Prenatal and Postnatal Animal Models of
Immune Activation: Relevance to a Range
of Neurodevelopmental Disorders
Louise Harvey, Patricia Boksa
Department of Psychiatry, McGill University, Douglas Mental Health University Institute, Verdun,
Quebec, Canada H4H 1R3
Received 15 June 2012; accepted 18 June 2012
lished links between immune activation during the pre-
natal or early postnatal period and increased risk of
developing a range of neurodevelopment disorders in
later life. Animal models have been used to great effect
to explore the ramifications of immune activation during
gestation and neonatal life. A range of behavioral, neu-
rochemical, molecular, and structural outcome meas-
ures associated with schizophrenia, autism, cerebral
palsy, and epilepsy have been assessed in models of pre-
natal and postnatal immune activation. However, the
epidemiology-driven disease-first approach taken by
some studies can be limiting and, despite the wealth of
data, there is a lack of consensus in the literature as to
the specific dose, timing, and nature of the immunogen
that results in replicable and reproducible changes
related to a single disease phenotype. In this review, we
Epidemiological evidence has estab-
highlight a number of similarities and differences in
models of prenatal and postnatal immune activation
currently being used to investigate the origins of schizo-
phrenia, autism, cerebral palsy, epilepsy, and Parkin-
son’s disease. However, we describe a lack of synthesis
not only between but also within disease-specific models.
Our inability to compare the equivalency dose of immu-
nogen used is identified as a significant yet easily rem-
edied problem. We ask whether early life exposure to
infection should be described as a disease-specific or
general vulnerability factor for neurodevelopmental dis-
orders and discuss the implications that either classifica-
tion has on the design, strengths and limitations
of future experiments.
' 2012 Wiley Periodicals, Inc. Develop
Neurobiol 72: 1335–1348, 2012
activation; neurodevelopment; disease
Early life events can have significant effects on an
organism’s long-term health and wellbeing during
adulthood. Since the \Barker hypothesis" drew atten-
tion to the impact of prenatal nutrition on the risk of
subsequent adult-onset disorders such as diabetes, cardi-
ovascular disease, and hypertension (Barker and Mar-
tyn, 1992), this field, also known as the developmental
origins of health and disease, has expanded to demon-
strate how influential the prenatal environment is on a
wide range of adult health outcomes (Barker, 2004;
Sinclair et al., 2007; Sinclair and Singh, 2007). With
respect to the central nervous system (CNS), early
events that have been implicated in altering the trajec-
tory of neurodevelopment include pregnancy and birth
complications, maternal/neonatal exposures to nutri-
tional deficiency, stress, drugs or toxins, and postnatal
social deprivation (Schlotz and Phillips, 2009). Infec-
tion with resulting immune activation is another such
insult, and the focus of the current article is on how
animal models of prenatal and postnatal immune activa-
tion are being used to study the role of early life infec-
tion in the etiology of neurodevelopmental disorders.
Prenatal or early postnatal immune activation has
been implicated in a number of major neurodevelop-
mental disorders, including schizophrenia, autism, cere-
Correspondence to: P. Boksa (firstname.lastname@example.org).
' 2012 Wiley Periodicals, Inc.
Published online 25 June 2012 in Wiley Online Library
bral palsy, and epilepsy (Pakula et al., 2009; Brown
and Derkits, 2010; Landrigan, 2010). While disorders
like schizophrenia and autism appear to be uniquely
human, certain structural, molecular, and behavioral
abnormalities found in these human disorders can be
assessed in animals species commonly used for preclin-
ical research. It will be recalled that the etiologies of
disorders like schizophrenia, autism, or cerebral palsy
are multifactorial, likely involving a complex interplay
between genetic and environmental factors. Thus, it
should not be surprising if CNS effects produced by a
single risk factor in isolation, like prenatal infection,
are rather subtle, and may not mimic the entire spec-
trum of abnormalities characteristic of the disorder.
Nonetheless, the development of animal models allows
us to address specific questions about effects of expo-
sure to early life infection, e.g., the timing of the criti-
cal period of exposure to the immune activation; the
duration and severity of the inflammatory response; the
trajectory of neurodevelopmental changes during
juvenile and adult life; the mechanisms mediating
effects of immune activation on neurodevelopment;
and responses to potential therapeutic intervention.
The approach taken by many researchers in this
area is to focus on a specific disorder and to work with
a particular model of prenatal or postnatal immune
activation, which attempts to mimic the epidemiology
of the disorder, concentrating on assessing end points
characteristic of that disorder as outcome measures.
The aim of this review is to provide a brief overview
of the range of models of early life immune activation
currently being used within the context of various neu-
rodevelopmental diseases. After a brief introduction to
the immunogens commonly used to induce immune
activation, we will describe some of the models,
mainly in rodents, that are commonly used to examine
effects of prenatal or postnatal immune activation in
relation to schizophrenia, autism, cerebral palsy, epi-
lepsy, and other disorders. Rather than aiming to be
exhaustive, we will use a selection of examples to
compare and contrast abnormalities in disease-specific
endpoints observed in these models.
It is possible that a better integration of findings
across specific disease-based models might enhance
our understanding of the overall effects of early life
exposure to infection on neurodevelopment. There-
fore, in the course of the review, we hope to highlight
some of the similarities and differences between these
models and suggest that a broadening of the outcome
measures assessed in some already well-established
models or collaboration between researchers with
interests in different diseases might be a valuable
option to consider.
IMMUNOGENS USED TO MODEL
PRENATAL OR POSTNATAL IMMUNE
The most common immunogens used to induce
inflammation in pregnant mice and rats are lipopoly-
saccharide (LPS) and polyinosinic:polycytidylic acid
[poly(I:C)]. LPS, a component of the cell wall of
Gram negative bacteria, is a molecular immunogen
used to mimic a bacterial infection whereas poly(I:C),
a synthetic, double-stranded RNA, mimics a viral
infection. Both immunogens bind to toll-like recep-
tors [LPS to TLR-4, poly(I:C) to TLR-3], initiating a
signaling cascade that leads to activation of transcrip-
tion factors, such as nuclear factor kappa B (NFjB)
and subsequent transcription of genes coding for pro-
and anti-inflammatory mediators such as cytokines
[interleukin (IL)-1, tumor necrosis factor (TNF)-a,
IL-6, and interferons (IFNs)], chemokines, and com-
plement proteins. IL-6 then acts in the brain to induce
cyclooxygenase-2-mediated synthesis of prostaglan-
dins in the hypothalamus, which can mediate a fever
response (Roth et al., 2009). Although there are broad
similarities in some components of the proinflamma-
tory cytokine cascade induced by both LPS and
poly(I:C), there can be significant differences in the
magnitude of cytokine responses induced by these
two types of immunogens, as well as both quantita-
tive and qualitative differences in the cell types that
respond to activation, the profile of cytokine induc-
tion, and activation of downstream signaling cascades
(e.g., Bsibsi et al., 2006; Reimer et al., 2008; Figueir-
edo et al., 2009). Also, importantly for models of
inflammation during pregnancy, TLR3 and TLR4
may be differentially modulated by hormones, includ-
ing progesterone (Jones et al., 2010), whose levels
increase throughout pregnancy, peaking during late
pregnancy in humans and rats and falling just before
parturition in the rat (Bridges, 1984).
When comparing findings within models of early
life immune activation, one simple but important fac-
tor to consider is the dosage of immunogen used.
These models make widespread use of LPS as the
immune activator; however, despite its frequency of
use, it is difficult to compare LPS dosages across stud-
ies as it is known that the bioactivity of LPS per milli-
gram is dependent on the lot and serotype of the strain
of Escherichia coli (Ray et al., 1991; Akarsu and
Mamuk, 2007). This results in an inability to effec-
tively compare and contrast between models, particu-
larly in instances where authors do not provide details
of in vivo or in vitro bioactivity assays. Similar to the
1336 Harvey and Boksa
case for LPS, we have recently reported that different
batches of poly(I:C) obtained from the same supplier
differ widely in their cytokinogenic activity when
measured by induction of plasma IL-6 (Harvey and
Boksa, 2012). Given this variability, the standardiza-
tion of LPS and poly(I:C) dosages, through reporting
of meaningful bioactivity levels in each publication,
would represent a significant step forward in helping
to increase the transparency and reproducibility of
models of early life immune activation.
Although seemingly out of favor compared to LPS
and poly(I:C), some of the first models of prenatal
infection were designed using intranasal influenza
administration (Shi et al., 2003). The advantage of
using this immunogen is that, as the infection is live,
the time course of propagation of the immune activa-
tion is naturalistic. This advantage is also a limitation,
however, as the researcher loses some control over
the dosing and window of exposure. Viruses have
also been used in neonatal rodent models to examine
the neurodevelopmental consequences of congenital
or neonatal infection with specific agents such as
Borna disease virus, cytomegalovirus, and lympho-
cytic choriomeningitis (Hornig and Lipkin, 2001;
Barry et al., 2006; Bonthius and Perlman, 2007).
Intramuscular injection of turpentine has an advanta-
geous feature as a model of maternal immune activa-
tion because, in contrast to systemic LPS and
poly(I:C) administration, the turpentine remains
localized at the site of injection (Wusteman et al.,
1990). This obviates direct effects of the immunogen
on maternal organs and on the placental–fetal unit.
Intramuscular turpentine causes an increase in circu-
lating IL-6 mediated by local production of IL-1 and
TNF-a and a robust febrile response (Luheshi et al.,
Direct Administration of Cytokines
A key question is whether activation of specific com-
ponents of the immune system can account for the
effects of prenatal or postnatal infection on the CNS.
Models using direct administration of cytokines to
pregnant rodents during gestation or to postnatal
rodents have been used to address this. For example,
the recognition of IL-6 as a key mediator in the
inflammatory response has led to the development of
models of maternal immune activation in which
IL-6 is administered directly to the pregnant dam
(Samuelsson et al., 2006; Smith et al., 2007).
DISEASE-BASED MODELS OF
PRENATAL AND EARLY POSTNATAL
Epidemiological evidence has described an increased
risk of schizophrenia in the offspring of mothers
exposed to viral infections (influenza, measles, herpes
simplex virus type 2, rubella, and polio), bacterial
infections (pneumonia, respiratory infections, genital,
and reproductive infections including bacterial vagi-
nosis), and parasites (notably Toxoplasmosis gondii)
(reviewed by Brown and Derkits, 2010). Both first
and second trimester exposures have been implicated
in increasing risk for schizophrenia. Accordingly, the
most common models of maternal immune activation
used in relation to schizophrenia are those in which
pregnant rodents are systemically administered either
poly(I:C) or LPS. It is generally assumed that the first
two trimesters of human gestation are roughly equiv-
alent to the entire gestation period in a rat or mouse.
Thus, the time of exposure to immunogen used varies
widely across rodent models, from one or two injec-
tions of immunogen early or late in gestation to once
daily injections throughout the entire gestation period
(reviewed by Boksa, 2010).
Outcome measures assessed in these models
include a wide range of behavioral, structural, and
molecular parameters deemed to be relevant to schiz-
ophrenia. The details of many of these studies were
comprehensively reviewed by Boksa (2010). At a be-
havioral level, three of the most germane measures
that have been examined are prepulse inhibition (PPI)
of startle, latent inhibition, and attentional set shift-
ing, since these can be assessed in both rodents and
humans using very similar paradigms, and deficits in
these have been consistently found in schizophrenia
patients. PPI is the most frequently measured behav-
ioral outcome assessed in rodent offspring from pre-
natal immune activation models. Consistent PPI defi-
cits have been reported in mice administered prenatal
poly(I:C), LPS, influenza virus or IL-6, and in rats
administered prenatal poly(I:C) or LPS (Shi et al.,
2003; Fortier et al., 2007; Smith et al., 2007; Meyer
et al., 2008c; Wolff and Bilkey, 2008; Romero et al.,
2010; Howland et al., 2012). Deficits in latent inhibi-
tion, a more subtle measure of attention and informa-
tion processing, have also been reported in mice and
Early Life Infection and Neurodevelopment1337
rats prenatally treated with poly(I:C) and mice prena-
tally treated with IL-6, however, to date, are unre-
ported in prenatal LPS models (Zuckerman et al.,
2003a; Zuckerman and Weiner, 2003; Meyer et al.,
2006a; Smith et al., 2007). Recently, alterations in
attentional set shifting, indicative of perseveration,
have also been observed in male rat offspring prena-
tally administered poly(I:C) (Zhang et al., 2012).
Social behavior has been shown to be impaired in
maternal poly(I:C)- and influenza-treated mice (Shi et
al., 2003; Smith et al., 2007), and deficits have been
described in a variety of learning and memory para-
digms, including spatial learning in the Morris Water
Maze, for both mice and rats as a result of prenatal
treatment with LPS, poly(I:C), and IL-6 (Meyer et
al., 2006b; Ozawa et al., 2006; Samuelsson et al.,
2006; Coyle et al., 2009).
Historically, dysfunction of the dopaminergic sys-
tem has been considered a hallmark of schizophrenia
neurochemistry. As such, amphetamine-induced loco-
motor activity and changes in brain tyrosine hydroxy-
lase and dopamine metabolite content have been used
as markers of dopaminergic activity in animal mod-
els. Extensive changes in these measures have been
reported in many models of maternal inflammation;
in particular, an increase in amphetamine-induced
locomotion has been reported in both the offspring of
mice and rats prenatally treated with poly(I:C) and of
rats prenatally treated with LPS (Zuckerman et al.,
2003b; Fortier et al., 2004; Meyer et al., 2008c).
More recently, schizophrenia research has focused on
excitatory and inhibitory amino acid transmission,
with reported hypofunction of N-methyl-D-aspartate
(NMDA) receptors and GABAergic interneurons in
the frontal cortex and hippocampus in clinical and
postmortem schizophrenia populations (Coyle and
Tsai, 2004; Nakazawa et al., 2011). Alterations in short-
term plasticity, long-term potentiation, and NMDA/
(AMPA) receptor activity have been reported in rats
prenatally treated with LPS during late gestation (Lante
et al., 2008; Lowe et al., 2008; Roumier et al., 2008).
Maternal administration of LPS to rats during late gesta-
tion led to reductions in hippocampal reelin and
GAD67, markers of GABAergic interneurons, in the
offspring (Nouel et al., 2012) while administration of
poly(I:C) in early gestation in mice led to decreases in
hippocampal reelin and immunogen- and sex-specific
increases in GAD67 (Meyer et al., 2006b; Harvey and
At the molecular level, there appear to be fewer
parallels between described changes in brains of peo-
ple with schizophrenia and the detailed molecular
alterations in brains from animals subject to prenatal
immune challenge. Some studies have described
long-term changes in synaptophysin, brain-derived
neurotrophic factor, parvalbumin, and Akt following
maternal immune activation in rodents (Golan et al.,
2005; Romero et al., 2007a; Makinodan et al., 2008;
Meyer et al., 2008c; Romero et al., 2010); however,
these studies tend to come from single laboratories
and have not yet been replicated across species or
Epidemiological evidence also links prenatal immune
activation with an increased risk of developing autism
in later life (reviewed by Patterson, 2012). For some
time, this link was less established than that between
schizophrenia and prenatal immune activation. Recent
studies have described associations between viral
infection in the first trimester, bacterial infection in
the second trimester, and an increased risk for autism
in the offspring (Atladottir et al., 2010) as well as a
link between elevated amniotic levels of monocyte
chemotactic protein-1 (Abdallah et al., 2012), TNF-a
and ?b (Abdallah et al., in press), and the risk of
developing autism in later life. A retrospective study
looking at banked sera from mothers whose offspring
were later diagnosed with autism found increased lev-
els of IFN-c, IL-4, and IL-5 in maternal serum at
midgestation (Goines et al., 2011). Given these recent
findings, much of the existing research regarding au-
tism and prenatal immune activation has been
appended onto existing animal models of prenatal
infection, notably prenatal poly(I:C) and prenatal
influenza administration, which have been developed
with respect to schizophrenia epidemiology. This
approach has proved fruitful given there are signifi-
cant similarities between the disorders, with a recent
article even suggesting that it may be neuroinflamma-
tory events during early fetal development which
result in this shared pathogenesis (Meyer et al., 2011).
The most well-known characteristic of autism in
humans is behavioral alterations including sensory
and motor deficits, elevated anxiety and impaired
social interaction, communication, and emotional
processing (Wing, 1997). As with schizophrenia, peo-
ple with autism often have an impaired ability to filter
environmental stimuli, which has been assessed using
PPI (Perry et al., 2007). The PPI deficits and altera-
tions in social interaction reported in various models
of prenatal immune activation are described in the
section on \Schizophrenia". Three mouse studies,
one using maternal influenza administration (Shi et
al., 2003) and two using maternal poly(I:C) adminis-
tration (Smith et al., 2007; Malkova et al., 2012),
1338 Harvey and Boksa
have described a number of behavioral changes in the
offspring, including deficits in PPI, decreased social
behavior, increased markers of anxiety, ultrasonic
vocalization deficits, and repetitive behaviors, which
mimic the behavioral outcomes of people with au-
tism. It is interesting to note that all of these studies
used a viral immunogen at early to midgestational
time points to induce these behaviors; this correlates
very well with recent autism epidemiology and sug-
gests further that targeted research using these immu-
nogens, at these time points, is warranted.
Disrupted neurochemistry and neurotransmitter
function is also evident in brains of people with au-
tism (reviewed by Lam et al., 2006). Alterations in
the serotonergic system have been reported, and there
is conflicting evidence on the altered function of the
dopaminergic system. We have described some of the
changes in dopamine system function in animal mod-
els of prenatal immune activation in the section on
\Schizophrenia". Deficits in serotonin and its main
metabolite have been described in various brain
regions of mice prenatally treated with poly(I:C) at
E9 (Winter et al., 2009), mice prenatally treated with
influenza at E16 or E18 (Fatemi et al., 2008; Winter
et al., 2008), and rats prenatally treated with LPS at
E10.5 (Wang et al., 2009a). However, none of the
included any behavioral measures of anxiety or social
interaction, which could have strengthened our under-
standing of the potential relationship between neuro-
chemical and behavioral changes that manifest as a
result of prenatal immune activation.
Markers of hippocampal, amygdalar, and cerebel-
lar dysfunction, including small cell size and
increased cell density in these regions, a reduction in
Purkinje cell number, and, developmentally, abnor-
mally enlarged neurons in the cerebellum, are consid-
ered hallmarks of autistic brain morphology (Bauman
and Kemper, 2005; Amaral et al., 2008). There have
been consistent reports of alterations in cerebellar
and hippocampal Purkinje cell number, size, and den-
sity in mouse models using prenatal influenza at both
early and late gestational time points (Fatemi et al.,
1999, 2002, 2008, 2009; Shi et al., 2009) and some
isolated descriptions of cerebellar and hippocampal
changes using maternal poly(I:C) in the mouse and
maternal IL-6 in the rat (Samuelsson et al., 2006; Shi
et al., 2009).
Finally, alterations in the central and peripheral
immune system, such as changes in plasma and brain
cytokines, dysregulation of immune-related genes,
alterations in gastrointestinal tract permeability, and
some autoimmune-like responses, have been reported
in people with autism (White, 2003; Ashwood and
Van de Water, 2004). In animal models of prenatal
immune activation, there are conflicting reports as to
whether cytokines are induced in the fetal brain fol-
lowing exposure to the immunogen (Cai et al., 2000;
Urakubo et al., 2001; Gayle et al., 2004; Ashdown et
al., 2006; Meyer et al., 2006b). Regardless of this
uncertainty, most of these markers were measured
just hours following the inflammatory event. With
respect to long-term effects, exposure of rats to single
doses of LPS either prenatally or neonatally has been
reported to result in reduced induction of plasma
cytokines (IL-6, TNF-a, and IL-1b) in response to
LPS challenge in adolescence or adulthood (Hodyl et
al., 2007; Galic et al., 2009b; Beloosesky et al.,
2010). However, prenatal LPS or poly(I:C) treatment
of rats has also been reported to result in elevated ba-
sal plasma cytokine levels later in life (Carvey et al.,
2003; Romero et al., 2007a, 2010; Han et al., 2011).
Currently, less is known about the long-term effect of
early life infection on cytokine content in the brain,
although Bilbo et al. (2005) have observed that rats
injected neonatally (P4) with E. coli show exagger-
ated IL-1 responses to LPS in the hippocampus and
cortex at adulthood.
Important primary risk factors for cerebral palsy,
including preterm birth and small birth weight, can
be a result of, or induced by, maternal infection. Epi-
demiological evidence has also suggested that mater-
nal infection during the first and second trimesters,
intrauterine infections (such as chorioamnionitis) dur-
ing the third trimester and labor, and neonatal infec-
tions (such as meningitis) are significant risk factors
for cerebral palsy (Nelson and Willoughby, 2000;
Reddihough and Collins, 2003; Nelson, 2008). Thus,
in contrast to models for autism and schizophrenia,
common animal models of prenatal immune activa-
tion for cerebral palsy involve direct intrauterine
administration of LPS which often, in turn, induce
preterm births (Bell and Hallenbeck, 2002). Systemic
models of gestational LPS administration are also
used (Cai et al., 2000; Paintlia et al., 2004).
Deep, focused and diffuse white matter injury
(periventricular leukomalacia, PVL) and damage to
the cortex, basal ganglia, and thalamus are the most
common markers of cerebral palsy-specific brain
damage (van de Bor et al., 1989; Folkerth, 2005). In a
preterm model, sheep were administered intravenous
injections of LPS over 5 days during late gestation.
This treatment regime resulted in a brain pathology
characterized by diffuse subcortical damage and PVL
(Duncan et al., 2002). In a rat model, intrauterine
Early Life Infection and Neurodevelopment1339
administration of LPS at E15 resulted in early cortical
cell death and dysmyelination, similar to PVL
lesions, at 3 weeks of age in the offspring (Bell and
Hallenbeck, 2002). A preterm model of intrauterine
LPS administration in mouse caused significant fetal
brain injury (Burd et al., 2010; Ernst et al., 2010) as
did a similar \to term" model (Elovitz et al., 2011).
Diffuse PVL lesions in cerebral palsy brains are
initiated by loss of developing oligodendrocytes and
subsequent hypomyelination, which is mediated, in
part, by free radical-induced oxidative stress, gluta-
mate toxicity, and circulating cytokines (Kinney and
Back, 1998). Administration of LPS to rats during
late gestation caused increases in markers of oxida-
tive stress in the fetal brain (Paintlia et al., 2008) and
in glutamate-induced hydroxyl radical release in the
striatum of P14 offspring (Cambonie et al., 2004), to-
gether with an increase in circulating proinflamma-
tory cytokines in the maternal and fetal compartments
(Cai et al., 2000; Paintlia et al., 2004). Moreover, sys-
temic LPS administration at E18 has been shown to
cause loss of developing oligodendrocytes in fetal
brain, as well as decreases in oligodendrocyte number
and expression of myelin-related proteins in the post-
natal brain (Paintlia et al., 2008). In a somewhat dif-
ferent model of white matter injury, using direct
injection of LPS into the neonatal rat corpus cal-
losum, decreases in preoligodendrocytes and hypo-
myelination have also been observed, together with
increases in callosal radial diffusivity measured with
in vivo magnetic resonance imaging (Pang et al.,
2003; Lodygensky et al., 2010).
Importantly, hypoxia-ischemia is one of the causes
of PVL, and a number of studies using rodent models
have shown that LPS exposure exacerbates detrimen-
tal effects of hypoxia on developing brain (Eklind
et al., 2005; Wang et al., 2009b). Hypoxia-ischemia
is known to directly contribute to the loss of oligo-
dendrocytes in PVL lesions (Levison et al., 2001).
Thus, as hypoxia-ischemia is also a risk factor for
cerebral palsy, it is likely that the combination of
prenatal inflammation and a hypoxic event during
gestation would result in a significant increase in oli-
godendrocyte loss in the fetal brain, increasing the
likelihood of cerebral palsy. Preterm birth is another
risk factor for cerebral palsy, and there is a strong asso-
ciation between in utero infection during pregnancy
and preterm birth (Romero et al., 2007b). Prenatal
administration of LPS or live bacteria in a variety of
animal species are well-established models of preterm
birth, and there is a rich literature describing effects of
prenatal infection on fetal physiology and well-being
from this perspective (Edwards and Tan, 2006; Kemp
et al., 2010; Adams Waldorf et al., 2011).
Epilepsy and Acute Seizures
Postnatal rodent models have been used to investigate
the role of immune activation in the generation of
acute seizures and epilepsy. Acute seizures can de-
velop as a proximate consequence of both bacterial
and viral CNS infections. Mechanisms involved in the
generation of such seizures have been investigated in
models of bacterial meningitis involving intracisternal
injection of group B streptococcus in infant rats (Kim
et al., 1995; Kolarova et al., 2003) and models of viral
CNS infection involving intracisternal administration
of Theiler’s murine encephalomyelitis virus in young
mice (Libbey and Fujinami, 2011). The notion that
inflammation can also enhance susceptibility to acute
seizures induced by other agents or enhance CNS dam-
age induced by seizures has been supported by studies
in immature rats indicating that LPS administration
enhances rapid kindling and facilitates acute seizures
or seizure-induced neuronal injury induced by agents
such as lithium-pilocarpine, kainic acid, or glutaric
acid (Auvin et al., 2007, 2010b; Magni et al., 2011).
Several groups of investigators have provided in-
triguing evidence that infection or immune activation
early in postnatal life can lead to long-lasting increased
seizure susceptibility in adulthood. Stewart et al. (2010)
showed that young mice (P28–35) receiving intracister-
nal injection of Theiler’s murine encephalomyelitis vi-
rus developed spontaneous epileptic seizures 2–7
months after the injection. Galic et al. (2008) have dem-
onstrated that rat pups receiving an intraperitoneal injec-
tion of LPS at a critical period in development (P7, P14)
show enhanced convulsant-induced seizure susceptibil-
ity at adulthood. In a similar vein, Galic et al. (2009a)
have shown that intracerebroventricular injection of
poly(I:C) in P14 rat pups also causes increased seizure
susceptibility at adulthood, indicating that early expo-
sure to either bacterial or viral immunogens can contrib-
ute to this lasting susceptibility. Using a somewhat
different approach, Auvin et al. have examined effects
of LPS administration in combination with seizure-
inducing agents such as hyperthermia or lithium-pilocar-
pine in immature rats; these studies showed that pairing
LPS with lithium-pilocarpine resulted in more severe
spontaneous seizures at adulthood compared to lithium-
pilocarpine alone (Auvin et al., 2010a), while the com-
bination of LPS and hyperthermia caused a long-term
reduction in convulsant-induced seizure threshold
compared with hyperthermia alone (Auvin et al., 2009).
Interestingly, a subset of studies looking at prenatal
administration of LPS was conducted by researchers
1340 Harvey and Boksa
interested in developing a model of Parkinson’s dis-
ease. A single dose of LPS at E12.5 resulted in long-
lasting and progressive dopaminergic cell death and a
lifelong elevation in serum TNF-a (Carvey et al.,
2003). Subsequent articles from the same group of
investigators have replicated these findings, describ-
ing a decrease in dopaminergic cells in the substantia
nigra that is evident at P21 and persists until the ani-
mal is over one year old (Ling et al., 2002, 2006,
2009). These results appear to be in contrast to those
described in the section on \Schizophrenia", in which
prenatal treatment with LPS or poly(I:C) results in an
increase in amphetamine-induced locomotion, a
marker of dopaminergic activity. However, a close
changes in models of prenatal immune activation
indicates that, while some researchers report an
increase in dopamine cell numbers or levels of dopa-
mine and its metabolites in various brain regions
(Ozawa et al., 2006; Meyer et al., 2008a,b,c; Winter
et al., 2008), others report decreases (Bakos et al.,
2004; Romero et al., 2007a, 2010). As might be
expected, these changes in dopaminergic parameters
appear to depend on the specific model used, the
stage of pregnancy at which the immunogen was
administered, and on the brain region and postnatal
age examined. So, although schizophrenia researchers
tend to focus on alterations in amphetamine-induced
locomotion as a measure of dopamine dysfunction,
the relationship between behavioral markers of dopa-
remains to be established in these models. The exam-
ple of dopamine offers a good illustration of the diffi-
culty in synthesizing findings on prenatal and early
postnatal inflammation across laboratories, given the
multiplicity of models and outcome measures used by
General CNS Effects
There is epidemiological evidence linking fetal and
early life infections with general cognitive deficits
and CNS abnormalities, as well as alterations in
hypothalamic-pituitary-adrenal (HPA) axis function
and immune regulation (reviewed by Bale et al.,
2010). Researchers have developed animal models of
pre- and postnatal immune activation to explore these
effects, without necessarily specifying a disease
model. A wide number of immunogens are used,
including LPS, as well as specific cytokines and
actual disease strains, such as Borna virus.
In the absence of a specific disease context, gen-
eral effects of immune activation on cognitive per-
formance in later life has been the subject of investi-
gation in studies using, for example, rat models of
prenatal or neonatal exposure to LPS or replicating
E. coli (Bilbo et al., 2005; Harre et al., 2008; Hao
et al., 2010). Since cognitive deficits are also a promi-
nent feature of schizophrenia, effects of prenatal
immune activation on cognition are also reported in
studies focused on this disorder (Ozawa et al., 2006;
Bitanihirwe et al., 2010; Howland et al., 2012; Zhang
et al., 2012).
Animal modeling studies have also provided
strong evidence that neonatal immune challenge can
result in persistent modifications to the HPA axis and
downstream innate immune function, resulting in
altered susceptibility to inflammatory disease (e.g.,
arthritis) later in life (reviewed by Spencer et al.,
2011). However, the implications of this neonatal
immune activation are not limited to the innate
immune system. For example, administration of LPS
during early life can affect the adult response to cere-
bral ischemia in later life (Spencer et al., 2006), a
measure which would, in humans, translate to a lon-
ger recovery period after stroke.
researchers to create focused models of immune acti-
vation and to observe the overall effects of specific
cytokines on CNS function. In rats, daily subcutane-
ous administration of IL-1a, IL-2, IL-6, TNF-a, and
IFN-c between P2 and P10 resulted in cytokine-spe-
cific behavioral changes (Tohmi et al., 2004). At 3
weeks of age, rats treated with IL-2 displayed
enhanced locomotor and exploratory activity (Tohmi
et al., 2004). At 8 weeks of age, rats treated with IL-
1a exhibited an increase in startle response, a
decrease in PPI, and an increase in social behavior.
Intriguingly, rats treated with IL-6, TNF-a, or IFN-c
showed no behavioral abnormalities. The prenatal
intramuscular turpentine model also appears to have
some degree of cytokine specificity, since only IL-6
is increased in the circulation while production of IL-
1 and TNF-a is confined to the local site of inflamma-
tion (Luheshi et al., 1997). Administration of turpen-
tine to pregnant rats at E15 resulted in a decrease in
PPI, an increase in amphetamine-induced locomo-
tion, prolonged fear conditioning and increased la-
tency in a cued task in the Morris Water Maze as well
as increased tyrosine hydroxylase expression in the
nucleus accumbens of offspring (Aguilar-Valles
et al., 2010; Aguilar-Valles and Luheshi, 2011).
When turpentine was administered at E18, only the
fear conditioning response was altered in the off-
spring (Aguilar-Valles and Luheshi, 2011). Together,
these findings suggest that there are windows of
vulnerability during fetal and neonatal development
Early Life Infection and Neurodevelopment1341
during which an increase in specific cytokines may
produce long lasting changes in dopamine related
biology and behavior.
Models using disease strains of viruses have a two-
fold advantage; while they can be used to generally
model and explore the ramifications of early immune
activation, they can also help us understand shared
pathways between seemingly unrelated diseases. For
example, neonatal infection with Borna disease virus
originated as a model to examine effects of viral
infection on brain development. Borna disease virus
is a single-stranded, negative sense RNA virus known
to cause Borna disease. With increasing insight into
the neonatal Borna virus model, researchers observed
that many of the downstream changes were relevant
to features of autism, as well as schizophrenia (Hornig
and Lipkin, 2001), although there is no epidemiologi-
cal evidence to suggest that autism or schizophrenia
is caused by actual exposure to Borna virus. Neonatal
exposure results in animals that are stunted in growth,
have altered sleep-wake cycles, decreased PPI,
decreased play behavior, and altered developmental
progression of motor skills (Hornig and Lipkin,
2001). As such, this model provides an intriguing al-
ternative for researchers who are interested in study-
ing the effects of neonatal viral infection.
IS EARLY LIFE INFECTION A DISEASE-
SPECIFIC RISK FACTOR OR A GENERAL
IMPLICATIONS FOR ANIMAL MODELS
As we have seen with prenatal infection, the line of
approach to develop animal models of disease pro-
ceeds (roughly) as follows. Begin with a disease of
interest and identify a risk factor that is associated
with that disease, develop an animal model of the risk
factor, and attempt to determine if the animal model
mimics the disease pathology seen in humans. How-
ever, difficulties with this approach may arise
because the risk factor identified for the disease of in-
terest is also a risk factor for a number of other dis-
eases. Therefore, when modeling that risk factor in an
animal, you might expect to observe pathology
related to several different diseases. As an example,
both schizophrenia and autism are associated with
prenatal infection as a risk factor but it is difficult to
envisage how the same risk factor will lead to one of
these disorders versus the other. As a corollary then,
it is difficult to determine if a specific animal model
of prenatal immune activation is relevant to schizo-
phrenia or autism or both. To move forward, we may
consider two possibilities. The first is that we believe
we will be able to create separate animal models of
these disorders using the same risk factor, but that
they will depend on identifying a specific immunogen
(or class or dose of immunogen) at a specific critical
time point of administration in order to lead to either
schizophrenia-like or autism-like pathology. The sec-
ond is that we would consider prenatal immune acti-
vation as a general vulnerability factor for both schiz-
ophrenia and autism, which requires other factors to
be present, either as genetic or other environmental
insults, in order to develop the disease. Differentiat-
ing between these two possibilities allows us to iden-
tify two distinct experimental plans of action.
Early Life Infection as a Disease-Specific
A major difficulty in formulating a synthesis from the
current literature on the effects of prenatal or post-
natal infection on neurodevelopment arises because
of the wide range of models used in separate experi-
ments by different groups of investigators (Boksa,
2010). The models differ with respect to the immuno-
gen used, the dose and number of administrations of
immunogen, the time during gestation or postnatal
life when immunogen is administered, the postnatal
age and sex of offspring examined, and the type of
outcome measure quantified.
Working on the idea that exposure to specific
immunogens at specific time points of gestation will
predispose toward different diseases, e.g., schizophre-
nia versus autism, fundamental information describ-
ing the effects of multiple immunogens, at multiple
time points during gestation, on the same outcome
measures, within the same study, is needed (see, for
example, Meyer et al., 2006b; Fortier et al., 2007;
Meyer et al., 2008c; Harvey and Boksa, 2012). These
types of experiments will help to discern if there are
differing effects of different immunogens or different
doses of immunogens and whether critical time win-
dows of administration are required to observe differ-
ential outcomes. These results will also help to
emphasize which specific parameters of prenatal
immune activation are essential to enable other inves-
tigators to replicate observed effects—as in many
fields of research, independent replication of results is
not a strong feature of the literature on prenatal infec-
tion and neurodevelopment. [A notable exception is
the consistent deficits in PPI following prenatal
poly(I:C), LPS, or influenza, which have been observed
by numerous independent laboratories, see the section
on \Schizophrenia".] Conversely, it will also be im-
portant to determine if a single protocol of immunogen
1342 Harvey and Boksa
administration gives rise to outcome measures specif-
ically related to one disorder (e.g., schizophrenia) but
not to others (e.g., autism). This would require that
researchers interested in a particular disorder not only
quantify outcome measures related to their disorder
of interest but also examine a wider range of CNS
outcome measures in order to examine the specificity
of effects in their model. These approaches should
help the field to form consensus regarding some of
the most integral questions in the prenatal inflamma-
tion literature, for example, how does prenatal infec-
tion change dopamine and other neurotransmitter bio-
chemistry, and do multiple behavioral phenotypes
converge, or not, within the same model?
A further potentially fruitful avenue of investigation
for researchers interested in neurodevelopmental disor-
ders would be to consider effects of prenatal infection
on systems other than the CNS that could influence
their disease of interest. For example, there is human
epidemiological evidence suggesting an association
between exposure to measles during gestation and the
onset of Crohn’s disease/inflammatory bowel disease
in offspring in later life (Ekbom et al., 1994, 1996;
Nielsen et al., 1998; Pardi et al., 1999). Hence, it has
been suggested that prenatal infection may be a risk
factor for both inflammatory bowel disease and autism,
and in fact, there is evidence that a subset of autistic
children experience gastrointestinal abnormalities,
which could be indicative of a decrease in mucosal in-
tegrity (Levy et al., 2007; de Magistris et al., 2010).
Recent work has highlighted that the microbiota of the
gut can play an important role in CNS function and
reactivity (Neufeld and Foster, 2009). However, to
date, few animal models of prenatal infection have
considered the role of the gut. In the rat, neonatal ex-
posure to LPS has been shown to result in a more
severe response to induced experimental colitis
(Spencer et al., 2007). No studies investigating effects
of prenatal infection on neurodevelopment have
included examination of the gut. Although this
research is ongoing in humans, the established models
of prenatal immune activation would provide an
obvious and accessible model in which to test the rela-
tionship between gut microbiota and behaviors rele-
vant to autism. One could envisage how investigations
of this sort might conceivably lead to a description of
specific situations in which prenatal infection could
lead to autism, as opposed to other CNS disorders.
Early Life Infection as a General
If we consider that prenatal infection is a general vul-
nerability factor conferring increased susceptibility to
a range of different neurodevelopmental disorders,
then it is more important than ever to create multifac-
torial models, combining either multiple environmen-
tal risk factors or an environmental risk factor on a
background of genetic risk. This approach has been
used by groups examining the interactive effects of
prenatal immune activation together with exposure to
environmental neurotoxins (Ling et al., 2004a,b,
2006) or effects of prenatal immune activation in
interaction with genes implicated in the etiology of
disorders such as schizophrenia and Parkinson’s dis-
ease (Granholm et al., 2010; Ibi et al., 2010; Ehninger
et al., 2012; Vuillermot et al., 2012). In theory, such
investigations may allow us to identify which combi-
nations of etiological factors are more likely to result
in one particular disorder (e.g., schizophrenia) versus
another (e.g., autism). Of course, this type of multi-
factorial animal model adds another layer of com-
plexity onto an already complicated prenatal infection
model and presents further challenges in attempts to
summarize how prenatal infection affects neurodevel-
opment using a synthesis of findings from various
Prenatal and early postnatal infections have been
associated with increased risk for a number of neuro-
developmental disorders. Animal models of prenatal
and early postnatal immune activation have contrib-
uted substantially toward indicating that this associa-
tion may be causal. However, for the most part, ani-
mal models have not yet shed light on the question of
which specific conditions are required in order for
early life infection to contribute to development of
one particular disorder as opposed to another (e.g.,
schizophrenia or autism?). We have discussed the di-
chotomy that exists between the belief that a specific
CNS disorder resulting from early life infection may
be determined by exposure to a particular class of
pathogen at a critical time point in fetal or postnatal
development, and the theory that the combination of
early life infection with specific additional genetic
and environmental insults will determine the neuro-
developmental outcome. In fact, these two ideas need
not be thought of as mutually exclusive. One can
readily envisage the scenario where the actual etiol-
ogy of a complex neurodevelopmental disorder like
schizophrenia or autism might require both exposures
to a specific infection at a specific time in early devel-
opment in combination with further environmental
and genetic \hits." A consideration of these two
approaches will be useful when designing robust and
Early Life Infection and Neurodevelopment1343
reproducible animal studies that can be integrated
into our current knowledge and contribute to the pro-
gression of the prenatal and postnatal immune activa-
Abdallah MW, Larsen N, Grove J, Norgaard-Pedersen B,
Thorsen P, Mortensen EL, Hougaard DM. Amniotic fluid
inflammatory cytokines: potential markers of immuno-
logic dysfunction in autism spectrum disorders. World
J Biol Psychiatry, in press.
Abdallah MW, Larsen N, Grove J, Norgaard-Pedersen B,
Thorsen P, Mortensen EL, Hougaard DM. 2012. Amni-
otic fluid chemokines and autism spectrum disorders: an
exploratory study utilizing a Danish Historic Birth
Cohort. Brain Behav Immun 26:170–176.
Adams Waldorf KM, Rubens CE, Gravett MG. 2011. Use of
nonhuman primate models to investigate mechanisms of
infection-associated preterm birth. BJOG 118:136–144.
Aguilar-Valles A, Flores C, Luheshi GN. 2010. Prenatal
function in the adult offspring in rat: relevance for schiz-
ophrenia. PLoS One 5:e10967.
Aguilar-Valles A, Luheshi GN. 2011. Alterations in cogni-
tive function and behavioral response to amphetamine
induced by prenatal inflammation are dependent on the
stage of pregnancy. Psychoneuroendocrinology 36:634–
Akarsu ES, Mamuk S. 2007. Escherichia coli lipopolysac-
charides produce serotype-specific hypothermic response
in biotelemetered rats. Am J Physiol Regul Integr Comp
Amaral DG, Schumann CM, Nordahl CW. 2008. Neuroan-
atomy of autism. Trends Neurosci 31:137–145.
Ashdown H, Dumont Y, Ng M, Poole S, Boksa P, Luheshi
GN. 2006. The role of cytokines in mediating effects of
prenatal infection on the fetus: implications for schizo-
phrenia. Mol Psychiatry 11:47–55.
Ashwood P, Van de Water J. 2004. A review of autism and
the immune response. Clin Dev Immunol 11:165–174.
Atladottir HO, Thorsen P, Ostergaard L, Schendel DE,
Lemcke S, Abdallah M, Parner ET. 2010. Maternal infec-
tion requiring hospitalization during pregnancy and
autism spectrum disorders. J Autism Dev Disord 40:
Auvin S, Mazarati A, Shin D, Sankar R. 2010a. Inflamma-
tion enhances epileptogenesis in the developing rat brain.
Neurobiol Dis 40:303–310.
Auvin S, Porta N, Nehlig A, Lecointe C, Vallee L, Bordet
R. 2009. Inflammation in rat pups subjected to short
hyperthermic seizures enhances brain long-term excit-
ability. Epilepsy Res 86:124–130.
Auvin S, Shin D, Mazarati A, Nakagawa J, Miyamoto J,
Sankar R. 2007. Inflammation exacerbates seizure-
induced injury in the immature brain. Epilepsia 48(suppl
Auvin S, Shin D, Mazarati A, Sankar R. 2010b. Inflamma-
tion induced by LPS enhances epileptogenesis in imma-
ture rat and may be partially reversed by IL1RA. Epilep-
sia 51(suppl 3):34–38.
Bakos J, Duncko R, Makatsori A, Pirnik Z, Kiss A, Jezova
D. 2004. Prenatal immune challenge affects growth,
behavior, and brain dopamine in offspring. Ann N Y
Acad Sci 1018:281–287.
Bale TL, Baram TZ, Brown AS, Goldstein JM, Insel TR,
McCarthy MM, Nemeroff CB, et al. 2010. Early life pro-
gramming and neurodevelopmental disorders. Biol Psy-
Barker DJ. 2004. The developmental origins of adult dis-
ease. J Am Coll Nutr 23:588S–595S.
Barker DJ, Martyn CN. 1992. The maternal and fetal ori-
gins of cardiovascular disease. J Epidemiol Community
Barry PA, Lockridge KM, Salamat S, Tinling SP, Yue Y,
Zhou SS, Gospe SM Jr., et al. 2006. Nonhuman primate
models of intrauterine cytomegalovirus infection. ILAR J
Bauman ML, Kemper TL. 2005. Neuroanatomic observa-
tions of the brain in autism: a review and future direc-
tions. Int J Dev Neurosci 23:183–187.
Bell MJ, Hallenbeck JM. 2002. Effects of intrauterine
inflammation on developing rat brain. J Neurosci Res
Beloosesky R, Maravi N, Weiner Z, Khatib N, Awad N,
Boles J, Ross MG, et al. 2010. Maternal lipopolysaccha-
ride-induced inflammation during pregnancy programs
impaired offspring innate immune responses. Am J
Obstet Gynecol 203:185 e181–e184.
Bilbo SD, Biedenkapp JC, Der-Avakian A, Watkins LR,
Rudy JW, Maier SF. 2005. Neonatal infection-induced
memory impairment after lipopolysaccharide in adult-
hood is prevented via caspase-1 inhibition. J Neurosci
Bitanihirwe BK, Weber L, Feldon J, Meyer U. 2010.
challenge in mice correlates with prefrontal cortical
AKT1 deficiency. Int J Neuropsychopharmacol 13:981–
Boksa P. 2010. Effects of prenatal infection on brain devel-
opment and behavior: a review of findings from animal
models. Brain Behav Immun 24:881–897.
Bonthius DJ, Perlman S. 2007. Congenital viral infections
of the brain: lessons learned from lymphocytic chorio-
meningitis virus in the neonatal rat. PLoS Pathog 3:e149.
Bridges RS. 1984. A quantitative analysis of the roles of
dosage, sequence, and duration of estradiol and proges-
terone exposure in the regulation of maternal behavior in
the rat. Endocrinology 114:930–940.
Brown AS, Derkits EJ. 2010. Prenatal infection and schizo-
phrenia: a review of epidemiologic and translational
studies. Am J Psychiatry 167:261–280.
Bsibsi M, Persoon-Deen C, Verwer RW, Meeuwsen S,
Ravid R, Van Noort JM. 2006. Toll-like receptor 3 on
adult human astrocytes triggers production of neuropro-
tective mediators. Glia 53:688–695.
1344Harvey and Boksa
Burd I, Bentz AI, Chai J, Gonzalez J, Monnerie H, Le Roux
PD, Cohen AS, et al. 2010. Inflammation-induced pre-
term birth alters neuronal morphology in the mouse fetal
brain. J Neurosci Res 88:1872–1881.
Cai Z, Pan ZL, Pang Y, Evans OB, Rhodes PG. 2000. Cyto-
kine induction in fetal rat brains and brain injury in neo-
natal rats after maternal lipopolysaccharide administra-
tion. Pediatr Res 47:64–72.
Cambonie G, Hirbec H, Michaud M, Kamenka JM, Barba-
nel G. 2004. Prenatal infection obliterates glutamate-
related protection against free hydroxyl radicals in neo-
natal rat brain. J Neurosci Res 75:125–132.
Carvey PM, Chang Q, Lipton JW, Ling Z. 2003. Prenatal
exposure to the bacteriotoxin lipopolysaccharide leads to
long-term losses of dopamine neurons in offspring: a
potential, new model of Parkinson’s disease. Front Bio-
Coyle JT, Tsai G. 2004. NMDA receptor function, neuro-
plasticity, and the pathophysiology of schizophrenia. Int
Rev Neurobiol 59:491–515.
Coyle P, Tran N, Fung JN, Summers BL, Rofe AM. 2009.
Maternal dietary zinc supplementation prevents aberrant
behaviour in an object recognition task in mice offspring
exposed to LPS in early pregnancy. Behav Brain Res
de Magistris L, Familiari V, Pascotto A, Sapone A, Frolli
A, Iardino P, Carteni M, et al. 2010. Alterations of the in-
testinal barrier in patients with autism spectrum disorders
and in their first-degree relatives. J Pediatr Gastroenterol
Duncan JR, Cock ML, Scheerlinck JP, Westcott KT,
McLean C, Harding R, Rees SM. 2002. White matter
injury after repeated endotoxin exposure in the preterm
ovine fetus. Pediatr Res 52:941–949.
Edwards AD, Tan S. 2006. Perinatal infections, prematur-
ity and brain injury. Curr Opin Pediatr 18:119–124.
Ehninger D, Sano Y, de Vries PJ, Dies K, Franz D, Gesch-
wind DH, Kaur M, et al. 2012. Gestational immune acti-
vation and Tsc2 haploinsufficiency cooperate to disrupt
fetal survival and may perturb social behavior in adult
mice. Mol Psychiatry 17:62–70.
Ekbom A, Daszak P, Kraaz W, Wakefield AJ. 1996.
Crohn’s disease after in-utero measles virus exposure.
Ekbom A, Wakefield AJ, Zack M, Adami HO. 1994. Peri-
natal measles infection and subsequent Crohn’s disease.
Eklind S, Mallard C, Arvidsson P, Hagberg H. 2005. Lipo-
polysaccharide induces both a primary and a secondary
phase of sensitization in the developing rat brain. Pediatr
Elovitz MA, Brown AG, Breen K, Anton L, Maubert M,
Burd I. 2011. Intrauterine inflammation, insufficient to
induce parturition, still evokes fetal and neonatal brain
injury. Int J Dev Neurosci 29:663–671.
Ernst LM, Gonzalez J, Ofori E, Elovitz M. 2010. Inflamma-
tion-induced preterm birth in a murine model is associ-
ated with increases in fetal macrophages and circulating
erythroid precursors. Pediatr Dev Pathol 13:273–281.
Fatemi SH, Earle J, Kanodia R, Kist D, Emamian ES, Pat-
terson PH, Shi L, et al. 2002. Prenatal viral infection
leads to pyramidal cell atrophy and macrocephaly in
adulthood: implications for genesis of autism and schizo-
phrenia. Cell Mol Neurobiol 22:25–33.
Fatemi SH, Emamian ES, Kist D, Sidwell RW, Nakajima
K, Akhter P, Shier A, et al. 1999. Defective corticogene-
sis and reduction in Reelin immunoreactivity in cortex
and hippocampus of prenatally infected neonatal mice.
Mol Psychiatry 4:145–154.
Fatemi SH, Folsom TD, Reutiman TJ, Abu-Odeh D, Mori
S, Huang H, Oishi K. 2009. Abnormal expression of
myelination genes and alterations in white matter frac-
tional anisotropy following prenatal viral influenza infec-
tion at E16 in mice. Schizophr Res 112:46–53.
Fatemi SH, Reutiman TJ, Folsom TD, Huang H, Oishi K,
Mori S, Smee DF, et al. 2008. Maternal infection leads to
abnormal gene regulation and brain atrophy in mouse
offspring: implications for genesis of neurodevelopmen-
tal disorders. Schizophr Res 99:56–70.
Figueiredo MD, Vandenplas ML, Hurley DJ, Moore JN.
2009. Differential induction of MyD88- and TRIF-de-
pendent pathways in equine monocytes by Toll-like re-
ceptor agonists. Vet Immunol Immunopathol 127:125–
Folkerth RD. 2005. Neuropathologic substrate of cerebral
palsy. J Child Neurol 20:940–949.
Fortier ME, Joober R, Luheshi GN, Boksa P. 2004. Mater-
nal exposure to bacterial endotoxin during pregnancy
enhances amphetamine-induced locomotion and startle
responses in adult rat offspring. J Psychiatr Res 38:335–
Fortier ME, Luheshi GN, Boksa P. 2007. Effects of prenatal
infection on prepulse inhibition in the rat depend on the
nature of the infectious agent and the stage of pregnancy.
Behav Brain Res 181:270–277.
Galic MA, Riazi K, Heida JG, Mouihate A, Fournier NM,
Spencer SJ, Kalynchuk LE, et al. 2008. Postnatal inflam-
mation increases seizure susceptibility in adult rats. J
Galic MA, Riazi K, Henderson AK, Tsutsui S, Pittman QJ.
2009a. Viral-like brain inflammation during development
causes increased seizure susceptibility in adult rats. Neu-
robiol Dis 36:343–351.
Galic MA, Spencer SJ, Mouihate A, Pittman QJ. 2009b.
Postnatal programming of the innate immune response.
Integr Comp Biol 49:237–245.
Gayle DA, Beloosesky R, Desai M, Amidi F, Nunez SE,
Ross MG. 2004. Maternal LPS induces cytokines in the
amniotic fluid and corticotropin releasing hormone in the
fetal rat brain. Am J Physiol Regul Integr Comp Physiol
Goines PE, Croen LA, Braunschweig D, Yoshida CK,
Grether J, Hansen R, Kharrazi M, et al. 2011. Increased
midgestational IFN-gamma, IL-4 and IL-5 in women
bearing a child with autism: a case-control study. Mol
Golan HM, Lev V, Hallak M, Sorokin Y, Huleihel M.
2005. Specific neurodevelopmental damage in mice off-
Early Life Infection and Neurodevelopment1345
spring following maternal inflammation during preg-
nancy. Neuropharmacology 48:903–917.
Granholm AC, Zaman V, Godbee J, Smith M, Ramadan R,
Umphlet C, Randall P, et al. 2010. Prenatal LPS
increases inflammation in the substantia nigra of GDNF
heterozygous mice. Brain Pathol 21:330–348.
Han X, Li N, Meng Q, Shao F, Wang W. 2011. Maternal
increases serum tumor necrosis factor-alpha in offspring.
Hao LY, Hao XQ, Li SH, Li XH. 2010. Prenatal exposure
to lipopolysaccharide results in cognitive deficits in age-
increasing offspring rats. Neuroscience 166:763–770.
Harre EM, Galic MA, Mouihate A, Noorbakhsh F, Pittman
QJ. 2008. Neonatal inflammation produces selective
behavioural deficits and alters N-methyl-D-aspartate re-
ceptor subunit mRNA in the adult rat brain. Eur J Neuro-
Harvey L, Boksa P. 2012. A stereological comparison of
GAD67 and reelin expression in the hippocampal stratum
oriens of offspring from two mouse models of maternal
inflammation during pregnancy. Neuropharmacology
Hodyl NA, Krivanek KM, Lawrence E, Clifton VL, Hodg-
son DM. 2007. Prenatal exposure to a pro-inflammatory
stimulus causes delays in the development of the innate
immune response to LPS in the offspring. J Neuroimmu-
Hornig M, Lipkin WI. 2001. Infectious and immune factors
in the pathogenesis of neurodevelopmental disorders: ep-
idemiology, hypotheses, and animal models. Ment
Retard Dev Disabil Res Rev 7:200–210.
Howland JG, Cazakoff BN, Zhang Y. 2012. Altered object-
in-place recognition memory,
locomotor activity in the offspring of rats exposed to a vi-
ral mimetic during pregnancy. Neuroscience 201:184–198.
Ibi D, Nagai T, Koike H, Kitahara Y, Mizoguchi H, Niwa
M, Jaaro-Peled H, et al. 2010. Combined effect of neona-
tal immune activation and mutant DISC1 on phenotypic
changes in adulthood. Behav Brain Res 206:32–37.
Jones LA, Kreem S, Shweash M, Paul A, Alexander J, Rob-
erts CW. 2010. Differential modulation of TLR3- and
TLR4-mediated dendritic cell maturation and function by
progesterone. J Immunol 185:4525–4534.
Kemp MW, Saito M, Newnham JP, Nitsos I, Okamura K,
Kallapur SG. 2010. Preterm birth, infection, and inflam-
mation advances from the study of animal models.
Reprod Sci 17:619–628.
Kim YS, Sheldon RA, Elliott BR, Liu Q, Ferriero DM,
Tauber MG. 1995. Brain injury in experimental neonatal
meningitis due to group B streptococci. J Neuropathol
Exp Neurol 54:531–539.
Kinney HC, Back SA. 1998. Human oligodendroglial de-
velopment: relationship to periventricular leukomalacia.
Semin Pediatr Neurol 5:180–189.
Kolarova A, Ringer R, Tauber MG, Leib SL. 2003. Block-
ade of NMDA receptor subtype NR2B prevents seizures
but not apoptosis of dentate gyrus neurons in bacterial
meningitis in infant rats. BMC Neurosci 4:21.
prepulse inhibition, and
Lam KS, Aman MG, Arnold LE. 2006. Neurochemical cor-
relates of autistic disorder: a review of the literature. Res
Dev Disabil 27:254–289.
Landrigan PJ. 2010. What causes autism? Exploring the
environmental contribution. Curr Opin Pediatr 22:219–
Lante F, Meunier J, Guiramand J, De Jesus Ferreira MC,
Cambonie G, Aimar R, Cohen-Solal C, et al. 2008. Late
N-acetylcysteine treatment prevents the deficits induced
in the offspring of dams exposed to an immune stress
during gestation. Hippocampus 18:602–609.
Levison SW, Rothstein RP, Romanko MJ, Snyder MJ,
Meyers RL, Vannucci SJ. 2001. Hypoxia/ischemia
depletes the rat perinatal subventricular zone of oligoden-
drocyte progenitors and neural stem cells. Dev Neurosci
Levy SE, Souders MC, Ittenbach RF, Giarelli E, Mulberg
AE, Pinto-Martin JA. 2007. Relationship of dietary
intake to gastrointestinal symptoms in children with
autistic spectrum disorders. Biol Psychiatry 61:492–497.
Libbey JE, Fujinami RS. 2011. Neurotropic viral infections
leading to epilepsy: focus on Theiler’s murine encephalo-
myelitis virus. Future Virol 6:1339–1350.
Ling Z, Chang QA, Tong CW, Leurgans SE, Lipton JW,
Carvey PM. 2004a. Rotenone potentiates dopamine neu-
ron loss in animals exposed to lipopolysaccharide prena-
tally. Exp Neurol 190:373–383.
Ling Z, Gayle DA, Ma SY, Lipton JW, Tong CW, Hong JS,
Carvey PM. 2002. In utero bacterial endotoxin exposure
causes loss of tyrosine hydroxylase neurons in the post-
natal rat midbrain. Mov Disord 17:116–124.
Ling Z, Zhu Y, Tong C, Snyder JA, Lipton JW, Carvey
PM. 2006. Progressive dopamine neuron loss following
supra-nigral lipopolysaccharide (LPS) infusion into rats
exposed to LPS prenatally. Exp Neurol 199:499–512.
Ling Z, Zhu Y, Tong CW, Snyder JA, Lipton JW, Carvey
PM. 2009. Prenatal lipopolysaccharide does not acceler-
ate progressive dopamine neuron loss in the rat as a result
of normal aging. Exp Neurol 216:312–320.
Ling ZD, Chang Q, Lipton JW, Tong CW, Landers TM, Car-
vey PM. 2004b. Combined toxicity of prenatal bacterial
endotoxin exposure and postnatal 6-hydroxydopamine in
the adult rat midbrain. Neuroscience 124:619–628.
Lodygensky GA, West T, Stump M, Holtzman DM, Inder
TE, Neil JJ. 2010. In vivo MRI analysis of an inflamma-
tory injury in the developing brain. Brain Behav Immun
Lowe GC, Luheshi GN, Williams S. 2008. Maternal infec-
tion and fever during late gestation are associated with
altered synaptic transmission in the hippocampus of juve-
nile offspring rats. Am J Physiol Regul Integr Comp
Luheshi GN, Stefferl A, Turnbull AV, Dascombe MJ,
Brouwer S, Hopkins SJ, Rothwell NJ. 1997. Febrile
response to tissue inflammation involves both peripheral
and brain IL-1 and TNF-alpha in the rat. Am J Physiol
Magni DV, Souza MA, Oliveira AP, Furian AF, Oliveira
MS, Ferreira J, Santos AR, et al. 2011. Lipopolysaccha-
1346 Harvey and Boksa
ride enhances glutaric acid-induced seizure susceptibility
in rat pups: behavioral and electroencephalographic
approach. Epilepsy Res 93:138–148.
Makinodan M, Tatsumi K, Manabe T, Yamauchi T, Maki-
nodan E, Matsuyoshi H, Shimoda S, et al. 2008. Maternal
immune activation in mice delays myelination and axo-
nal development in the hippocampus of the offspring. J
Neurosci Res 86:2190–2200.
Malkova NV, Yu CZ, Hsiao EY, Moore MJ, Patterson PH.
2012. Maternal immune activation yields offspring dis-
playing mouse versions of the three core symptoms of
autism. Brain Behav Immun 26:607–616.
Meyer U, Engler A, Weber L, Schedlowski M, Feldon J.
2008a. Preliminary evidence for a modulation of fetal do-
paminergic development by maternal immune activation
during pregnancy. Neuroscience 154:701–709.
Meyer U, Feldon J, Dammann O. 2011. Schizophrenia and
autism: both shared and disorder-specific pathogenesis
via perinatal inflammation? Pediatr Res 69:26R–33R.
Meyer U, Feldon J, Schedlowski M, Yee BK. 2006a. Immu-
nological stress at the maternal-foetal interface: a link
between neurodevelopment and adult psychopathology.
Brain Behav Immun 20:378–388.
Meyer U, Nyffeler M, Engler A, Urwyler A, Schedlowski
M, Knuesel I, Yee BK, et al. 2006b. The time of prenatal
immune challenge determines the specificity of inflam-
mation-mediated brain and behavioral pathology. J Neu-
Meyer U, Nyffeler M, Schwendener S, Knuesel I, Yee BK,
Feldon J. 2008b. Relative prenatal and postnatal maternal
contributions to schizophrenia-related neurochemical
dysfunction after in utero immune challenge. Neuropsy-
Meyer U, Nyffeler M, Yee BK, Knuesel I, Feldon J. 2008c.
Adult brain and behavioral pathological markers of
prenatal immune challenge during early/middle and late
fetal development in mice. Brain Behav Immun 22:469–
Nakazawa K, Zsiros V, Jiang Z, Nakao K, Kolata S, Zhang
S, Belforte JE. 2011. GABAergic interneuron origin of
Nelson KB. 2008. Causative factors in cerebral palsy. Clin
Obstet Gynecol 51:749–762.
Nelson KB, Willoughby RE. 2000. Infection, inflammation
and the risk of cerebral palsy. Curr Opin Neurol 13:133–
Neufeld KA, Foster JA. 2009. Effects of gut microbiota on
the brain: implications for psychiatry. J Psychiatry Neu-
Nielsen LL, Nielsen NM, Melbye M, Sodermann M, Jacob-
sen M, Aaby P. 1998. Exposure to measles in utero and
Crohn’s disease: Danish register study. BMJ 316:196–
Nouel D, Burt M, Zhang Y, Harvey L, Boksa P. 2012. Pre-
natal exposure to bacterial endotoxin reduces the number
of GAD67- and reelin-immunoreactive neurons in the
hippocampus of the rat offspring. Eur Neuropsychophar-
Ozawa K, Hashimoto K, Kishimoto T, Shimizu E, Ishikura
H, Iyo M. 2006. Immune activation during pregnancy in
mice leads to dopaminergic hyperfunction and cognitive
impairment in the offspring: a neurodevelopmental
animal model of schizophrenia. Biol Psychiatry 59:546–
Paintlia MK, Paintlia AS, Barbosa E, Singh I, Singh AK.
degeneration of oligodendrocyte progenitors and hypo-
myelination in developing rat brain. J Neurosci Res
Paintlia MK, Paintlia AS, Contreras MA, Singh I, Singh
AK. 2008. Lipopolysaccharide-induced peroxisomal dys-
function exacerbates cerebral white matter injury: attenu-
ation by N-acetyl cysteine. Exp Neurol 210:560–576.
Pakula AT, Van Naarden Braun K, Yeargin-Allsopp M.
2009. Cerebral palsy: classification and epidemiology.
Phys Med Rehabil Clin N Am 20:425–452.
Pang Y, Cai Z, Rhodes PG. 2003. Disturbance of oligoden-
drocyte development, hypomyelination and white matter
injury in the neonatal rat brain after intracerebral injec-
tion of lipopolysaccharide. Brain Res Dev Brain Res
Pardi DS, Tremaine WJ, Sandborn WJ, Loftus EV Jr, Poland
GA, Melton LJ III. 1999. Perinatal exposure to measles vi-
rus is not associated with the development of inflamma-
tory bowel disease. Inflamm Bowel Dis 5:104–106.
Patterson PH. 2012. Maternal infection and autism. Brain
Behav Immun 26:393.
Perry W, Minassian A, Lopez B, Maron L, Lincoln A.
2007. Sensorimotor gating deficits in adults with autism.
Biol Psychiatry 61:482–486.
Ray A, Redhead K, Selkirk S, Poole S. 1991. Variability in
LPS composition, antigenicity and reactogenicity of
phase variants of Bordetella pertussis. FEMS Microbiol
Reddihough DS, Collins KJ. 2003. The epidemiology and
causes of cerebral palsy. Aust J Physiother 49:7–12.
Reimer T, Brcic M, Schweizer M, Jungi TW. 2008.
poly(I:C) and LPS induce distinct IRF3 and NF-kappaB
signaling during type-I IFN and TNF responses in human
macrophages. J Leukoc Biol 83:1249–1257.
Romero E, Ali C, Molina-Holgado E, Castellano B, Guaza
C, Borrell J. 2007a. Neurobehavioral and immunological
consequences of prenatal immune activation in rats.
Influence of antipsychotics. Neuropsychopharmacology
Romero E, Guaza C, Castellano B, Borrell J. 2010. Ontog-
eny of sensorimotor gating and immune impairment
induced by prenatal immune challenge in rats: implica-
tions for the etiopathology of schizophrenia. Mol Psychi-
Romero R, Gotsch F, Pineles B, Kusanovic JP. 2007b.
Inflammation in pregnancy: its roles in reproductive
physiology, obstetrical complications, and fetal injury.
Nutr Rev 65:S194–S202.
Roth J, Rummel C, Barth SW, Gerstberger R, Hubschle T.
2009. Molecular aspects of fever and hyperthermia.
Immunol Allergy Clin North Am 29:229–245.
Early Life Infection and Neurodevelopment1347
Roumier A, Pascual O, Bechade C, Wakselman S, Poncer
JC, Real E, Triller A, et al. 2008. Prenatal activation of
microglia induces delayed impairment of glutamatergic
synaptic function. PLoS One 3:e2595.
Samuelsson AM, Jennische E, Hansson HA, Holmang A.
2006. Prenatal exposure to interleukin-6 results in inflam-
matory neurodegeneration in hippocampus with NMDA/
GABA(A) dysregulation and impaired spatial learning.
Am J Physiol Regul Integr Comp Physiol 290:R1345–
Schlotz W, Phillips DI. 2009. Fetal origins of mental health:
evidence and mechanisms. Brain Behav Immun 23:905–
Shi L, Fatemi SH, Sidwell RW, Patterson PH. 2003. Mater-
nal influenza infection causes marked behavioral and
pharmacological changes in the offspring. J Neurosci
Shi L, Smith SE, Malkova N, Tse D, Su Y, Patterson PH.
2009. Activation of the maternal immune system alters
cerebellar development in the offspring. Brain Behav
Sinclair KD, Lea RG, Rees WD, Young LE. 2007. The de-
velopmental origins of health and disease: current theo-
ries and epigenetic mechanisms. Soc Reprod Fertil Suppl
Sinclair KD, Singh R. 2007. Modelling the developmental
origins of health and disease in the early embryo. Therio-
Smith SE, Li J, Garbett K, Mirnics K, Patterson PH. 2007.
Maternal immune activation alters fetal brain develop-
ment through interleukin-6. J Neurosci 27:10695–10702.
Spencer SJ, Auer RN, Pittman QJ. 2006. Rat neonatal
immune challenge alters adult responses to cerebral
ischaemia. J Cereb Blood Flow Metab 26:456–467.
Spencer SJ, Galic MA, Pittman QJ. 2011. Neonatal pro-
gramming of innate immune function. Am J Physiol
Endocrinol Metab 300:E11–E18.
Spencer SJ, Hyland NP, Sharkey KA, Pittman QJ. 2007.
Neonatal immune challenge exacerbates experimental
colitis in adult rats: potential role for TNF-alpha. Am J
Physiol Regul Integr Comp Physiol 292:R308–R315.
Stewart KA, Wilcox KS, Fujinami RS, White HS. 2010.
Development of postinfection epilepsy after Theiler’s vi-
rus infection of C57BL/6 mice. J Neuropathol Exp Neu-
Tohmi M, Tsuda N, Watanabe Y, Kakita A, Nawa H. 2004.
Perinatal inflammatory cytokine challenge results in dis-
tinct neurobehavioral alterations in rats: implication in
psychiatric disorders of developmental origin. Neurosci
Urakubo A, Jarskog LF, Lieberman JA, Gilmore JH. 2001.
Prenatal exposure to maternal infection alters cytokine
expression in the placenta, amniotic fluid, and fetal brain.
Schizophr Res 47:27–36.
van de Bor M, Guit GL, Schreuder AM, Wondergem J,
Vielvoye GJ. 1989. Early detection of delayed myelina-
tion in preterm infants. Pediatrics 84:407–411.
Vuillermot S, Joodmardi E, Perlmann T, Ove Ogren S, Fel-
don J, Meyer U. 2012. Prenatal immune activation inter-
acts with genetic nurr1 deficiency in the development of
attentional impairments. J Neurosci 32:436–451.
Wang S, Yan JY, Lo YK, Carvey PM, Ling Z. 2009a. Do-
paminergic and serotoninergic deficiencies in young
adult rats prenatally exposed to the bacterial lipopolysac-
charide. Brain Res 1265:196–204.
Wang X, Stridh L, Li W, Dean J, Elmgren A, Gan L, Eriks-
son K, et al. 2009b. Lipopolysaccharide sensitizes neona-
tal hypoxic-ischemic brain injury in a MyD88-dependent
manner. J Immunol 183:7471–7477.
White JF. 2003. Intestinal pathophysiology in autism. Exp
Biol Med (Maywood) 228:639–649.
Wing L. 1997. The autistic spectrum. Lancet 350:1761–
Winter C, Djodari-Irani A, Sohr R, Morgenstern R, Feldon
J, Juckel G, Meyer U. 2009. Prenatal immune activation
leads to multiple changes in basal neurotransmitter levels
in the adult brain: implications for brain disorders of neu-
rodevelopmental origin such as schizophrenia. Int J Neu-
Winter C, Reutiman TJ, Folsom TD, Sohr R, Wolf RJ,
Juckel G, Fatemi SH. 2008. Dopamine and serotonin lev-
els following prenatal viral infection in mouse-Implica-
tions for psychiatric disorders such as schizophrenia and
autism. Eur Neuropsychopharmacol 18:712–716.
Wolff AR, Bilkey DK. 2008. Immune activation during
mid-gestation disrupts sensorimotor gating in rat off-
spring. Behav Brain Res 190:156–159.
Wusteman M, Wight DG, Elia M. 1990. Protein metabo-
lism after injury with turpentine: a rat model for clinical
trauma. Am J Physiol 259:E763–E769.
Zhang Y, Cazakoff BN, Thai CA, Howland JG. 2012. Pre-
natal exposure to a viral mimetic alters behavioural flexi-
bility in male, but not female, rats. Neuropharmacology
Zuckerman L, Rehavi M, Nachman R, Weiner I. 2003a.
Immune activation during pregnancy in rats leads to a
postpubertal emergence of disrupted latent inhibition,
dopaminergic hyperfunction, and altered limbic morphol-
ogy in the offspring: a novel neurodevelopmental model
of schizophrenia. Neuropsychopharmacology 28:1778–
Zuckerman L, Rimmerman N, Weiner I. 2003b. Latent inhi-
bition in 35-day-old rats is not an \adult" latent inhibition:
implications for neurodevelopmental models of schizo-
phrenia. Psychopharmacology (Berl) 169:298–307.
Zuckerman L, Weiner I. 2003. Post-pubertal emergence of
disrupted latent inhibition following prenatal immune
activation. Psychopharmacology (Berl) 169:308–313.
1348 Harvey and Boksa