passionless Droning about autism

Hello Friends – There are (at least) two big
classifications of microglia findings in autism, an altered
morphology (i.e., shape and function, or ‘activated’ versus
‘quiescent’), and an increased number (i.e., more), with both
parameters varying with each other and spatially.  In other
words, disparate parts of the brain have different numbers of
microglia in them, and the functional profile of those microglia
also varies from one area to another. 
[Note: There is ongoing
discussion regarding the appropriate definition of
‘activation’ of microglia, with evidence of (at least) four states
of microglial morphology.] Recently I saw a discussion on the SFARI
site about the fancy in
vivo study of microglia numbers in high functioning males with
autism.  (I believe I am growing
increasingly jaded, as it occurs to me that radio tracing against
[¹¹C](R)-PK11195) to
show microglial activation is a fancy trick, but one leaving us
open to detecting other stuff too.)  In any case, the findings
are not especially unexpected by now, well not to me anyways, but a
comment at the SFARI site really got me thinking about the chain of
events that could lead to different spatial and
morphological characteristics of microglia.  Perhaps we could
gain insight into the question of what the microglia are doing by
trying to understanding how they got there. Do we have any
biologically plausible models that might educate us on how a
different morphology and distribution of microglia could be
achieved?  A while ago I got a copy of a few
articles that don’t have autism in them per se, but they kept
coming to the forefront of my mind when I thought about that
question.  The first is Distribution
of microglia in the postnatal murine nigrostriatal
system, which had a disease focus on
Parkinson’s, but what really grabbed my attention is what they
learned about the developmental pathway microglia took to populate,
and then depopulate the substantia nigra (SN), a little wedge of
brain involved with motor skills, reward seeking, and addiction.

Interestingly, the SN
has been shown to contain more microglia than
adjacent structures. We have analysed
changes in microglia numbers and in microglial morphology in the
postnatal murine nigrostriatal system at various stages ranging
from postnatal day 0 (P0) up to 24 months of age. We
clearly show that the microglia numbers in the SN and in the
striatum dramatically increase from P0 to P15 and
significantly decrease in both areas in 18-month-old and
24-month-old animals.

[Note: There seems to be
some variance in the appropriate ‘rat-to-human-age’ approximations;
especially when trying to do something as
expeditionary as comparing brain development.  We should
extrapolate only with caution.] The part that makes me grin is that
it illustrates our nascent understanding of the process of
microglial colonization into the CNS, the hows,
whens, wheres, and whys are still
shrouded in mystery. The broadest outlines tell us that microglial
penetration into the brain is a long running, dynamic process; the
microglia are slow infiltrators, gaining access into parts of the
brain in concert with a swath of proliferating and inhibitory
factors, all at a time of once in a lifetime neurodevelopmental
modifications. Regulation
of postnatal forebrain amoeboid microglial cell proliferation and
development by the transcription factor Runx1 paints a
beautiful portrait of functionality.  Runx1 is a chemical
messenger that participates in phenotyopic determination of blood
cell progenitors into mature cells.  The researchers observed
spatial, time dependent expression of Runx1 in the developing
forebrain, and differential levels following injury.

Here we show that the mouse
transcription factor Runx1, a key regulator of myeloid
cell proliferation and differentiation, is expressed in forebrain
amoeboid microglia during the first two postnatal weeks.
Runx1 expression is then downregulated in ramified microglia. Runx1
inhibits mouse amoeboid microglia proliferation and promotes
progression to the ramified state. We show further that
Runx1 expression is upregulated in microglia following nerve injury
in the adult mouse nervous system. These findings provide
insight into the regulation of postnatal microglia activation and
maturation to the ramified state and have implications for
microglia biology in the developing and injured brain.

It doesn’t really tell us much about a
persistent change in microglia per se, but it does render a picture
of proliferation and differentiation as an easily
disrupted symphony.  When we think about the
developing brain, I won’t pretend to have more than a lightyear
close guess at what microglia might be doing
differently between amoeboid and ramified
morphologies in this locale, at this time, but I
very highly doubt there isn’t
a functional impact on microenvironment neurodevelopment; our
developing brains are using opportunities like the Indians used the
buffalo, no waste, no excess, and because balance is important,
everything is important. Moving back to the
question of the plausibility of a pathway to the autism state,
luckily (or unluckily?) the literature is veritably littered with
insults that perturb microglial development, leading to
 persistent changes to microglial morphology, ultimately
percolating up to behavioral changes. Prenatal stress is a bad, bad
thing, and here is a study that finds that extreme mice stress can
persistently alter the mice activation profile of mice microglia.
Prenatal
stress increases the expression of proinflammatory cytokines and
exacerbates the inflammatory response to LPS in the hippocampal
formation of adult male mice, was just published, and
comes wrapped up with a double hit, and
different resting and stimulated neuroimmune environments.

Under basal conditions,
prenatally stressed animals showed increased expression of
interleukin 1ß and tumor necrosis factor-a (TNF-a) in the
hippocampus and an increased percentage of microglia cells with
reactive morphology in CA1 compared to non-stressed males.
Furthermore, prenatally stressed mice showed increased TNF-a
immunoreactivity in CA1 and increased number of Iba-1
immunoreactive microglia and GFAP-immunoreactive astrocytes in the
dentate gyrus after LPS administration. In contrast, LPS did not
induce such changes in non-stressed animals. These
findings indicate that prenatal stress induces a basal
proinflammatory status in the hippocampal formation during
adulthood that results in an enhanced activation of microglia and
astrocytes in response to a proinflammatory
insult.

Note: I have not read this
paper so I do not know if a qualitative number of microglia, or
just more immune-targeted microglia were found, but likely the
latter. A similar, full free paper, Prenatal
stress causes alterations in the morphology of microglia and the
inflammatory response of the hippocampus of adult female
mice, found broadly similar results; perturbed resting
and stimulated states in the treatment group.

Prenatal stress, per se,
increased IL1ß mRNA levels in the hippocampus, increased the total
number of Iba1-immunoreactive microglial cells and increased the
proportion of microglial cells with large somas and retracted
cellular processes. In addition, prenatally stressed and
non-stressed animals showed different responses to peripheral
inflammation induced by systemic administration of LPS.
LPS induced a significant increase in mRNA levels of IL-6,
TNF-a and IP10 in the hippocampus of prenatally stressed mice but
not of non-stressed animals.  

Going back to my
ramblings on glial priming, it seems that here we have an
example of a type of cross system priming (sweet!), where
disturbing the stress response system changed the immune system;
such is the way of the polyamorous chemical families interacting in
our brain.  It also occurs to me that given the
delicate nature of the developing brain, and the
crazy important
tasks going on in there, we might want to think very
carefully before we ‘induced a significant increase in
mRNA levels of IL-6, TNF-a and IP10 in the hippocampus‘
on subgroups who might be environmentally predisposed to react with
exaggerated vigor.  But what do I know? Of course, the
prenatal immune challenge arena holds a ton of studies on
persistent microglial function, and ‘consequences’.  There are
way too many to list, but a quick overview of some very recent ones
would include: Enduring
consequences of early-life infection on glial and neural cell
genesis within cognitive regions of the brain, an early
life real infection model with e coli that
concludes, “Taken together, we have provided evidence that
systemic infection with E. coli early in life has significant,
enduring consequences for brain development and subsequent adult
function.”  (Staci Bilbo!)  This paper was sort
of a quinella, as it showed both changes in immune responsiveness
into adulthood; it also demonstrated the ability
of an immune insult to alter
the developmental trajectory of the
microglia, i.e., E. coli increased the number of newborn
microglia within the hippocampus and PAR compared to controls. The
total number of microglia was also significantly increased in E.
coli-treated pups, with a concomitant decrease in total
proliferation. Neonatal
lipopolysaccharide exposure induces long-lasting learning
impairment, less anxiety-like response and hippocampal injury in
adult rats very directly blasted rats with some LPS
immune activation action, and includes, ”Neonatal LPS
exposure also resulted in sustained inflammatory responses in the
P71 rat hippocampus, as indicated by an increased number of
activated microglia and elevation of interleukin-1ß content in the
rat hippocampus.”  (Sound familiar?) Interleukin-1
receptor antagonist ameliorates neonatal lipopolysaccharide-induced
long-lasting hyperalgesia in the adult rats took the
extra step of adding a set of animals that got inhibited
inflammatory responses.  Results are increasingly
unsurprising.

Neonatal
administration of an IL-1 receptor antagonist (0.1mg/kg)
significantly attenuated long-lasting hyperalgesia induced by LPS
and reduced the number of activated microglia in the adult rat
brain. These data reveal that neonatal intracerebral LPS
exposure results in long-lasting hyperalgesia and an elevated
number of activated microglia in later life. This effect is similar
to that induced by IL-1ß and can be prevented by an IL-1 receptor
antagonist

I love how (once again) we
can see how interrupting the immune response can have an effect.
Environmental impacts outside of the immune
activation realm may also find a place within the ‘big tent’ of
microglial agitation with consequent developmental impacts. 
The people who made the first big neuroimmune / autism splash at
Johns Hopkins later came out with Neuroinflammation
and behavioral abnormalities after neonatal terbutaline treatment
in rats: implications for autism, which found that an
agent used to prevent labor in some situations could
“produced a robust increase in microglial activation on PN
30 in the cerebral cortex” in treatment animals. 
The drug in question, terbutaline, has been weakly associated with
increased incidence of autism, i.e., Prenatal
exposure to ß2-adrenergic receptor agonists and risk of autism
spectrum disorders, and beta2-adrenergic
receptor activation and genetic polymorphisms in autism: data from
dizygotic twins. And now, in 2013, Beta-adrenergic
receptor activation primes microglia cytokine production,
displays another example of cross system
priming.

To determine
if ß-AR stimulation is sufficient to prime microglia, rats were
intra-cerebroventricularly administered isoproterenol (ß-AR
agonist) or vehicle and 24h later hippocampal microglia were placed
in culture with media or LPS. Prior isoproterenol treatment
significantly enhanced IL-1ß and IL-6, but not TNF-a production
following LPS stimulation. These data suggest that central
ß-AR stimulation is sufficient to prime microglia cytokine
responses.

In other words, they gave
the rats a drug in the class of terbutline, and subsequently
observed an increased microglia responsiveness in cultured
cells.  What a crazy coincidence. Detecting total
populations of microglia in adulthood, either regionally
or in the brain as a whole is a little more difficult, the little
buggers are a lot easier to detect when we light them up with neon
green tracers that stick to proteins expressed at ‘activation’
time, and it just doesn’t look like the question has been asked too
many times.  I did, however, find something that has a sort of
chip shot on this analysis, Prenatal
stress alters microglial development and distribution in postnatal
rat brain, which looked at regional microglia populations
and phenotypes at two time periods following prenatal stress
events.

Prenatal
stress consisting of 20 min of forced swimming occurred on
embryonic days 10–20. On postnatal days 1 and 10, stressed and
control pups were killed. Microglia were identified using Griffonia
 simplicifolia lectin and quantified in the whole encephalon.
In addition, plasma corticosterone was measured in dams at
embryonic day 20, and in pups on postnatal days 1 and 10.
At postnatal day 1, there was an increase in number of
ramified microglia in the parietal, entorhinal and frontal
cortices, septum, basal ganglia, thalamus, medulla oblongata and
internal capsule in the stressed pups as compared to controls, but
also there was a reduction of amoeboid microglia and the total
number of microglia in the corpus callosum. By postnatal
day 10, there were no differences in the morphologic type or the
distribution of microglia between the prenatal stress and control
groups, except in the corpus callosum; where prenatal stress
decreased the number of ramified microglia. The stress procedure
was effective in producing plasma rise in corticosterone levels of
pregnant rats at embryonic day 20 when compared to same age
controls. Prenatal stress reduced the number of immature
microglia and promoted an accelerated microglial differentiation
into a ramified form.

They did a lot
of clever stuff at analysis time, taking samples from several
locations after birth and ten days later, and
also did some fine grained classification of the
shape of the microglia.  They include spatial and temporal
mappings of four microglial developmental profiles.  It looks
as if prental stress was able to alter the developmental speed of
microglia from one morphology to another in different parts of the
brain.  There was as small section in the discussion that
speculated on what such changes might mean for neurodevelopment.

Given that during the
early postnatal period occur numerous brain developmental processes
(e.g. neurogenesis, myelination, synaptogenesis, astrogliogenesis,
neuronal cell death and blood–brain barrier maturation) [6, 19, 22,
25, 36, 52] it is possible that altered microglial
development induced by in utero stress may affect other
developmental processes either changing microenvironment molecular
constitution or triggering earlier inflammatory changes secondary
to the blood–brain barrier opening induced by prenatal
stress .  Although punctual, the altered microglial
development might alter extensively the other
neurodevelopmental processes ensuing perdurable
structural changes; for example it is possible that the
change in the distribution pattern of microglia in the prenatal
stress group may render vulnerable some neuroanatomic
regions due to the reduction of neurotrophic factors,
such as the corpus callosum where there is a continuous axonal
growth

No kidding! [There is also some very
interesting notes regarding microglial participation in purkinje
cell death that deserves and entire post. . .] This should be the
point that any rational observer must accept that we several lines
of evidence that early life experiences can persistently alter
microglial function with plausible mechanisms that could affect
neurodevelopment.  Our data concerning total population
numbers in adulthood is a lot more difficult to come by, but I
think this will probably be getting looked at soon enough. Of
course, in any particular individual it is difficult (or
impossible?) to know how they may have arrived at a state of
increased microglial activation, but at the same time, it is not as
if we have no clue on possible pathways to this destination; our
short list of environmental factors includes immune insult, stress,
and chemical agents. If the question is, ‘what are the microglia
doing in the autism population?’, one plausible answer is ‘their
phenotype was persistently altered by an early life event through a
developmental programming model’. As I was mulling all of this
over, two things happened.  First, a maternal CRP
study came out, and found a pretty strong relationship
between direct measurements of mommy inflammation with increased
risk of baby autism.  The nice part is that they had a
gigantic data set (1.2M births!) to work with thanks to a few
decades of single payer medicine.  (Very
nice!)

For maternal CRP
levels in the highest quintile, compared with the lowest quintile,
there was a significant, 43% elevated
risk. This finding suggests that
maternal inflammation may have a significant role in autism, with
possible implications for identifying preventive strategies and
pathogenic mechanisms in autism and other neurodevelopmental
disorders.

Just after that paper came out, I
made some Fred Flintstone style beef ribs.  I ‘primed’ the
meat with a Moroccan inspired spice rub overnight, then
slow, slow, slow cooked them with a
low, low, low heat all day
long, and blasted away with a date glaze under the
broiler just before go time and they were caveman style
primal fucking awesome.  The key to arriving there
was the slow cooking. The rib preparation
process got me thinking about our population wide experiment
in replacing infection with inflammation where we have
traded in death by pathogens or other once fatal ailments in
exchange for a longer life frequently plagued by conditions
associated with higher inflammation.  Our analysis on long
term alterations to microglial proliferation and morphology is
largely comprised of studying acute insults
(sound familiar?), i.e., injection of purified bacterial cell
components known to trigger a robust immune response, ten sessions
of mouse based pregnant forced swimming, or exposure to chemicals
with rare and particular exposure routes in humans.  Mostly I
think this is due to the black swan nature of the developmental programming
model alongside the very new idea that microglia are
doing jobs other than responding to infections; our models are
crude because of our relative ignorance.  What will we find
when our filters are appropriately powered to detect for chronic,
but subtle insults? It occurs to me that there may be a ribs model
of altered microglial colonization of the fetal brain; it seems
clear that proliferation and differentiation of microglia can
clearly be changed by powerful inputs, but the
chemical messengers that impact that change are closely related (or
the same) as the measurement points in the maternal CRP study.
Could a slow cooking of slightly higher but not acutely
increased maternal inflammation be participating in the
genesis of autism (in some children) through altering the migration
and proliferation of microglia into the neonatal brain?  Could
the same chemical messengers of inflammation be subtly
priming the microglia to respond with increased vigor to
insults later in life?  Has our replacement of infection with
inflammation included an unanticipated effect that alters the
developmental pathway of the very cells that help shape our
children’s brains? I don’t think we are (quite) clever enough to
answer these types of questions yet, but we are at least starting
to generate the right kind of data to inform us on how to get
started.  I don’t know what we will find, but the initial data
doesn’t look very good.  In the meantime, I am recommending
you go get some ribs and let them cook all day long. –     
–
pD