering the understanding of cognitive and metabolic effects of pollution exposure, elucidating potential mechanisms behind these effects, and proposing population level solutions to attenuate air pollution derived morbidity and mortality. Chapter one addresses the knowledge gap of aging interactions with TRAP exposure, and explores hippocampal neuronal and neuroinflammatory effects of TRAP exposure. Chapter two begins to elucidate the pathways activated by pollution exposure that lead to neuroinflammation. Chapter three explores the developmental role of pollution exposure, focusing on hippocampal dependent memory impairment and neuroinflammation. Finally, chapter four utilizes current knowledge of spatial dispersion patterns of airborne particulate to propose policy interventions to reduce the mortality risk from pollution exposure in vulnerable older adult populations. It is the hope of this researcher that the knowledge presented here will be valuable in furthering our understanding of air pollution, and in reducing the current societal burden imposed by pollution exposure.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Neuroinflammatory effects of urban traffic-derived nanoparticulate matter on neural systems

PDF

Air pollution neurotoxicity throughout the lifespan: studies on the mechanism of toxicity and interactions with effects of sex and genetic background

PDF

TLR4-mediated innate immune response and neuroinflammation: focus on APOE and obesity

PDF

Self-perceptions of Aging in the Context of Neighborhood and Their Interplay in Late-life Cognitive Health

PDF

Statistical algorithms for examining gene and environmental influences on human aging

PDF

Examination of the long-term psychosocial and functioning effects of a healthy living, life management behavior intervention for older adults

PDF

Intergenerational transmission of values and behaviors over the family life course

PDF

Computational approaches to identify genetic regulators of aging and late-life mortality

PDF

Investigating brain aging and neurodegenerative diseases through omics data

PDF

Life course implications of adverse childhood experiences: impacts on elder mistreatment, subjective cognitive decline, and caregivers' health

PDF

Toxicity of urban particulate matter: long-term health risks, influences of surrounding geography, and diurnal variation in chemical composition and the cellular oxidative stress response

PDF

Effects of western dietary factors during early life on glucose metabolism, the gut microbiome, and neurocognition

PDF

The evolution of decision-making quality over the life cycle: evidence from behavioral and neuroeconomic experiments with different age groups

PDF

Chronic eye disease epidemiology in the multiethnic ophthalmology cohorts of California study

Asset Metadata

Creator Woodward, Nicholas (author)

Core Title Neuroinflammation and the behavioral consequences of air pollution over the life course

School Leonard Davis School of Gerontology

Degree Doctor of Philosophy

Degree Program Gerontology

Publication Date 03/26/2017

Defense Date 08/17/2016

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Air pollution,astrocyte,behavior,brain,cognitive,microglia,neuroinflammation,neuron,Neuroscience,OAI-PMH Harvest,particulate matter

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Finch, Caleb (committee chair), Curran, Sean (committee member), Kanoski, Scott (committee member)

Creator Email nwoodwar@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-307646

Unique identifier UC11281132

Identifier etd-WoodwardNi-4815.pdf (filename),usctheses-c40-307646 (legacy record id)

Legacy Identifier etd-WoodwardNi-4815.pdf

Dmrecord 307646

Document Type Dissertation

Format application/pdf (imt)

Rights Woodward, Nicholas

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA