Prenatal inflammation and fetal brain development: 2 key actors in autism
During pregnancy, the developing fetus is not totally protected from its
environment. It can interact and respond to it and in return the environment
can have permanent and dramatic effects on organ development at very sensitive
and crucial periods. In particular, the cognitive functions could be affected
by prenatal factors thus leading to long-lasting brain damage.
Maternal health and immune environment during pregnancy.
Maternal health is crucial to maintain a favorable environment suitable
to fetal growth during pregnancy. Alterations in maternal immune homeostasis
during pregnancy (in response to infection, disease or other environmental
influences) are associated with miscarriage, preterm delivery or preeclampsia (1).
Maternal immune dysregulation during pregnancy: the "cytokine theory"
of neurodevelopmental disorder.
In the New York Times, this theory was reviewed two years ago by the author
Moises Velasquez-Manoff: " At least a subset of autism — perhaps one-third, and very
likely more — looks like a type of inflammatory disease. And it
begins in the womb"
(An immune disorder at the root of Autism).
The maternal immune system play a pivotal role during pregnancy, protecting
the mother and the fetus from deleterious environmental pathogens. Moreover,
the maternal cellular immune responses must be suppressed to prevent rejection
of the fetus. A healthy pregnancy switches the maternal immune system
toward a more tolerant, low inflammatory state that decreased the production
of inflammatory cytokines and increased the production of cytokines that
are more regulatory.
Maternal immune alteration have been linked to ASD where immune dysfunction
in the mothers of autistic children during pregnancy could alter brain
development as seen in affected children (2-4). A large Danish epidemiologic
study has correlated maternal infection during pregnancy (viral infection
during the first trimester or a maternal bacterial infection during the
second trimester) to autism in children (5).
An acute maternal infection induces an inflammatory response with an abnormal
cytokine and chemokine secretion that could affect fetal brain development (6-8).
Cytokines and chemokines are proteins that control the type of immune response
(with variation in the intensity and the duration of the response), they
are involved in pregnancy maintenance (9) and have a strong role in neurodevelopment (10-11).
Dysregulations of inflammatory cytokine and chemokine levels such as IL4, IL-5,
IL-8 IFN-γ, TNF
MCP-1 have been reported in the maternal serum or amniotic fluid during early-
or mid-gestation of autistic children (12-15).
IL-6 is a very interesting cytokine that can cross the human placenta (unlike
many other cytokines) and may directly alter the placental environment
thus impacting the fetus (16). A healthy pregnancy is characterized by
a shift in cytokine production and by immune cells regulation to prevent
fetal rejection (17). IL-6 has pleiotropic effects on the utero-placental
compartment: modulating the placental immunological balance, activating
uterine immune cells, and maintaining maternal tolerance (2).
Several clinical studies suggest that IL6 levels in the amniotic fluid
predict development of cerebral lesions (18-20) and impairment in spatial
learning (21). IL-8, a neutrophil attractant chemokine is also associated
with neurodevelopmental disorder such as schizophrenia (22) that has similar
etiology with autism (23).
An increased level of MCP-1 has also been seen in different neuroinflammatory
diseases or traumatic brain damage (24). As the composition of the amniotic
fluid is more from fetal than maternal origin, an elevated MCP-1 could
reflects an inflammation in the fetal brain (for more information see
the paragraph "Blood-Brain barrier, brain inflammation and autism").
Interestingly, MCP-1 induces IL-6 (8) creating an amplification loop in
the inflammation process.
Another pro-inflammatory biomarker, C-reactive protein (CRP) has been found
to be elevated in maternal serum of autistic children (25).
Maternal autoimmune disease and anti-brain autoantibodies.
Several epidemiologic studies showed a clear link between maternal autoimmune
diseases and autism (26). The high prevalence of autoimmune diseases in
mothers of autistic children suggests that these disorders may affect
the developing fetus in utero (27).
Several studies have shown that mothers of autistic children have a higher
frequency of anti-brain autoantibodies in their plasma than mothers of
healthy children (28) and this is associated with maternal autoimmune
diseases in particular rheumatoid arthritis and systemic lupus erythematosus
(29). Furthermore, autoantibodies specifically present in women with systemic
lupus erythematosus, have been shown in mouse models to be neurotoxic
to the developing brain (30). In utero exposure to maternal anti-brain
autoantibodies could be an important trigger for abnormal brain development
as suggested by several studies (31-33). Recent work has demonstrated
that IgG (antibodies) from mother of autistic children trigger autistic-like
behavior in mice (34) or in rhesus monkey (35).
Anti-paternal HLA antibodies.
During pregnancy, maternal adaptative immune system could be activated
by allogenic fetal cells. This leads to maternal immunization against
paternally inherited HLA antigens (36), particularly against male-specific
minor histocompatibility (HY) inherited antigens when carrying a male
fetus (37). Anti-HY immunization (HY antibodies) may be generated and
although these responses are generally well tolerated, they can have deleterious
effects on the fetus and the pregnancy outcome (38-39) in immune sensitized
patient. Anti-HY antibodies could induce an embryonic toxic response with
the induction of pro-inflammatory cytokines such as TNFα, CD8+ T
lymphocyte and a decrease in regulatory T cells (40).
To date, there is no scientific evidences showing a link between anti-paternal
HLA antibodies and neurodevelopmental disorder. Nevertheless, we strongly
believe that the inflammation resulting from an anti-paternal HLA antibodies
production during pregnancy could play a role in autism and could explain
the high prevalence of autism in males.
Fig. 1 Probable mechanisms of fetal brain injury with in utero exposure to maternal
inflammation. Adapted from the article " Models of fetal brain injury,
intrauterine inflammation, and preterm birth" by Burd et
al , 2012 (
AJRI , 67 (4): 287-94).
Several animal models, showing symptoms of autism, have been generated
by altering the maternal environment during pregnancy (41-44).This can
be done by infecting the mother or by activating the immune system with
no pathogens involved. These animal models are very useful tools to decipher
the molecular mechanisms leading to autism. Maternal immune activation
(MIA) in mice or rats give rise to a prenatal exposure to altered levels
of cytokines such as IL-1, IL-2, IL-6 or TNF-α, significantly elevated
in the placenta (45-47) but also in the fetal brain (48-49). This inflammation
is sufficient to induce fetal brain injuries with autistic-like behavioral
changes in the offspring (50). The maternal immune response to MIA will
determine the severity of the behavioral symptoms in the offspring (51).
As an example, a loss of weight after MIA in the pregnant mice is correlated
with a higher production of TNF-α leading to aggravated long term
developmental brain and behavioral abnormalities in the offspring.
Blood-Brain-Barrier (BBB), Brain inflammation and Brain damage in autism.
Placenta allows maternal antibodies, such as immunoglobulin G (IgG), to
reach the fetus and provide passive immunity throughout the pregnancy.
In addition to the passage of protective antibodies, maternal antibodies
reactive to fetal antigens (autoantibodies) could also reach and harm
the fetus (52). Moreover, some inflammatory cytokines such as IL-6 could
also cross the placenta (46, 53).
The blood brain barrier (BBB) separates the central nervous system (CNS)
from the immune system, to create an immune privileged environment for
the brain (54).
The BBB is developing during the fetal period thus maternal antibodies
are able to have direct access to the brain during gestation (29, 33).
Maternal autoantibodies targeting fetal brain have been detected in autistic
children brain (54, 55) highlighting the transfer of these factors from
the maternal to the fetal blood stream to finally reach fetal brain (55,
56). A large number of clinical studies strongly established the presence
of maternal autoantibodies in the serum (57-59) or in the brain of autistic
patients in post mortem specimens (60). Brain injury by circulating brain–specific
maternal autoantibodies may play a key role in developmental disorders (30, 61).
After a prenatal exposure to maternal inflammation, the brain proteome
(all the proteins expressed in the brain) is severely altered in mice
(62). Indeed elevated cytokines and chemokines were found in postmortem
brain specimens of individuals who had ASD (63-66). The inflammatory cytokine
levels are significantly elevated in ASD patients compared to the control
population (67, 68) which was linked to impaired communication and social
Placental inflammation induces irreversible perinatal brain damage either
directly by damaging neurons or by affecting brain development at different
- it damages the white-matter (71) with an alteration of astrocytes development
(72) and an activation of the microglia (73) that in turn will secrete
- it induces neuron cell death (74)
- it affects the fetal blood–brain barrier efficiency (75).
The severity of white matter damage is accentuated when the inflammation
exposure occurs earlier putting into light a notion of window of sensitivity
during fetal neurodevelopment (76). The brain damage in autistic children
can be morphologically assessed as abnormal cortical development has been
reported (77), Magnetic Resonance Imaging also showed differences between
autistic children and control individuals (78).
View All References > >