My son was a ‘gut kid’.  The irony is, for a while, because we were first time parents, we didn’t even know.  My son was flagged for evaluation for autism around a year of age and we met with the autism center people several times between his first and third birthdays, with his official diagnosis coming just after he turned three.  My wife came home from one of the early meetings convinced that his evaluators didn’t know the first thing about our son, autism, or anything else, and that in fact, they might be insane.

 ‘Do you know what those idiots asked me today?’

‘What?’

‘What his shits look like.  My kid can’t talk and they want to ask me about his diapers!’

‘Who fucking cares?‘

We wound up caring, a lot.  It turns out, this wasn’t a stupid question, it just seemed like one to us.   The answer to their question was that our son was having at least four or six very messy diapers a day, his stools were never firm logs that look like they came from an spherical filter, but always, always more liquid than solid, and frequently contained chunks of identifiable food.  But from our viewpoint, within the context of a child who was not speaking,  hurting himself, and never looked at anyone, the idea that we should be worrying about his shit was the stupidest thing we’d ever heard. 

But.  When we started paying attention, starting reading, and started meeting more people with children with autism, our incredulity waned.  We  tried GF/CF and probiotics.  We paid for lab tests to analyze the bacterial populations in his intestines. We experienced a life saving miracle with anti-fungal agents wherein my son essentially stopped hurting himself over the course of weeks after persistently banging his head dozens of times a day for six months.  For nearly a year we removed all complex carbohydrates from our son’s diet, an intervention that makes GFCF feel like a Sunday afternoon after college but before kids and autism.  We saw changes in our son based on how his GI tract was performing.  For our son, for us, we knew that by some mechanism, what got put in his mouth, and what happened along the way was tightly coupled with how our son felt and behaved.

This is why my vision with spots of rage when I see the ideas of GI and dietary involvement with autism mocked by pseudo-skeptics so rampantly on the Internet.  I cannot stand the thought of a single child continuing to suffer the way I watched my son suffer because they were told that there was no basis of GI interaction in autism.  That thought hurts.

Those biases stated, we are now, finally, starting to see research indicating that in some cases of autism, there are very real, non imaginary differences in GI function.

A few months ago, Impaired Carbohydrate Digestion and Transport and Mucosal Dysbiosis in the Intestines of Children with Autism and Gastrointestinal Disturbances was published [full, dense, but very cool paper available online].  Here is the abstract.

Gastrointestinal disturbances are commonly reported in children with autism, complicate clinical management, and may contribute to behavioral impairment. Reports of deficiencies in disaccharidase enzymatic activity and of beneficial responses to probiotic and dietary therapies led us to survey gene expression and the mucoepithelial microbiota in intestinal biopsies from children with autism and gastrointestinal disease and children with gastrointestinal disease alone. Ileal transcripts encoding disaccharidases and hexose transporters were deficient in children with autism, indicating impairment of the primary pathway for carbohydrate digestion and transport in enterocytes. Deficient expression of these enzymes and transporters was associated with expression of the intestinal transcription factor, CDX2. Metagenomic analysis of intestinal bacteria revealed compositional dysbiosis manifest as decreases in Bacteroidetes, increases in the ratio of Firmicutes to Bacteroidetes, and increases in Betaproteobacteria. Expression levels of disaccharidases and transporters were associated with the abundance of affected bacterial phylotypes. These results indicate a relationship between human intestinal gene expression and bacterial community structure and may provide insights into the pathophysiology of gastrointestinal disturbances in children with autism.

I’ll admit it.  From the outside, from the don’t-have-a-kid-with-autism-and-GI-problems perspective, that is some dense and seemingly bland stuff.  Essentially children with GI distress and children with GI distress and autism were compared and it was found that there were distinctly qualitative differences regarding the GI function in the groups.  This is validation of what a lot of us have been saying for a long time, that the GI problems our kids were experiencing weren’t coincidental to the autism, but somehow related.

For anyone who has been paying attention to the details of the autism-gut debate, these are dynamite findings.  These observations are the death knell for the overused, oversimplified notion that the GI connection to autism was a function of some kids having autism, some kids having GI distress, and that therefore, some kids with autism also have GI distress.  This research tells us that the reality is not so simple.

This study is the view from the microscope as opposed to the telescope, and took care not to study just anyone with an autism diagnosis, but those with an autism diagnosis and GI distress, problems so severe that invasive procedures to obtain tissue samples from the GI tract was warranted.   This is a critically important facet of the study design in my opinion, a lot of the negative findings in this arena have been epidemiological, and cast the widest possible net, capturing everyone with autism and comparing them with a sample of everyone else.  This is a great strength of the paper; for too long everyone has acknowledged the heterogeneous nature of autism, but few studies have tried to understand differences at a phenotype level.  This paper is different.

As evidence of the non-random population, the autism patient group had a regression incidence of over eighty percent, and nearly as many of the children in both groups were reported to have food allergies.

The details of the findings in the paper get deep pretty fast, but at a high level there were differences found in proteins involved with the digestion of carbohydrates and changes in bacterial populations between the groups, with some differences found with regard to specific locations in the intestine.  Based on these findings, the authors speculate that alterations in carbohydrate processing could result in abnormal bacterial populations by way of altered microbial food availability in parts of the gut.

Based on these findings, we propose a model whereby deficiencies in disaccharidases and hexose transporters alter the milieu of carbohydrates in the distal small intestine (ileum) and proximal large intestine (cecum), resulting in the supply of additional growth substrates for bacteria. These changes manifest in significant and specific compositional changes in the microbiota of AUT-GI children (see Figure 7A–C).

The authors discuss a potential feedback loop of effects of intestinal microbes and nutritional processing, and of the known and potential effects of altered bacterial populations.

Additionally, intestinal microbes can influence the expression of disaccharidases and transporters [59] through the influence of pathogen-associated molecular patterns (PAMPs) and butyrate (a byproduct of bacterial fermentation) on CDX2 expression and activity [60], [61], [62], [63]. In this regard, the observation that CDX2 was decreased in AUT-GI children with increased levels of Betaproteobacteria may be important.

Whatever the underlying mechanisms, reduced capacity for digestion and transport of carbohydrates can have profound effects. Within the intestine, malabsorbed carbohydrates can lead to osmotic diarrhea [64]; non-absorbed sugars may also serve as substrates for intestinal microflora that produce fatty acids and gases (methane, hydrogen, and carbon dioxide), promoting additional GI symptoms such as bloating and flatulence [65].

This is very similar to the underlying theory of the Specific Carbohydrate Diet, impaired carbohydrate digestion promotes bacterial imbalances in the intestine by altered food availability, leading to gastrointestinal distress.

Because of the varied nature of the protein imbalances found and absence of the common alleles associated with such conditions, the authors report that it is unlikely that the underlying cause of the imbalances is genetically based.

In our study, 93.3% of AUT-GI children had decreased mRNA levels for at least one of the three disaccharidases (SI, MGAM, or LCT). In addition, we found decreased levels of mRNA for two important hexose transporters, SGLT1 and GLUT2. Congenital defects in these enzymes and transporters are extremely rare [40], [41], and even the common variant for adult-type hypolactasia was not responsible for reduced LCT expression in AUT-GI children in this cohort. Therefore, it is unlikely that the combined deficiency of disaccharidases (maldigestion) and transporters (malabsorption) are indicative of a primary malabsorption resulting from multiple congenital or acquired defects in each of these genes.

There are a couple of ideas presented on what might be causing the altered disaccharide transporter levels, with food composition intake, immune or hormonal irregularities, and bacterial populations and their associated fermentation byproducts listed as possible candidates.  This study did not attempt to determine if any of these things were actually responsible, but an upcoming paper will also detail qualitative differences in expression of genes involved with inflammation in the autism group.

Regarding bacterial populations found, there were several differences identified by bacterial classification and location as well as some associations with onset of autistic behaviors and GI distress.

Pyrosequencing analysis of mucoepithelial bacteria revealed significant multicomponent dysbiosis in AUT-GI children, including decreased levels of Bacteroidetes, an increase in the Firmicute/Bacteroidete ratio, increased cumulative levels of Firmicutes and Proteobacteria, and increased levels of bacteria in the class Betaproteobacteria.

Stratification of AUT-GI children based on the timing of GI symptom development relative to autism onset revealed that the levels of Clostridiales and cumulative levels of Lachnospiraceae and Ruminococcaceae were significantly higher in AUT-GI children for whom GI symptoms developed before or at the same time as the onset of autism symptoms compared to AUT-GI children for whom GI symptoms developed after the onset of autism and compared to Control-GI children. However, we cannot discern whether changes in Clostridiales occurred before the onset of autism in this subgroup. We can only conclude that increased levels of Clostridiales members in biopsies taken after the development of both GI symptoms and autism are associated with the timing of GI onset relative to autism onset in this cohort. Although the reason for this association remains unclear, this finding may suggest that the timing of GI onset relative to autism is an important variable to consider in the design of future prospective studies investigating the microbiota of children with autism.

I am in love with the appreciation of the subtlety on display at the end, it may not be sufficient to simply categorize by GI and non GI autism, but also by the temporal relationship to onset of behavioral symptoms.  It makes for a messy outlook going forward, but one based on pragmatism as far as coming to valid conclusions.

As is appropriate, the authors end with an admission that we are still largely groping in the dusk about how the microbiome interacts with our tightly coupled systems, but give a variety of reasons to believe that what we do know makes system wide effects reasonable and a relationship with autism plausible.

Metabolic interactions between intestinal microflora and their hosts are only beginning to be understood. Nonetheless, there is already abundant evidence that microflora can have system-wide effects [76], [77], [78], [79], [80], [81], [82], [83] and influence immune responses, brain development and behavior [24], [25], [26], [84], [85].

No kidding!

It should be noted that this paper is a child of a 2010 IMFAR abstract, Intestinal Inflammation, Impaired Carbohydrate Metabolism and Transport, and Microbial Dysbiosis in Autism.  If I understand correctly, another paper is being prepared regarding the findings of intestinal inflammation that will be complimentary to Impaired Carbohydrate Digestion and Transport and Mucosal Dysbiosis in the Intestines of Children with Autism and Gastrointestinal Disturbances.  I’ll try to post something when it is published.

This study was small, with only twenty two participants, largely as a result of the difficulty in obtaining tissue specimens.  While this does give us cause for caution, it is important to note that this research does not exist in a vacuum, but rather, as a much larger set of research that tell us that the relationship between GI complaints and autism is more than the inceptions of DAN doctors.  Previously, Gastrointestinal abnormalities in children with autistic disorder, performed similar biochemistry and reported broadly consistent carbohydrate digestion problems, ‘Low intestinal carbohydrate digestive enzyme activity was reported in 21 children (58.3%), although there was no abnormality found in pancreatic function.’  Several other papers analyzing fecal samples have reported altered bacterial populations, including Low relative abundances of the mucolytic bacterium Akkermansia muciniphila and Bifidobacterium spp. in feces of children with autism, Gastrointestinal flora and gastrointestinal status in children with autism–comparisons to typical children and correlation with autism severity, Fecal lactoferrin and Clostridium spp. in stools of autistic children, and Pyrosequencing study of fecal microflora of autistic and control children, among others.

If the findings from this latest paper are spurious finding based on sample size problems, a lot of other studies are coincidentally finding the same type of thing in the wrong way.   Does anyone think that is likely?

I entered the autism world and online autism debate from a place of seeing with my own eyes the failures of a toddlers GI function and the difficult to overstate changes in that toddler alongside improvements in his GI health.  On one of the first autism blogs on which I participated I got into a discussion (argument?) with a blogger who I came to respect very much, but has since moved on.  I described the fact that my son had six or more diarrhea stools, a day, every day, for months on end, and that when we added dietary changes, probiotics, and later antifungal agents, the changes to his GI function were profound and impossible to misinterpret.  He told me something along the lines that humans were susceptible to illusions and sleight of hand, and I thought, ‘as if not knowing the difference between diarrhea and a log was along the lines of figuring out where the jack of spades went!’.   I couldn’t believe, could not fucking believe, someone would try to convince me that I had imagined my sons problems, and associated recovery.   This wasn’t a sugar pill study where I was asked if my child acted more or less hyperactive, this was a matter of asking myself, ‘How many diarrhea diapers did I change today?  Six?  Or Zero?’  [Repeat once a day for 180 days.]

I doubt this is necessary, but just in case, I will go on the record to state that it is easy, very easy, to tell the difference between a condition of chronic diarrhea and normal GI function.  There might not be a more simpleminded determination to make on Planet Earth or indeed, our perceptible universe.   This is a situation that is susceptible to placebo effects only in the most elaborate imaginations of people who have never experienced chronic GI problems.

From that time on, with nearly zero exceptions, I have become a little less shocked, but a little more saddened by the doublethink style skepticism applied to GI distress and autism in nearly every single conversation I have ever seen on the Internet.  I’ve put some time thinking toward this, why so many otherwise intelligent people house such extreme hostility on a relationship between GI function and autism.  I believe that the Wakefield / MMR autism debacle is at the heart of this disconnect; his ill fated and now retracted paper that launched a thousand Internet scribbles has seemingly forever tied GI complaints and autism to bad science.

It doesn’t have to be this way.  As a community, the vaccine wars and kissing cousin prevalence question has done a lot to fracture us, and very little to unite us.  That is a sad statement, and nothing makes it more unfortunate than the fact that it does not have to be this way.  Wakefield can be wrong about the MMR and there can still be very real differences in GI function in some cases of autism.  We can respectfully disagree about how well our existing prevalence studies inform us on the incidence of autism without also needing to accept a world view where every child with autism has raging bowel problems.

We should have the intellectual honesty to admit that there is nothing inherently dangerous about acknowledging what the data tells us; GI function seems to be abnormal in a subset of children with autism, and the underlying features of that GI distress are qualitatively different than what is found in ‘normal’ children.

-          pD

Lately I’ve found myself reading papers and knowing and owning several of the references; tragically I can’t tell if I’m reading the right research and am onto something, or I am chasing phantoms and my web of pubmed alerts and reading interests are funneling my reference list into a narrowing echo chamber of sorts.   With that warning in mind, we can proceed to poking around several papers, only some of which mention autism per se.  Along the way, we will see evidence supporting the possibility of a biologically plausible mechanism of developmental programming of the neuroimmune environment, a sequence of events that may lead to impaired synaptic pruning in (some cases of?) autism.

By now, everyone has seen/read/heard about one form or another of the ‘a massive asteroid is going to destroy the world’ story.  One of the common survival strategies from an asteroid strike involves altering the path of the asteroid so that it misses the Earth.  The thoughtful analysis of this problem allows for the physics based reality of the problem, moving an asteroid out of an extinction based trajectory involves just a little work when the asteroid is ten thousand gazillion miles away, but a lot more work when it is only a gazillion miles away.  Upon careful evaluation living organisms display similar behavior, relatively minor disturbances in early life can alter the developmental trajectory, while that same disturbance later in life is unable to materially affect the organism beyond a transient effect.   The accumulated evidence that early life experiences can shape the adult outcome is nearly impossible to dispute with any remaining intellectual honesty, the question is instead, is how large is the effect in autism?

This analogy adequately symbolizes one of the more beautiful and terrifying concepts I’ve come across researching autism, that of ‘developmental programming’, which I blogged some about here, but essentially is the idea that there are critical timeframes during which environmental impacts can have long term persistent effects on a wide range of outcomes.  The most robustly replicated findings involve changes to metabolic profiles in response to abnormal prenatal nutritional environments, but there is also evidence of various other effects, including neurological, and reputable speculation, that autism, may in fact, be in part, a disorder of developmental programming.

Secondarily, there has long been speculation of problems in the removal of ‘excess’ synapses, i.e., ‘synaptic pruning’ in the autism population.   This culling of synapses begins in fetal life continuing throughout adolescence and the repeated observations of increased head circumference during infancy as a risk factor for autism has resulted in the idea that altered synaptic pruning maybe involved in autism.

In the last month or so several rather serendipitously themed papers have been published with tantalizing clues about some of the finer grained mechanisms of synaptic pruning, the possibility of impaired synaptic pruning in the autism population, and a known risk factor for autism that models a developmental programming event sequence that may tie them together.

First off, we have Synaptic pruning by microglia is necessary for normal brain development, (Paolicelli et all) with a very straightforward title, that has this dynamite in the abstract: (snipped for length)

These findings link microglia surveillance to synaptic maturation and suggest that deficits in microglia function may contribute to synaptic abnormalities seen in some neurodevelopmental disorders.

This paper is short, but pretty cool, and very nice from a new territory perspective.  It also speaks directly towards one of the increasingly hilarious obfuscations you will sometimes see raised in online discussions about immunological findings in autism, namely, that we can’t know if the state of chronic inflammation in the CNS observed in autism is harmful or beneficial.   [hint: It might not be causative, but it isn’t beneficial.]

Here’s is a snippet from the Introduction:

Time-lapse imaging has shown that microglia processes are highly motile even in the uninjured brain and that they make frequent, but transient contact with synapses. This and other observations have led to the hypothesis that microglia monitor synaptic function and are involved in synapse maturation or elimination.  Moreover, neurons during this period up-regulate the expression of the chemokine fractalkine, Cx3cl1, whose receptor in the central nervous system is exclusively expressed by microglia and is essential for microglia migration. If, in fact, microglia are involved in scavenging synapses, then this activity is likely to be particularly important during synaptic maturation when synaptic turnover is highest.

Nice.  A time dependent participation by microglia in the critical process of optimization of neuron numbers, a process we are still very much groping our way in the dark towards untangling.  The researchers focused in on a particular molecular target, a chemical messenger of the immune system, fractalkine, and found that without fractalkine, the process of synaptic turnover was impaired.

A couple of tests were performed, first immunohistochemistry (i.e., exceedingly clever manipulation of antibodies to determine the presence or absence of proteins in very specific locations) which demonstrated that microglia were, in fact, ‘engulfing synaptic material’ in animals during periods of synaptic maturation.

Secondly, so called ‘knock out mice’ (i.e., genetically engineered mice constructed without the ability to make a specific protein, in this case, fractalkine) were used evaluate for changes in synaptic form and function based on a lack of fractalkine.  Changes in dendritic spine density were observed in the knock out mice group, with much higher densities in a very specific type of neuron during the second and third postnatal week of life.  The authors indicate this is a key timeframe in synaptic pruning, and state their findings are “suggesting a transient deficient synaptic pruning in Cx3cr1 knockout mice “.  The effect of not having fractalkine on spine density was time dependent as shown below.

Several other measurements were taken, including synaptic firing frequencies, which also implicated an increased surface area for synapses on dendritic spines, consistent with impaired pruning.  Time dependent effects on synaptic efficiency and seizure susceptibility were also found, which the led the authors to conclude that the findings were “consistent with a delay in brain circuit development at the whole animal level.”

For additional evidence of fractalkine participation in synaptic maintenance, we can look to the opposite direction, where researchers evaluating neuron loss in an Alzheimers model reported “Knockout of the microglial chemokine receptor Cx3cr1, which is critical in neuron-microglia communication, prevented neuron loss”.  Taken together, the conclusion that fractalkine processing is involved with neuron maintenance is highly likely, and correspondingly, highly unlikely to be a set of spurious findings.

There’s a couple paragraphs on potential mechanisms by which fractalkine could be interacting with microglia to achieve this effect, with the authors claiming that their data and other data generally supports a model wherein microglia were not effectively recruited to appropriate locations in the brain due to a lack of fractalkine, or, a ‘transient reduction in microglia surveillance.’

The conclusion is a good layman level wrap up that speaks toward the Interconnectedness of the brain and the immune system:

In conclusion, we show that microglia engulf and eliminate synapses during development. In mice lacking Cx3cr1, a chemokine receptor expressed by microglia in the brain, microglia numbers were transiently reduced in the developing brain and synaptic pruning was delayed. Deficient synaptic pruning resulted in an excess of dendritic spines and immature synapses and was associated with a persistence of electrophysiological and pharmacological hallmarks of immature brain circuitry. Genetic variation in Cx3cr1 along with environmental pathogens that impact microglia function may contribute to susceptibility to developmental disorders associated with altered synapse number. Understanding  microglia-mediated synaptic pruning is likely to lead to a better understanding of synaptic homeostasis and an appreciation of interactions between the brain and immune system

That’s all pretty cool, but there was precious little discussion of autism, except in the general sense of a ‘developmental disorder associated with altered synapse number’.   [But the references do speak to autism, the first reference provided, Dendritic Spines in Fragile X Mice displays a significant relationship to autism, and it describes how another flavor of knock out mice, this time designed to mimic Fragile-X, exhibit a ‘developmental delay in the downregulation of spine turnover and in the transition from immature to mature spine subtypes.’  Go figure!]

The other reason Paolicelli is of particular interest to the autism discussion is one of the major players in this study, the microglia (i.e., the resident immune cells of the CNS), have been found to be ‘chronically activated’ in the autism brain by direct  measurement in two studies (here, and here, [and by me, here]), and tons of other studies have shown indirect evidence of an ongoing state of immunological alertness in the autism brain.

Considering this is a brand new paper, I do not believe that there are any studies illuminating the results of a state of chronic activation of microglia on the process of synaptic pruning per se.  I will, however, go on the record that such an effect is very, very likely, and the logical leap is microscopically small that there will be some detrimental impact to such a state.  The inverse argument, a scenario wherein there could be a state of chronic microglial activation that does not interfere with microglia participation in the synaptic pruning requires logical acrobatics worthy of Cirque Du Soleil.  I am open to evidence, however.

So, from Paolicelli, we know that a ‘transient reduction in microglial surveillance’ induced by a reduction in the ability to production fractalkine can result in a condition ‘consistent with a delay in brain circuit development at the whole animal level’.

Next up, we have a paper that was all over the JerkNet in the days and weeks following its release, Neuron number and size in prefrontal cortex of children with autism.  This is a cool study, and likely a very important paper, but I must say that a lot of the online commentary exhibits an irrational exuberance towards one part of the findings.   Here is part of the abstract.

Children with autism had 67% more neurons in the PFC (mean, 1.94 billion; 95% CI, 1.57-2.31) compared with control children (1.16 billion; 95% CI, 0.90-1.42; P = .002), including 79% more in DL-PFC (1.57 billion; 95% CI, 1.20-1.94 in autism cases vs 0.88 billion; 95% CI, 0.66-1.10 in controls; P = .003) and 29% more in M-PFC (0.36 billion; 95% CI, 0.33-0.40 in autism cases vs 0.28 billion; 95% CI, 0.23-0.34 in controls; P = .009). Brain weight in the autistic cases differed from normative mean weight for age by a mean of 17.6% (95% CI, 10.2%-25.0%; P = .001), while brains in controls differed by a mean of 0.2% (95% CI, -8.7% to 9.1%; P = .96). Plots of counts by weight showed autistic children had both greater total prefrontal neuron counts and brain weight for age than control children.  [PFC == prefrontal cortex]

Essentially the authors used a variety of mechanisms to measure neuron number in a specific area of the brain, the prefrontal cortex, and found large variations (increases) in the autism group.   The prefrontal cortex is thought to be involved in ‘planning complex coginitive behaviors’, and ‘moderating correct social behavior’, among others, so this was a smart place to look.

The implicit hype on the internet is that this firmly indicates a ‘prenatal cause’ to autism, but if you read the paper, read what Courchense has said, and read recent literature, you know that the simplicity of this as a singular prenatal cause of autism is long broad strokes, and short on appreciation of the subtlety that textures reality.

A link @ LBRB sent me to the team at The Thinking Person’s Guide To Autism, who had a very nice transcription of a talk given by Courchesne at IMFAR 2011.  Here is a snipet that started my wheels turning.

What we see in autism is either an excess proliferation, producing an overabundance of neuron numbers, or the excess might be due to a reduced ability to undergo naturally occurring cell death. Or it could be both. We don’t know which and our data don’t speak to that, although our data do suggest that it’s probably both.

Finally, our evidence shows that across time, there’s a prolonged period of apoptosis, removal and remodeling of circuits. In order to get back to where neuron numbers are supposed to be, it takes a very long time for the autistic brain. In the normal developing brain, this takes just a few months. In autism, it’s a couple of decades.

[Note how well this fits within the model described by Paolicelli, i.e., “consistent with a delay in brain circuit development at the whole animal level”.  ]

I would highly recommend anyone who has read this far to go read the entire post @ TPGTA sometime.

As far as synaptic pruning goes, here is the associated segment of the paper:

Apoptotic mechanisms during the third trimester and early postnatal life normally remove subplate neurons, which comprise about half the neurons produced in the second trimester. A failure of that key early developmental process could also create a pathological excess of cortical neurons. A failure of subplate apoptosis might additionally indicate abnormal development of the subplate itself. The subplate plays a critical role in the maturation of layer 4 inhibitory functioning as well as in the early stages of thalamocortical and corticocortical connectivity development.inhibitory functioning and defects of functional and structural connectivity are characteristic of autism, but the causes have remained elusive.

Nearly half of the neurons in the area studied are expected to be removed through pruning, a process that extends well after birth.  That is something that you didn’t see referenced in too many places trumpeting this study as ‘proof’ that autism was caused by disturbances in the prenatal environment.  I’m not coming down on the prenatal environment as a critical timeframe for autism pathogensesis, just the difficult to defend underlying notion that this is the only time the environment should be evaluated, or the idea that if something is initiated prenatally other timeframes are therefore, unimportant.

So, I’d read that microglia were actively involved in proper synaptic pruning, contingent on utilization of fractalkine, and then read that impaired synaptic apoptotic mechanisms could be participating in autism, with a consequence of an over abundance of neurons.

Then, I got myself a copy of Microglia and Memory: Modulation by Early-Life Infection, which is another study in a growing body of evidence that immune challenges early in life can have unpredictable physiological consequences.  (This is another very cool paper with Staci Bilbo as an author, whom I think is seriously onto something.)  This study, in particular, focused on interactions microglia and formation of memories.   Here is the abstract:

The proinflammatory cytokine interleukin-1ß (IL-1ß) is critical for normal hippocampus (HP)-dependent cognition, whereas high levels can disrupt memory and are implicated in neurodegeneration. However, the cellular source of IL-1ß during learning has not been shown, and little is known about the risk factors leading to cytokine dysregulation within the HP. We have reported that neonatal bacterial infection in rats leads to marked HP-dependent memory deficits in adulthood. However, deficits are only observed if unmasked by a subsequent immune challenge [lipopolysaccharide (LPS)] around the time of learning. These data implicate a long-term change within the immune system that, upon activation with the “second hit,” LPS, acutely impacts the neural processes underlying memory. Indeed, inhibiting brain IL-1ß before the LPS challenge prevents memory impairment in neonatally infected (NI) rats. We aimed to determine the cellular source of IL-1ß during normal learning and thereby lend insight into the mechanism by which this cytokine is enduringly altered by early-life infection. We show for the first time that CD11b+ enriched cells are the source of IL-1ß during normal HP-dependent learning. CD11b+ cells from NI rats are functionally sensitized within the adult HP and produce exaggerated IL-1ß ex vivo compared with controls. However, an exaggerated IL-1ß response in vivo requires LPS before learning. Moreover, preventing microglial activation during learning prevents memory impairment in NI rats, even following an LPS challenge. Thus, early-life events can significantly modulate normal learning-dependent cytokine activity within the HP, via a specific, enduring impact on brain microglial function.

Briefly, the authors infected rats four days after birth with e-coli, and then challenged them with LPS in adulthood to simulate the immune system to evaluate if memory formation was affected.   As further evidence of an immune mediated effect, prevention of microglial activation in adulthood was sufficient to attenuate the effect.  Clearly the effect on memory formation was based on the immune system.  (Note:  Most of the studies I’ve read would indicate [i.e., educated guess] that a four day old rat is brain developmentally similar to the third trimester of a human fetus.)  While a terrifying and beautiful expression of developmental programming in its own right, there isn’t much to speak towards synaptic pruning in this paper, except maybe, potentially, one part of their findings.

In our study, CX3CL1 did not differ by group, whereas its receptor was decreased basally in NI rats, implicating a change at the level of microglia.

This is where things get either highly coincidental, or connected.  CX3CL1 is another name for fractalkine, i.e., animals that were infected in early life had decreased expression of the receptor for fractalkine compared to placebo animals, i.e., fractalkine is the same chemical messenger found to be integral in the process of synaptic pruning in Synaptic pruning by microglia is necessary for normal brain development!  From a functionality standpoint, having less receptor is very similar to having less fractalkine; as the animals in Microglial Cx3cr1 knockout prevents neuron loss in a mouse model of Alzheimer’s disease tell us.

If, if synaptic apoptotic processes are impaired in autism, perhaps this is one mechanism of action. The timeline would involve a prenatal immune challenge, which causes a persistent decrease fractalkine receptor expression, which in turn, causes a consequent impairment in synaptic pruning through interference in microglial targeting.  There is near universal agreement that immune disturbances in utero are capable of altering developmental trajectory undesirably, and here, in an animal model, we have evidence that infections are capable of reducing availability of receptors of ligands known to play a critical role in synaptic pruning, the absence of which leads to conditions which are “consistent with a delay in brain circuit development at the whole animal level”. 

Only time, and more research, will tell if this is a pattern, a phantom, or a little of both.

-          pD

These have been rough times for the people who are heavily invested in the kissing cousin theories of autism as a predominantly genetic disorder and the static, or near static rate of autism.  The California twin study that is old news by the time I get this finished showed much different rates of genetic participation than previously believed.  These findings exposed the underlying frailty of gene-based causation theories, namely that some of the most widely referenced studies in the autism literature, studies used repeatedly as a basis for the notion that autism was ‘the most highly heritable neurodevelopmental disorder’, were, in fact, relatively underpowered, and suffered from serious temporal and methodological shortcomings.    

By contrast, the California study looked at two hundred twin pairs, a lot more twins than any previous study and actually performed autism diagnostics on all of the participating children, whereas other studies relied on medical records.  Performing dedicated ADOS diagnosis prospectively on the children allowed the researchers to discern between autism and PDD-NOS, something that not all previous studies were not able to perform, if for no other reason than the DSM-IV wasn’t even released when several of the most often cited studies were published.   This is from the Comment section of the California twin study:

The results suggest that environmental factors common to twins explain about 55%  of the liability to autism. Although genetic factors also play an important role, they are of substantially lower magnitude than estimates from prior twin studies of autism. Nearly identical estimates emerged for ASD, suggesting that ASD presents the same liability spectrum as strict autism.

This is on top of the fact that there is a quiet, but growing acknowledgement of the fact that literally decades of genetic studies have failed to be able to explain more than a fraction of autism cases despite sequencing of tens of thousands of genomes.   This is a very similar situation to a great number of other disorders which we thought we would cure once the human genome was decoded.   [Note: That isn’t to say that we haven’t learned a lot from sequencing the genome, just that we didn’t quite get what we thought we were going to get.]

This ‘double hit’, so to speak, has reached a critical mass such that health officials are making politically shrewd, but refreshingly realistic statements, and dare I say, a sliver of common sense may be about to infiltrate the discussion about autism prevalence.  For example, as pointed out by Sullivan, Tom Insel, head of the National Institute of Mental Health keeps a blog where he recently blogged ‘Autism Spring’, which included this nugget within the context of continued failure of genetic studies to explain any substantial part of autism, “It is quite possible that these heritability estimates were too high. . .” Ouch. (I would recommend the entire blog posting by Mr. Insel.) 

The high heritability estimates, and implicit genetically-mediated cause of autism, are foundational pillars of the argument that autism rates have not changed over time.  Though overused, or used wrongly in many instances, there is a kernel of dispassionate reality behind the statement, ‘there is no such thing as a genetic epidemic’.  Without the crutch of exceedingly high heritability to rely on, the notion of a stable rate of autism loses the only hard science (read: replicable, biologically-plausible), i.e.,genetics, it ever had, and must place complete reliance on the softer sciences (read: unquantifiable, ‘greater awareness’), i.e.,sociology.  This is great news if you love impossible to verify estimates of prevalence and anecdotes about crazy uncle George who would have been diagnosed with autism forty years ago.  However, if you think we should be relying less on psychologists and cultural anthropologists to answer critical questions, and rely more on hard science, this means that the old narrative on autism prevalence holds even less allure than it did in the past, for those of you who thought this was possible.

Before Kid Autism came around, I would occasionally read discussion boards on the creationism versus evolution ‘debate’.  One thing that I noticed was that the creationists would often employ a ‘God of the Gaps’-style argument: anything that couldn’t be explained by science (yet), or anything necessary to support whatever fanciful construct had been erected to protect biblical creation fables, was ascribed to the work of God.  That’s one thing you have to give to God, he (or she!) can handle it all; it didn’t matter what primitive logical test biblical creation was failing to pass, the golden parachute clause was always that God could have just made things that way.  It was a nifty out on the part of the creationists, kind of like a get out of jail free card. The autism prevalence discussion has been working just like this, and the funny part is that the people that are always claiming to have the intellectual high ground, the supposed skeptics, are playing the part of the creationists!  Zing! 

Here is how it works:

Concerned Parent: It sure does seem like there is more autism than there used to be, what with there being X in a thousand kids with it!  That’s much, much more than even ten years ago!  My brothers, sisters and I all knew kids with mental retardation and Down’s syndrome, but we just don’t remember kids like we see today.

Supposed Skeptic: It is diagnostic substitution and ‘greater awareness’; autism incidence has been stable.  The DSM was changed which resulted in more children being labeled.

Concerned Parent:  It sure does seem like there’s more autism than there used to be.  Now there are Y kids in a thousand having autism!  Why does my son’s preschool teacher keep insisting something is changing?

Supposed Skeptic: It is diagnostic substitution and ‘greater awareness’; autism incidence has been stable.  The DSM was changed which resulted in more children being labeled.

Concerned Parent:  What the hell?  Now there are Z kids in a thousand having autism!  When are those genetic studies going to figure autism out, anyway? 

Supposed Skeptic: It is diagnostic substitution and ‘greater awareness’; autism incidence has been stable.  When does the new DSM come out again? 

(Replace X/Y/Z with any progressively larger numbers.)

It doesn’t matter what prevalence number is thrown about–even the astronomical one in thirty-eight figure bandied about for South Korean children didn’t cause so much as a raised eyebrow; the autism equivalent of God of the Gaps, greater awareness and loosening of diagnostic criteria can handle any amount of increase gracefully.  It is the equivalent of an uber-absorbent autism paper towel, capable of soaking up any number of new children with a diagnosis; there is, literally, no amount of an increase that the God of the Gaps can’t handle.   

If, instead the question was posed like this, ‘How much of the apparent increase in autism is real?’, the answer was always, ‘Zero’, regardless of what the current rates of autism were when you asked the question

Then a funny thing happened, a series of studies from several researchers showed a consistent trend of older parents giving rise to more children with autism than younger parents. There were differences between the studies on just how much of an effect an older parent had, but the overall direction of association was clear.  In this instance, there was also the luxury of a plausible biological mechanism that involved the mediator in favor, genetics.  The idea is that advancing age in the parent meant more years for gametes to get knocked by a random cosmic zap or other environmental nastygram and this disturbance created genetic problems down the line for the offspring, a theory I think is probably pretty good.   Once a couple of these studies started to pile up, there was a small shift in the narrative regarding autism prevalence; after all, nobody could bother to try to deny that parents were getting older compared to past generations.  Here is how it looked:

Concerned Parent:  What the hell?  Now there are X kids in a thousand having autism! 

Supposed Skeptic: Greater awareness and diagnostic substitution are primarily responsible for our observations of increased autism, although, ‘a real, small increase’ cannot be ruled out.   

And with that, there was a little less autism prevalence for the God of the Gaps to handle.   It never seemed to bother anyone that implicit in this argument is an impossible to quantify concept ‘small increase’.  If you were to ask someone what rate of autism ‘a small increase’ amounted to with more precision, the answer is whatever amount rises to the level of autism minus the difficult to quantify effect of older parents.  That is some lazy stuff.

Here are some examples of prominent online skeptics discussing the possibility of a true rise in autism.  See if you can detect a pattern.

Here is Stephen Novella pushing The Fairytale in 2009:

While a real small increase cannot be ruled out by the data, the observed increase in diagnostic rates can be explained based upon increased surveillance and a broadening of the definition – in fact autism is now referred to as autism spectrum disorder.

[Here we see the notion that everything can be explained by the God of the Gaps.]

Here is an example of Orac toying around with this filibuster just the other day, in August of 2011:

True, the studies aren’t so bulletproof that they don’t completely rule out a small real increase in autism/ASD prevalence, but they do pretty authoritatively close the door on their being an autism “epidemic.”

These aren’t the only examples, far from it.   Check it out:

It should be noted that the data cannot rule out a small true increase in autism prevalence. (Stephen Novella in 2008)

If the true prevalence rate of autism and ASDs has increased, it has not increased by very much. (David Gorski, 2010)

It should also be noted that all of this research, while supporting the hypothesis that the rise in autism diagnoses is not due to a true increase in the incidence but rather is due to a broadening of the definition  increased surveillance, does not rule out a small genuine increase in the true incidence. A small real increase can be hiding in the data. (Stephen Novella, 2008)

We should have the curiosity to wonder, what, exactly, does small mean in these contexts?  What percentage size increase should we consider small enough to hide within the data?  Five percent?  Ten percent?  What does ‘small’ mean, numerically, within a range?   Is a ten to twenty percent rise in autism rates reason for us to take comfort in the fact that the effect of greater awareness is real?  At what level does the percentage of ‘real’ autism increase mandate more than superficial lip service, more than a paragraph about ‘gene-environment interactions’ at the end of a two-thousand word blog post that takes pride in the intellectual chops of outthinking Jenny McCarthy?  You won’t get anyone to answer this question; they can’t, because they don’t really know what they mean when they say, ‘small’, other than, ‘it can’t be vaccination’. 

How do we know the amount of this increase must, in fact, even be ‘small’?  This becomes especially problematic when we consider the smackdown that the canard of autism as ‘among the most heritable neurological conditions’ has taken as of late.  If the high heritability estimates of autism are incorrect, yet so often repeated as gospel, why should we also assign confidence to the idea that the increase is trivial?  Isn’t one argument the foundation of the other?   Did either really have quality data behind them? 

This is a terrible, awful, horrible, completely fucking idiotic way to address a question as important as whether or not a generation of children is fundamentally different.  We cannot afford the ramifications of being wrong on this, but we seem to find ourselves in an epidemic of otherwise intelligent people willing to accept the pontifications of cultural anthropologists and the feebleness of social scientists on this critical question.   I am not arguing against the realities of diagnostic switching and greater awareness affecting autism diagnosis rates.  But we can understand that while they are a factor, we must also admit that we have little more than a rudimentary understanding of these impacts, and when we consider the implications of being incorrect, the potential disaster of a very real, not ‘small’ increase in the number of children with autism, we shouldn’t be overselling our knowledge for the sake of expedient arrival at a comforting conclusion.   We should be doing the opposite.

If we can’t have the robustly defendable values on autism rates right now, that’s fine, because that is the reality, but we should at least have the courage to acknowledge this truth.  This is the nature of still learning about something, which we are obviously doing in terms of autism, but in that situation, we don’t have the currency of scientific debate, decent data, to be saying with authority that any true increase in autism is small. 

Unfortunately for the purveyors of The Fairytale, things are going to get a lot worse.  The problem is that we are starting to identify extremely common, in some cases, recently more common, environmental influences that subtly increase the risk of autism.  These are further problems for a genetic dominant model and effectively mandate that the ‘small increase’ is going to have to start getting bigger as a measurement, with a correlated decrease in the amount of autism that cultural shuffling can be held responsible for.  Will anyone notice?

By way of example, we now have several studies that link the seasons of gestation with neurodevelopmental disorders including autism and schizophrenia; i.e., Season of birth in Danish children with language disorder born in the 1958-1976 period, Month of conception and risk of autism, or Variation in season of birth in singleton and multiple births concordant for autism spectrum disorders, which includes in the abstract, “The presence of seasonal trends in ASD singletons and concordant multiple births suggests a role for non-heritable factors operating during the pre- or perinatal period, even among cases with a genetic susceptibility.”  Right!  As I looked up some of these titles, I found that the evidence for this type of relationship has been well known for a long time; schizophrenia, in particular has a lot of studies in this regard, i.e., Seasonality of births in schizophrenia and bipolar disorder: a review of the literature, which is a review of over 250 studies that show an effect, and I also found Birth seasonality in developmentally disabled children, which includes children with autism and was published in 1989, which is like 1889 in autism research years. 

Our seasons have remained constant (but probably won’t stay too constant for much longer. . . ), but this still throws a whole barrel of monkey wrenches into the meme of a disorder primarily mediated through genetics. 

More damning for the Fairytale are some studies presented at this year’s IMFAR, and some others just published, that tell us that abnormal immune profiles during pregnancy appear to provide slightly increased risk for autism, roughly doubling the chance of a child receiving a diagnosis.  The groovy part is that the studies utilized both direct and indirect measurements of an activated immune system to draw similar conclusions, a sort of biomarker / phenotype crossfire.

From the direct measurement end, we have Cytokine Levels In Amniotic Fluid : a Marker of Maternal Immune Activation In Autism?, which reports that mothers with the highest decile of tnf-alpha levels in the amniotic fluid had about a one and a half times increased risk for autism in their children.  This makes a lot of sense considering the robustness of animal models of an acute inflammatory response during pregnancy and its impact on behavior. 

Another study, this one from the MIND Institute in California (which I love), is Increased mid-gestational IFN-gamma, IL-4, and IL-5 in women giving birth to a child with autism: a case-control study (full paper). They found that in pregnant mothers, increased levels of IFN-gamma led to a roughly 50% increased risk of an autism diagnosis.  Here is a snipet:

The profile of elevated serum IFN-γ, IL-4 and IL-5 was more common in women who gave birth to a child subsequently diagnosed with ASD. An alternative profile of increased IL-2, IL-4 and IL-6 was more common for women who gave birth to a child subsequently diagnosed with DD without autism.

This study took a lot of measurements, and goes to great lengths to explicitly call for additional analysis into the phenomena.   IFN-gamma is typically considered pro-inflammatory, while IL-4 and IL-5 are considered regulatory cytokines.  In order to determine if these findings were chance or not, the researchers determined if there was a correlation between the levels of IFN-gamma, IL-4, and IL-5, which they reported with very robust results.    Less clear is what might be causing these profiles, or how, precisely, they might give rise to an increased risk of autism.  The interconnectedness of the brain and the immune systemwould be a good place to start looking for an answer to the last question though. 

What about indirect measurements? It just so happens, another paper was published at IMFAR this year that observed the flip side of the coin, conditions associated with altered cytokine profiles in the mother and this study also found an increased risk of autism.  The Role of Maternal Diabetes and Related Conditions In Autism and Other Developmental Delays, studied a thousand children and the presence of diabetes, hypertension, and obesity in their mothers in regards to the risk of a childhood autism diagnosis.   The findings indicate that having a mother with one or more of those conditions roughly doubles the chances of autism in the offspring.  Obesity, in particular, has an intriguing animal model Enduring consequences of maternal obesity for brain inflammation and behavior of offspring, a crazy study that I blogged about when it was published.   A variety of auto immune disorders in the parents have been associated with an autism diagnosis in several studies. 

The obesity data is particularly troublesome for the idea of a ‘small’ increase in autism, just like parents have been getting older, parents have also been getting fatter, waaaay fatter, (and more likely to have diabetes)  the last few decades.  There isn’t any squirming out of these facts.  If, indeed, being obese or carrying associated metabolic profiles is associated with an increased risk of autism, ‘small’ is getting ready to absorb a big chunk of real increase.  But is there any clinical data to support this possible relationship, do we have any way to link obesity data with this autism data from the perspective of harder figures?

It further turns out, there are some very simple to navigate logical jumps between the above studies.  Remembering that our clinical measurements indicated that increased INF-gamma, IL-4, and IL-5 from the plasma of the mothers was associated with increased risk, we can see very similar patterns in Increased levels of both Th1 and Th2 cytokines in subjects with metabolic syndrome (CURES-103).  Here is part of the abstract, with my emphasis.

Metabolic syndrome (MS) is a cluster of metabolic abnormalities associated with obesity, insulin resistance (IR), dyslipidemia, and hypertension in which inflammation plays an important role. Few studies have addressed the role played by T cell-derived cytokines in MS. The aim of the tudy was to look at the T-helper (Th) 1 (interleukin [IL]-12, IL-2, and interferon-gamma [IFN-gamma]) and Th2 (IL-4, IL-5, and IL-13) cytokines in MS in the high-risk Asian Indian population.

Both Th1 and Th2 cytokines showed up-regulation in MS. IL-12 (5.40 pg/mL in MS vs. 3.24 pg/mL in non-MS; P < 0.01), IFN-gamma (6.8 pg/mL in MS vs. 4.7 pg/mL in non-MS; P < 0.05), IL-4 (0.61 pg/mL in MS vs. 0.34 pg/mL in non-MS; P < 0.001), IL-5 (4.39 pg/mL in MS vs. 2.36 pg/mL in non-MS; P < 0.001), and IL-13 (3.42 pg in MS vs. 2.72 pg/mL in non-MS; P < 0.01) were significantly increased in subjects with MS compared with those without. Both Th1 and Th2 cytokines showed a significant association with fasting plasma glucose level even after adjusting for age and gender. The Th1 and Th2 cytokines also showed a negative association with adiponectin and a positive association with the homeostasis model of assessment of IR and high-sensitivity C-reactive protein.

Check that shit out!  Seriously, check that out; increased IFN-gamma, IL-4, and IL-5 in the ‘metabolic syndrome’ group, comprised of people with, among other things, obesity, insulin resistance, and hypertension; the same increased cytokines and risk factors found to increase the risk of autism. 

If we look to studies that have measured for TNF-alpha in the amniotic fluid during pregnancy, we quickly find,  Second-trimester amniotic fluid proinflammatory cytokine levels in normal and overweight women

There were significant differences in amniotic fluid CRP and TNF-alpha levels among the studied groups: CRP, 0.018 (+/-0.010), 0.019 (+/-0.013), and 0.035 (+/-0.028) mg/dL (P=.007); and TNF-alpha, 3.98 (+/-1.63), 3.53 (+/-1.38), and 5.46 (+/-1.69) pg/mL (P=.003), for lean, overweight, and obese women, respectively. Both proinflammatory mediators increased in women with obesity compared with both overweight and normal women (P=.01 and P=.008 for CRP; P=.003 and P=.01 for TNF-alpha, respectively). There were significant correlations between maternal BMI and amniotic fluid CRP (r=0.396; P=.001), TNF-alpha (r=0.357; P=.003) and resistin (r=0.353; P=.003).

Nice. 

What we are really looking at are five studies the findings of which speak directly to one another; a link to metabolic syndrome during pregnancy and increased IFN-gamma, IL-4, and IL-5, a link to obesity and hypertension in pregnant mothers and autism risk, and an increased risk of autism in mothers wherein IFN-gamma, IL-4, and IL-5 were found to be increased outside of placenta.   Further, we have a link between amniotic fluid levels of TNF-alpha and metabolic syndrome, metabolic syndrome in mothers and autism risk, and increased risk from increased tnf-alpha in the amniotic fluid. 

As I have said previously, one thing that I have learned during this journey is that when we look at a problem in different ways and see the same thing, it speaks well towards validity of the observations.  What we see above is a tough set of data to overcome; we need several types of studies looking at the relationship between metabolic syndrome, immune profiles during pregnancy, and autism from different angles to have reached the same wrong conclusion, something that is increasingly unlikely.  We are in an epidemic of obesity and the associated endocrine mish mash of metabolic syndrome, there simply isn’t any diagnostic fuzziness on this.  It is happening all around us.  Even though the total increase in risk is relatively small, the sheer quantity of people experiencing this condition of risk mandates that the numbers game looks favorable towards a real increase in autism.  If we acknowledge this, how can we continue to have faith in the concept that any true increase in the autism rates must be ‘small’?

Is the next argument going to be that besides increased parental age, and heavier or more diabetic mothers, the rest of the autism increase is the result of diagnostic three card monte?  (Just how much is the rest, anyways?)

And even though these studies, and likely more in the future, expose the crystal delicate backbone of the ‘small true increase’ argument, I have great pessimism that the people so enamored with invoking this phrase will ever acknowledge its shifting size, much less the implications of being wrong on such a grand scale.

-          pD

One of the more beautiful and terrifying concepts I’ve come across in the last year or so is the idea of ‘developmental programming’, or sometimes fetal programming, or as I imagine it will eventually be recognized, the realization of subtle change is still change, and subtle change during critical timeframes can amplify into meaningful outcomes.  The underlying hypothesis is that environmental influences during early life, gestation, infancy, or even childhood, have the capacity to permanently influence physiological and behavioral state into adulthood.  The available evidence implicates the potential for developmental programming to be involved with an assortment of conditions that on the whole, you’d rather not have than have, including the spectrum sized set of disorders grouped as ‘metabolic syndrome’ that incorporates several risk factors for cardiovascular disorders, obesity, type II diabetes.  There is also less pronounced evidence for some autoimmune disorders, and perhaps, autism. 

Here is the most concise explanation of developmental programming I’ve seen so far, from Developmental Programming of Energy Balance and Its Hypothalamic Regulation

The concepts of nutritional programming, fetal programming, fetal origins of adult disease, developmental origins of health and disease, developmental induction, and developmental programming were all conceived to explain the same phenomenon: a detrimental environment during a critical period of development has persistent effects, whereas the same environmental stimulus outside that critical period induces only reversible changes.

I am absolutely in love with the importance of time dependent effects, a sort of combo pack of why the dose doesn’t always make the poison, and the importance of understanding subtle interactions in developing systems. 

The area of developmental programming that has a ton of research in the human field and animal models is the link between metabolic syndrome and a differently structured uterine and/or early postnatal environment.  A nice review from 2007, Developmental programming of obesity in mammals (full paper) has this:

Converging lines of evidence from epidemiological studies and animal models now indicate that the origins of obesity and related metabolic disorders lie not only in the interaction between genes and traditional adult risk factors, such as unbalanced diet and physical inactivity, but also in the interplay between genes and the embryonic, fetal and early postnatal environment. Whilst studies in man initially focused on the relationship between low birth weight and risk of adult obesity and metabolic syndrome, evidence is also growing to suggest that increased birth weight and/or adiposity at birth can also lead to increased risk for childhood and adult obesity. Hence, there appears to be increased risk of obesity at both ends of the birth weight spectrum.

And

Childhood and adult obesity are amongst the cardiovascular risk factors now considered to be ‘programmed’ by early life and, perhaps counter-intuitively, babies subjected either to early life nutritional deprivation or to an early environment over-rich in nutrients appear to be at risk. Supportive evidence includes the observation of a ‘U-shaped’ curve which relates birthweight to risk of adult obesity (Curhan et al. 1996).

[Check out that example of a hormetic dose curve!  Totally sweet!]

The list of papers supporting a link between abnormal gestational or birth parameters and subsequent obesity in the offspring is very, very voluminous.   The satellite level high view of the research starts with Dutch mothers during a time of famine, and the observations that these children were much more likely to be obese at nineteen in Obesity in young men after famine exposure in utero and early infancy.  Later, infants in England were found to have birth weight positively correspond to adult weight in Birth weight, weight at 1 y of age, and body composition in older men: findings from the Hertfordshire Cohort Study (full paper).  A study with twin pairs, Birth weight and body composition in young women: a prospective twin study  had similar findings, but with the additional coolness factor of being able to detect differences between genetically identical twins who happened to be born at different weights.  There are studies on infants that are born light but then ‘catch up’are consistently more likely to be obese, a review of which can be found in Rapid infancy weight gain and subsequent obesity: systematic reviews and hopeful suggestions.  Startlingly, Weight Gain in the First Week of Life and Overweight in Adulthood observed that formula fed babies who gained considerable weight during the first eight days after birth were more likely to be obese as adults, similar to other findings implicating formula fed babies with adult obesity.

Therearealsoconservativelya bazillionanimalmodelsthattellusthatthestudiesin humans are accurate.

Part of me hates the deterministic nature of these findings, it’s really just an extension of the fatalism of genetic assignment, but on the other hand, the data is the data.  I must admit, I am in love with the underlying evolutionary cleverness of the thrifty phenotype end of the U curve on display; a fetus or neonate that is deprived of nutrients, or perhaps, some types of nutrients, programs itself for an environment in which food is scarce, handling calories differently at a very fine grained metabolic level.  From a survival standpoint this modification is most definitely the smart move; all inbound indicators are signaling to the fetus that calorie acquisition is going to be tough on the outside, and as a result, the physiology is tweaked so that baby is ready to make the absolute most of any available nutrients.  If that child, however, is raised in a world with plentiful calories, if not always, beneficial calories, they tend to store fat more readily than a baby/child/adult that did not receive the same messages in utero.  Neat.

Like lots of things I seem to be running into, our observations of what is happening seem to be more advanced than our understanding of how it is happening.  The ideas of developmental programming have been around for a while, but we are still very much in the learning phase regarding mechanism of action, a very thorough review that I ran into can be found here:  Mechanisms of developmental programming of the metabolic syndrome and related disorders.   (full paper). 

Another example of programming a bit closer to home to the autism world has been in the news lately, namely the replication of findings that children who grow up around farm animals, or in some cases, pets, are less likely to suffer from allergies and /or asthma than children who grow up without that exposure.  These findings are also very robust, and appear to implicate similar critical developmental timeframes including the gestational environment, infancy, and toddlerhood. 

Here is an example of the kind of thing in this area,  Farming environment and prevalence of atopy at age 31: prospective birth cohort study in Finland

Cross-sectional studies have shown an association between the farming environment and a decreased risk of atopic sensitization, mainly related to contact with farm animals in the childhood. Objective Investigate the association of a farming environment, especially farm animal contact, during infancy, with atopic sensitization and allergic diseases at the age of 31. Methods In a prospective birth cohort study, 5509 subjects born in northern Finland in 1966 were followed up at the age of 31. Prenatal exposure to the farming environment was documented before or at birth. At age 31, information on health status and childhood exposure to pets was collected by a questionnaire and skin prick tests were performed. Results Being born to a family having farm animals decreased the risk of atopic sensitization [odds ratio (OR) 0.67; 95% confidence interval (CI) 0.56-0.80], atopic eczema ever (OR 0.77; 95% CI 0.66-0.91), doctor-diagnosed asthma ever (OR 0.74; 95% CI 0.55-1.00), allergic rhinitis at age 31 (OR 0.87; 95% CI 0.73-1.03) and allergic conjunctivitis (OR 0.86; 95% CI 0.72-1.02) at age 31. There was a suggestion that the reduced risk of allergic sensitization was particularly evident among the subjects whose mothers worked with farm animals during pregnancy, and that the reduced risk of the above diseases by farm animal exposure was largely explained by the reduced risk of atopy. Having cats and dogs in childhood revealed similar associations as farm animals with atopic sensitization. Conclusion and Clinical Relevance Contact with farm animals in early childhood reduces the risk of atopic sensitization, doctor-diagnosed asthma and allergic diseases at age 31.

That is one hell of a long running study and the findings are consistent with a wealth of similar studies across populations, including Exposure to environmental microorganisms and childhood asthma, and Effect of animal contact and microbial exposures on the prevalence of atopy and asthma in urban vs rural children in India.  These findings are part and parcel with the Hygiene Hypothesis, the idea that a relative reduction in ‘training’ of the immune system can lead to disturbances in normal immune system development and consequent development of autoimmune disorders.   (Here’s a nice review of the evidentiary backing for the Hygiene Hypothesis) From a clinical viewpoint, there are reasons to suspect this is a biologically plausible pathway; in Environmental exposure to endotoxin and its relation to asthma in school-age children the researchers reported an inverse relationship between the amount of endotoxin (i.e., a bacterial fingerprint that is recognized by the immune system) and the immune  response, stating, “Cytokine production by leukocytes (production of tumor necrosis factor alpha, interferon-gamma, interleukin-10, and interleukin-12) was inversely related to the endotoxin level in the bedding, indicating a marked down-regulation of immune responses in exposed children.”  We can also see immunomodulatory effects of farm or rural living in the cytokine profiles of breast milk between two populations, as reported in Immune regulatory cytokines in the milk of lactating women from farming and urban environments, which found much higher concentrations of TGF-Beta1, a critical immune modulator, in breastmilk and collustrum of ‘farm mothers’.  The concentration of TGF-Beta1 in breastmilk had already been implicated in infant development of atopic disease in Transforming growth factor-beta in breast milk: a potential regulator of atopic disease at an early age. 

The evidence supporting developmental programming in these instances is very problematic to overcome, clearly there are mechanisms by which the events of very early life can cause persistent changes to physiology into adulthood; be they changes ‘designed’ to be adaptive, or disturbed trajectories of usually tightly regulated systems that find inappropriate targets in an environment different than what our ancestors evolved in.  I’d note that none of what is above invalidates any findings of genetic involvement with cardiovascular problems, obesity, or asthma, but it should serve as a portrait of how genetic recipes are only part of the process. 

So, what about autism?  This is, admittedly, where things get a bit more speculative, there isn’t the same type of epidemiological evidence in the autism arena as what we see above.  Part of this discrepancy is an artifact of the fuzzy nature of autism, a bazillion different conditions each with their own personalized manifestation, a much more daunting set of variables to detangle compared with measuring BMI, triglyceride levels or asthma.  Those caveats in place, there is still room to discuss some potential examples wherein early life experiences might be participating in ‘programming’ some of what we see in autism. 

A nice review paper that speaks directly towards a developmental programming model that involves autism is Early life programming and neurodevelopmental disorders that includes as an author, Tom Insel, head of the National Institute of Mental Health, and generally, one of the good guys.   This is part of the abstract.

Although the hypothesized mechanisms have evolved, a central notion remains: early life is a period of unique sensitivity during which experience confers enduring effects. The mechanisms for these effects remain almost as much a mystery today as they were a century ago (Insel and Cuthbert 2009). Recent studies suggest that maternal diet can program offspring growth and metabolic pathways, altering lifelong susceptibility to diabetes and obesity. If maternal psychosocial experience has similar programming effects on the developing offspring, one might expect a comparable contribution to neurodevelopmental disorders, including affective disorders, schizophrenia, autism and eating disorders. Due to their early onset, prevalence and chronicity, some of these disorders, such as depression and schizophrenia, are among the highest causes of disability worldwide (World Health Organization, 2002). Consideration of the early life programming and transcriptional regulation in adult exposures supports a critical need to understand epigenetic mechanisms as a critical determinant in disease predisposition.

A concise explanation of the concept of developmental programming and the need for more finely detailed understandings of the likely epigenetic underpinnings.  Also included is a discussion of things like maternal stress during gestation, childhood environmental enrichment (or more specifically, ‘de-enriched’ or otherwise, terrible situations), and prenatal infection models.  Nice.  

What about specifics for the autism arena?  One environmental event that most everyone agrees can increase risk of an autism diagnosis is an immune challenge in the gestational period.  The animal models are robust and have been replicatedacross laboratories and epidemiological data supports an association.  A lot of groups have been studying the effects of maternal immune activation in animal models the past few years, what we can see are some striking parallel veins to what is observed in autism that involve the concept of developmental programming. 

One paper, with a title I love, is  Neonatal programming of innate immune function.  Here is a snipet of the abstract from the first paper:

There is now much evidence to suggest that perinatal challenges to an animal’s immune system will result in changes in adult rat behavior, physiology, and molecular pathways following a single inflammatory event during development caused by the bacterial endotoxin lipopolysaccharide (LPS). In particular, it is now apparent that neonatal LPS administration can influence the adult neuroimmune response to a second LPS challenge through hypothalamic-pituitary-adrenal axis modifications, some of which are caused by alterations in peripheral prostaglandin synthesis. These pronounced changes are accompanied by a variety of alterations in a number of disparate aspects of endocrine physiology, with significant implications for the health and well-being of the adult animal.

Another very cool, and very dense, paper with a salient title and content by the same group is  Early Life Activation of Toll-Like Receptor 4 Reprograms Neural Anti-Inflammatory Pathways (full paper) which reports that a single early life immune challenge results in persistently altered response to immune stimulants into adulthood, with differential responses in the CNS compared to the periphery.  Especially interesting in this paper is that the researchers have dug down a layer into the biochemical changes affected by early life immune challenge and found that alterations to HPA-Axis metabolites are responsible for the changes. 

Tinkering around with the HPA-Axis, an entangled neuroendicrine system that touches on stress response, immune function, mood, and more can have a lot of disparate effects.  It turns out, there is evidence that early life immune challenges can also modify behaviors in a way consistent with altered stress responses.

For example, the very recently published Peripheral immune challenge with viral mimic during early postnatal period robustly enhances anxiety-like behavior in young adult rats has a short, but to the point abstract:

Inflammatory factors associated with immune challenge during early brain development are now firmly implicated in the etiologies of schizophrenia, autism and mood disorders later in life. In rodent models, maternal injections of inflammagens have been used to induce behavioral, anatomical and biochemical changes in offspring that are congruent with those found in human diseases. Here, we studied whether inflammatory challenge during the early postnatal period can also elicit behavioral alterations in adults. At postnatal day 14, rats were intraperitoneally injected with a viral mimic, polyinosinic:polycytidylic acid (PIC). Two months later, these rats displayed remarkably robust and consistent anxiety-like behaviors as evaluated by the open field/defensive-withdrawal test. These results demonstrate that the window of vulnerability to inflammatory challenge in rodents extends into the postnatal period and offers a means to study the early sequelae of events surrounding immune challenge to the developing brain.

The methodology is very similar to what we see in a lot of animal models of early life immune activation, convince a young animals immune system that they are under microbial attack by mimicking either bacterial or viral invaders, and then measure behaviors, or physiology, later in life. This study could be seen as a complement to a much earlier (2005) paper, Early life immune challenge–effects on behavioural indices of adult rat fear and anxiety, which used a different immune stimulant (bacterial fingerprint/LPS versus viral fingerprint/Poly:IC), but which found generally consistent results.

There are more, for example, Early-Life Programming of Later-Life Brain and Behavior: A Critical Role for the Immune System (full paper), which reviews animal study evidence that early life immune challenges can have lifelong effects.  Here is part of the Introduction:

Thus, the purpose of this review is to: (1) summarize the evidence that infections occurring during the perinatal period can produce effects on brain and subsequent behavior that endure throughout an organism’s life span, and (2) discuss the potential role of cytokines and glia in these long-term changes. Cytokines are produced within the brain during normal brain development, but are expressed at much higher levels during the course of an immune response. In contrast to overt neural damage, we present data indicating that increased cytokine exposure during key periods of brain development may also act as a “vulnerability” factor for later-life pathology, by sensitizing the underlying neural substrates and altering the way that the brain responds to a subsequent immune challenge in adulthood. In turn, this altered immune response has significant and enduring consequences for behavior, including social, cognitive, and affective abilities. We discuss the evidence that one mechanism responsible for enduring cytokine changes is chronic activation of brain microglia, the primary immunocompetent cells of the CNS.

Check that out!  We have several papers showing, indeed, a ‘chronic activation of brain microglia’ in the autism population; one way, it seems, to achieve this, is ‘increased cytokine exposure during key periods of brain development’.  (Ouch!) 

Is developmental programming the mechanism by which gestational immune activation raises the risk of autism?  I don’t think we can answer that question with any authority yet, but the logical jumps to arrive at that conclusion are small, and  are supported by a great deal of evidence.  No doubt, we’ll be learning more about this in the years to come.

Ultimately, I think what all of this means is that, as usual, there is another layer of complexity thrown into the mix.  As far as autism goes, it seems likely that at least some of our children are manifesting behaviors consistent with autism as a result of things that happened to them very, very early in their life.  Figuring out if this is happening, how it is happening, and to which individuals, is a daunting, very difficult task; but at least we are approaching a level of knowledge to allow for such an endeavor.

This posting focused on the bad stuff, but the inverse is just as meaningful, having a ‘normal’ gestational period as far as nutrients go, programs you towards a more healthy weight, and being born to a mother exposed to a variety of microbial agents, as the overwhelming majority of mothers were for most of human existence, programs you away from asthma.  But from a broader standpoint, from a ‘every human on the planet’ view, I think we must begin to recognize that everyone is being programmed, in some ways for good, in others, for not so good.  Curiosity and thoughtful analysis is our way to illuminate the beautiful and dispassionate gears that propel the machinations of nature; developmental programming is one of the cogs in the natural world, hopefully, one day, we will acquire the wisdom to refine the program for our benefit, but in the meantime, it is still exciting to witness the discovery of the inner workings.

-          pD

The osmotic pressure of cool people and pop culture tells me that what we used to call one night stands are now called ‘hookups’, casual sexual encounters as convenient that don’t necessarily mean people are dating, but some release can be found, and everyone moves on with their lives until the next time.  This reminds me a lot of how people that ought to know better have been treating autism prevalence studies lately.  The results are useful in cementing an already reached conclusion, but ultimately, the findings are only used as isolated ejaculations of the same ideological tweets.  Last week’s hookup doesn’t mean anything come this Saturday night, and there is absolutely no reason, no reason, anyone should be troubled to compare this weeks findings used to trumped a static rate of autism with last weeks findings.  What we are witnessing is the equivalent of a scientific one night stand, and anyone who doesn’t think the scientific method should be framed for the sake of expediency ought to be furious.

These posts can oftentimes take me a long while to complete, so dating my start point a bit, about two weeks ago, the NHS study from England came out that described a near 1% prevalence of ‘autism’ in adults.  The ‘findings’ from this study actually came to light and received attention in the autism community over a year ago, but the real publication happened in May 2011, so there you are.  

About a week ago, the Korea ‘study’ on autism came out; it hit the web with a large footprint, and amazingly, described an atmospheric autism ‘prevalence’ of autism of near 2.5%, with 1 in 38 (!!!!) Korean children ‘estimated’ to be on the autism spectrum.   If it has not happened already, this study and ‘conclusions’ will soon became part of the autism lexicon; an uber-Kevlar argument, impervious to any concerns involving the possibility of an actual increase in the number of children with autism. 

Both of these studies share very similar methodologies; essentially a lot of people were screened through a questionnaire, a subset of people with ‘high’ scores on the questionnaire were subsequently retested with standard tools for assessing autism.  Based on how well the questionnaire did at predicting autism spectrum diagnosis, an extrapolation, with various ‘corrections’, was made as towards how many people in the general public are on the spectrum.  In both studies, the overwhelming majority of people ‘estimated’ with autism were previously undiagnosed and were not receiving any services. 

Here’s the thing that is driving me up the wall crazy, apeshit mystified and enraged. Nobody cared.  Let’s look again at what these studies found and see if we can detect anything of potential interest in their conclusions when compared between one another.

Nobody, and I mean nobody, took these two studies as evidence of an autism epidemic, despite the fact that here we have two supposedly (?) well designed studies that found entire spectrum sized differences in the number of children and adults with autism!  You could literally drive the old spectrum through the hole in the new spectrum!  If both of these two studies are meaningful, if both have accurately captured autism in their respective target populations, we have no choice but to admit that the epidemic is real, and we have proof that children have an autism spectrum disorder two and a half times more frequently than adults.  There is an epidemic of autism in our children; or at least, in Korean children!

Did anyone see those headlines that I somehow missed?  Did the online skeptical community acknowledge that we now finally have some solid evidence that indeed, autism rates are higher in children than adults, and somehow I failed to see those conversations? 

Here’s what really confuses me.  Some of the same people, same ‘skeptics’, and same news organizations breathlessly reported both of these findings without, apparently, understanding their implications alongside one another.  For example, in 2009, here’s a post from Stephen Novella at Science Based Medicine that touched on the England study that includes this nugget:

They found a consistent prevalence of 1% in all age groups they surveyed. This is remarkable for two reasons – first, they found the exact same 1% figure as the CDC US survey (assuming the CDC data is more accurate than the phone survey published in Pediatrics). This supports the conclusion that the 1% figure may be close to the true prevalence of ASD in the population.

Second, the NHS study found that the prevalence of autism was the same in all age groups, strongly suggesting that true ASD incidence has not been increasing over recent decades and supporting the increased surveillance and definition hypothesis.

Check out how ‘remarkable’ Mr. Novella thinks the 1% matchup between English adults and American children is in terms of making the case for a static rate of autism.  This is a guy whose posts outside the autism realm I tend to enjoy in many instances, he is clearly a superior intellect, and applies a very skeptical eye towards his non-autism posts.  My presumption is that he was well aware that the NHS study actually diagnosed a grand total of 19 adults, and had good reasons, which he declined to illuminate in that post, for why this relatively low number of results was immune to significant confounding problems, which is why it provided such ‘remarkable’ evidence ‘strongly suggesting that true ASD incidence has not been increasing’. 

Then, in May 2011, Mr. Novella posted Autism Prevalence Higher than Thought, concerning the Korea study.  Here is a snippet from the conclusions:

This study adds an interesting data point to the whole picture of ASD. If correct, then the theoretically upper limit of ASD prevalence is about 2.6% of the population, more than twice the previous estimate. It also indicates that when you undergo a program of thorough searching, you will find more diagnoses.

What is going on here?  The England study, which found a prevalence of 1%, the study that previously was found to be remarkable evidence of a static rate of autism was exactly the same type of study, wide-scale screening for likely candidates within the general population, followed by targeted autism assessment of people with high scores, and backwards extrapolation.  Does anyone think that the Korea study was that much more thorough than the England study?  If a study came out tomorrow that reported 5%, or 10% prevalance, would we simply assign this to a even more strenously executed methodology?   Is there any evidence that we might use to suspect a 5% prevalance reported next week in Columbia is faulty that could not also be applied against Korea?

For what reason should we, now, believe that the England study of adults was so fatally flawed that it missed more than one autistic adult for every one it found?  Surely a study capable of missing more than half of the autistic adults had some type of warning signs back in 2009 that might indicate that the evidence might be less than remarkable, maybe questionable, or that, in fact, it might be a Fairytale?

Am I cynical to suggest that what really made the England study such remarkably ‘strong evidence’ of a static rate of autism was that, at the time, it had findings within the statistical range of existing CDC numbers in children?   Was the online and media love affair with the England NHS study little more than prevalence hookup?  Have I reached the theoretical limit of jadedness?

There really isn’t a way to reconcile these two findings without either accepting a two and a half times increase in autism in children versus adults, a sort of epidemic-lite, or accepting that one or both of the studies suffer from serious flaws.  But if we start accepting that the studies might have serious problems, we shouldn’t be saying they are ‘strong evidence’ of anything, except, perhaps, the difficult to overstate problems of autism prevalence studies.  Of course, it is a different ballgame if you are relieved of the intellectual responsibility of actually trying to reconcile the two findings; if you allow yourself the prevalence doublethink that England has meaningful data, and so does Korea, and that the rate of autism isn’t increasing, then, no harm, no foul Big Brother.

One prevalence study that didn’t get the booty call was Brief Report: Prevalence of Pervasive Developmental Disorder in Brazil: A Pilot Study, which came out in February, 2011; just three months before Korea.  Methodology wise, this study is a kissing cousin to Korea and England, a screening was performed in the general population, and assessments were subsequently performed and then statistical extrapolations were performed to reach a prevalence rate.   Let’s see what these values look like up against each other, and see if we can detect a pattern.

Can anyone see a pattern here? 

Now the skeptic might tell you that the Brazil study was a lot smaller, which is true; the initial screening of children only contained a little less than 1,500 children.  But it hardly matters; just to get to the level of English adults ‘found’, they would have had to miss two children for every child they found, and to approach Korea values, they needed to have missed almost nine children for every child actually diagnosed.  Does anyone think this is reality?  Why would prospective screening and backwards extrapolation be so accurate in one population, and so wildly inaccurate in another population?  The Brazil and England study used versions of the same screening questionnaire!

I understand that being partially funded by Autism Speaks, and having a ‘cultural anthropologist’ with a book on the subject of autism carries some weight in the press conference area; so that might explain why one study got press, and another didn’t.  Forgetting the press issue, where are the calls that we should try throwing four thousand Brazilian genomes at a sequencer to see what in their genetic makeup appears to be protecting them from autism so effectively?  Why aren’t these studies meaningful evidence of some environmental force acting to create wildly different rates of autism in these different populations?  

I would note that the press releases, media regurgitations, and skeptical viewpoints nearly all contained the boilerplate note that more studies are needed.    Consider, however, if our need for ‘more study’ is so extensive, if we place so little confidence in our methodologies that papers published within months of each other, with nearly identical study methods, find literally nine times higher rates of autism in one population aren’t a warning sign of an real difference in incidence, what this ought to be telling us is that all of our prevalence data are crapshoots, at best.  We shouldn’t get to pick and choose which studies we think are meaningful because they happen to meet comforting quotas, or discard those that fail to support those palliative notions.

It is tempting to look at the Brazil study and evaluate for design or implementation problems that could cause such startlingly low rates of autism; the authors go into some discussion about the reasons their findings might seem so low.  Complicating matters along this line, however, is that the Brazil and Korea studies, shared a researcher, the relatively well known psychiatrist with a large pubmed autism prevalence footprint, Eric Fombonne.    It occurred to me that it might be a fun experiment to see how reliable Mr. Fombonne has been regarding autism prevalence. 

[Click on the image to get a bigger view / stupid wordpress template]  Note that I have omitted review papers, or papers that had no abstracts, but it doesn’t really help.  (How could it?)

All of these findings were wholly or partially authored by the same person.  Is there anything more damning for the state of autism prevalence research than this person continues to be considered a source of reliable information?  

I used to live with a fun dude in college; he went to engineering school and went on to work at a manufacturing facility near our town.  One of the funniest things he told me about engineering was this quote:

Dilution is the solution to pollution!

In other words, if you have a hundred pounds of diethyl-pthylate-poisonate to dispose of, ship in a hundred thousand gallons of water, and start pumping; if you have two hundred pounds to eject, ship in two hundred thousand gallons of water.  This is what is happening to the definition of autism, the quirky element, the ‘broad autistic phenotype’ is seeping into these studies.   After dozens, or hundreds of prevalence studies we are ultimately left with as many portraits of different entities as envisioned by the researcher and width of spectrum de jour.  The upshot of this, however, is that it makes no sense to try to compare these studies.  

In the meantime, we are told time and time again that even though our common sense, our memories of childhood, and the repeated lamentations from every person who has worked with children for the last few decades, all of which are warning us that something is different; all of these things are all supposedly subject to an array of biases so strong that we cannot trust them to reach any conclusions.  Only through carefully planned, objective analysis can we reach any conclusions on autism incidence.  The results of this choreographed investigation looks like this:

 Does anyone really think there aren’t some pretty serious biases operating here?  If we cannot use common sense to try to reconcile the picture above, what can we use?  If trusting common sense is dangerous to valid conclusions, so is trusting this. 

If anyone really thought that Korea and Brazil were measuring the same condition, a condition that until very, very recently has been considered lifelong and severely debilitating, the two wildly different findings would be cause for alarm, undeniable evidence of a massive environmental force influencing the development of autism in some populations.  But no one thinks this, no one cares, and that is because; no one really believes these studies are measuring the same thing.  But admitting this is dangerous to too many, it is the implicit acknowledgement of just how little we understand, how beholden our policies and research prioritizations are guided by the softest of science and scientists, and ultimately, how frequently we’ve been sold a narrative with the scientifically defendable value of a set of  monetized South Florida mortgages.

Such is the way of the prevalence hookup, transiently entertaining, but without meaning from week to week.   Until we can find a way past this, past reliance on the shifting sands of behavioral assessments that can vary from researcher to researcher (or by the same researcher!), we can perform all of the ‘thorough investigations’ that we can afford and repeat the ‘findings’ that support our meme until we are blue in the face.  None of it will mean a goddamned thing, though we may lose a generation of children while we bounce from one set of findings to another, feeling pleased with the ones that make doom seem unlikely, and discarding the ones that should be cause for great alarm.

-pD

I’ve gotten into a lot of discussions online about the vaccines and autism; generally with very poor, if not nonexistent, evidence of having changed any opinions, but relatively strong evidence ( p >  .001) that persisting in making my arguments can get you called ‘an antivaccine loon’, ‘idiot’, someone who engages in ‘Gish Gallop’, or the worst insult I’ve received so far, ‘anti-science’.   While I am really torn on the vaccine issue, I am certain that both peripheries of this debate are at least somewhat wrong in the conclusions that they’ve drawn from the available evidence.  I do believe that lots of parents have witnessed a very real change in their children post vaccination, and I also don’t believe for a single second that vaccines are the cause of an epidemic of autism.  It’s a mess and I’ve been poking around the Internet almost five years into journey autism and from my eyes, it hasn’t improved any in the past half decade.  This is very sad.

That being said, while I do think we need to have a rational and dispassionate discussion about what our existing vaccine studies can and cannot tell us about autism, I’m really concerned about the fact that the vaccine wars seem to have inoculated otherwise intelligent people from any semblance of intellectual curiosity regarding the immunological findings in the autism realm.  That’s a problem, because there are lots of things other than vaccines that can modify the immune response, various environmental agents and cultural changes that are relatively new, and ignoring immunological findings in autism because they happen to intersect with the function of vaccination is a huge, massive, supernova sized disservice to what history will view us poorly on, refusing to perform honest evaluation due to fear and the comfort of willful ignorance.

Here, in this post, I will make the case that this lack of curiosity on immunological findings in autism is either born of a lack of understanding on how much we know about the ties between the immune system and the brain, or alternatively, originates from a deep seated desire to avoid honest interactions.  This isn’t to make the case that vaccines can cause autism, or even that the immunological disturbances observed in autism are causative, but rather that an obstinate refusal to consider these as possibilities is the sign of someone who cannot, or will not accept, the biological plausibility of immunologically driven behaviors despite a constellation of evidence.

One of the things that jumps out to me why the autism population might be a subgroup of the population susceptible to changes as a result of immune dysfunction (and thus, potentially adversely affected as a result of vaccination), is the sheer volume of evidence we now have available to us indicating an altered immune response, and indeed, an ongoing state of inflammation within the brain in the autism population, and most strikingly, repeated observations of a correlation between the degree of immune dysregulation as a propensity of an inflammatory state, and the severity of autism behaviors.  Again and again we’ve seen that as markers indicative of an inflammatory state increase, so too, do severity of autism behaviors.  Not only that, but there are instances wherein the decrease of components known to regulate the immune response decrease, autistic behaviors are more severe.  Subtle shifts in either the start or the resolution of the immune response seems to affect autistic behavior severity in the same way.   I know coincidences happen all the time, but that doesn’t mean that everything is a coincidence. 

We also have a large number of studies that tell us that in vitro, similar levels of stimulation with a variety of agents cause exaggerated or dysregulated production of immune markers in the autism population.   

A large percentage of the time that I mention these findings, usually within discussions with an origin in vaccination, someone decides to educate me on one of the most rudimentary scientific axioms:

Correlation does not equal causation.

It must be stated, the above statement is absolutely true.  Unfortunately for the people for whom this accurate, but simplistic catchphrase comprises the entirety of their argument, it completely ignores a wealth of research that tells us in very unambiguous terms that there is incontrovertible evidence that crosstalk between the immune system and central nervous system can modify behavior. The research indicating a relationship between immune dysregulation and autism does not exist in a vacuum, but rather, is only a tiny fragment of evidence, mostly accumulated within the last few years, that tells us that the paradigm of the past decades, that of the brain as a immune privileged organ without communication to the immune system, is as antiquated as refrigerator moms and a one in ten thousand prevalence. 

From a common sense, why didn’t I think of that standpoint, the best example of the interaction between the brain and the immune response is the old standard, just plain old getting sick.  You live in the dirty world, you pick up a pathogen, you get sick, and suddenly you get lethargic and you start to run a fever.  But is it the pathogen itself that is actually making you feel like staying in bed all day? 

What is being learned is that it is not necessarily the microbial invader that is causing you to get tired and feel sore, but rather, that your decreased energy levels are centrally mediated through your brain, and the triggers for your brain to start a fever include molecules our bodies use for a wide range of communications, including immune based messaging, cytokines.  Some of the most common cytokines in the research to follow include IL-6, IL-1B, and TNF-Alpha; so called ‘pro-inflammatory’ cytokines.  Researchers have been plugging away at just how the immune response is capable of modifying behaviors, i.e., inducing, sickness behavior for a while now, at least in terms of autism research.  From 1998, we have Molecular basis of sickness behavior:

Peripheral and central injections of lipopolysaccharide (LPS), a cytokine inducer, and recombinant proinflammatory cytokines such as interleukin-1 beta (IL-1 beta) induce sickness behavior in the form of reduced food intake and decreased social activities. Mechanisms of the behavioral effects of cytokines have been the subject of much investigation during the last 3 years. At the behavioral level, the profound depressing effects of cytokines on behavior are the expression of a highly organized motivational state. At the molecular level, sickness behavior is mediated by an inducible brain cytokine compartment that is activated by peripheral cytokines via neural afferent pathways. Centrally produced cytokines act on brain cytokine receptors that are similar to those characterized on peripheral immune and nonimmune cells, as demonstrated by pharmacologic experiments using cytokine receptor antagonists, neutralizing antibodies to specific subtypes of cytokine receptors, and gene targeting techniques. Evidence exists that different components of sickness behavior are mediated by different cytokines and that the relative importance of these cytokines is not the same in the peripheral and central cytokine compartments.

The first sentence in this abstract references a practice that is extremely common in studying the immune system, intentionally invoking a robust immune response by exposing either animals, or cells in vitro, to the components that comprise the cell wall of certain types of bacteria; lipopolysaccharide, or LPS.  LPS could be considered a sort of bacterial fingerprint, a pattern that our immune systems, and the immune system of almost everything, has evolved to recognize, and correspondingly initiates an immune response. 

Because this is a conversation that frequently has an origin in vaccination, essentially the act of faking an infection, it is salient to remember that the animals or cell cultures aren’t really getting sick when exposed to LPS; there is no pathology associated with whatever type of bacteria might be housed within a cell membrane containing LPS.  Usually, when the body is exposed to a gram negative bacteria, and the consequent LPS exposure, there are also effects of the bacteria that interact with the organism, but by only incorporating the alert signal for a bacterial invader, we can gain insight into the effect of the immune response itself; there isn’t anything else to cause any changes.   This means that similarly to LPS administration, straight administration of these pro-inflammatory cytokines are similar to the result of getting sick with a pathogen, at least as far as the immune response is concerned. 

In the above instance, administration of LPS, or simply cytokines, had been shown to be capable of causing reduced food intake and ‘decreased social activities’. 

Later in 1998, Central administration of rat IL-6 induces HPA activation and fever but not sickness behavior in rats  (full version), was published wherein the authors report that central administration (i.e., directly into the CNS), of cytokines in isolation (IL-6) or in combination (IL-6 + IL-1B) were capable of inducing altered HPA activation, fevers, and sickness behaviors.  Effects of peripheral administration of recombinant human interleukin-1 beta on feeding behavior of the rat was published a few years later, and observed that peripheral administration (i.e., not the CNS) of IL-1B could affect how much a rat ate, with sucrose ingestion being consistently altered during periods of sickness. 

Jumping ahead a few years, a review paper Expression and regulation of interleukin-1 receptors in the brain. Role in cytokines-induced sickness behavior reviewed how cytokines participate in sickness behavior, Interleukin-6 and leptin mediate lipopolysaccharide-induced fever and sickness behavior examined the interactions of IL-6 and leptin in sickness behavior, and Behavioral and physiological effects of a single injection of rat interferon-alpha on male Sprague-Dawley rats: a long-term evaluation reported “these data suggest that a single IFN-alpha exposure may elicit long-term behavioral disruptions”. 

Much more recently, Sickness-related odor communication signals as determinants of social behavior in rat: a role for inflammatory processes more elegantly found that behavior was modified by LPS exposure, and that this effect was neutralized by concurrent administration of the anti-inflammatory cytokine, IL-10.  Similarly, Inhibition of peripheral TNF can block the malaise associated with CNS inflammatory diseases observed another distinct means by which interfering with the immune response could attenuate the effect of faux sickness, in part, concluding, “Thus behavioral changes induced by CNS lesions may result from peripheral expression of cytokines that can be targeted with drugs which do not need to cross the blood-brain barrier to be efficacious.”  In other words, what is happening in the periphery, outside of the protective boundaries of the blood brain barrier, can none the less manipulate behaviors that are controlled by the brain. 

There are tons, tons more studies like this, but the point should be clear by now; it is accepted that you can achieve some of the same behaviors the come alongside illness, such as fever and lethargy, without the presence of an actual bacteria or virus; all you need is for your brain to think that you are sick. 

While it must be acknowledged that the behavioral disturbances observed in autism are a lot different than feeling the need to watch TV all day, these types of studies were among the first clues that the traditional view of the CNS as a separate entity within the gated community of the blood brain barrier needed revision.   

Measuring how much sugar water a rat drank is great stuff, but the reality is that we have conservatively a gazillion studies telling us that disorders that manifest behaviorally have strong, strong ties to the immune system; and once we begin to understand the vast scope of these findings, the utter frailty of “correlation does not equal causation” becomes painfully clear to the intellectually honest observer. 

The big problem I found myself with in crafting this posting was that the sheer volume of studies available really makes a complete illustration of the literature impossible; I started looking and pubmed nearly puked trying to return to me a listing of all of the things I wanted to summarize.  So here is some of the best of the best; to keep things interesting, I thought I’d only include findings from 2007 or later as a mechanism to show just how nascent our understanding of the connections between the brain and the immune system really are. 

Initially, we can start with a condition that nearly everyone agrees is diagnosed based on behavior, depression.  It turns out, the number of findings establishing a link between immune system markers and depression is wide and deep.   

Here’s a great one, Elevated macrophage migration inhibitory factor (MIF) is associated with depressive symptoms, blunted cortisol reactivity to acute stress, and lowered morning cortisol, which reports, that MIF can modify HPA axis function and is tied to depression; a particularly compelling finding considering well documented alterations in HPA axis metabolites in autism, and the fact that increased MIF has also been found in the autism population, and as levels increased, so too did autism severity. 

Here is part of the abstract for Inflammation and Its Discontents: The Role of Cytokines in the Pathophysiology of Major Depression (full paper)

Patients with major depression have been found to exhibit increased peripheral blood inflammatory biomarkers, including inflammatory cytokines, which have been shown to access the brain and interact with virtually every pathophysiologic domain known to be involved in depression, including neurotransmitter metabolism, neuroendocrine function, and neural plasticity. Indeed, activation of inflammatory pathways within the brain is believed to contribute to a confluence of decreased neurotrophic support and altered glutamate release/reuptake, as well as oxidative stress, leading to excitotoxicity and loss of glial elements, consistent with neuropathologic findings that characterize depressive disorders.

Somewhere along the way, researchers discovered that some anti-depressants can exert anti-inflammatory effects, for examples of these findings we could look to Fluoxetine and citalopram exhibit potent antiinflammatory activity in human and murine models of rheumatoid arthritis and inhibit toll-like receptors, or Plasma cytokine profiles in depressed patients who fail to respond to selective serotonin reuptake inhibitor therapy, which concludes in part, “Suppression of proinflammatory cytokines does not occur in depressed patients who fail to respond to SSRIs and is necessary for clinical recovery”. 

In Investigating the inflammatory phenotype of major depression: focus on cytokines and polyunsaturated fatty acids, the authors report that, “The findings of this study provide further support for the view that major depression is associated with a pro-inflammatory phenotype which at least partially persists when patients become normothymic.”  A nice review of the evidence of immunological participation in depression can be found in The concept of depression as a dysfunction of the immune system (full paper).

Moving forward, we can look to schizophrenia, we have similar findings, including Serum levels of IL-6, IL-10 and TNF-a in patients with bipolar disorder and schizophrenia: differences in pro- and anti-inflammatory balance, which observed an imbalanced baseline cytokine profile in the schizophrenic group; findings very similar in form with An activated set point of T-cell and monocyte inflammatory networks in recent-onset schizophrenia patients involves both pro- and anti-inflammatory forces. Similarly, the findings from Dysregulation of chemo-cytokine production in schizophrenic patients versus healthy controls, (full paper) which states, in part:

Growing evidence suggests that specific cytokines and chemokines play a role in signalling the brain to produce neurochemical, neuroendocrine, neuroimmune and behavioural changes. A relationship between inflammation and schizophrenia was supported by abnormal cytokines production, abnormal concentrations of cytokines and cytokine receptors in the blood and cerebrospinal fluid in schizophrenia

Their findings include differentially increased and decreased production of chemokines and cytokines as a result of LPS stimulations in the case group.  Of particular note, a similarly dysregulated immune profile of cytokine and chemokine generation has been found in the autism population in several studies.  

We also have several trials of immunomodulatory drugs in the schizophrenic arena that further implicate the immune system in pathology, including Adjuvant aspirin therapy reduces symptoms of schizophrenia spectrum disorders: results from a randomized, double-blind, placebo-controlled trial, a ‘gold standard’ trial which found that, “Aspirin given as adjuvant therapy to regular antipsychotic treatment reduces the symptoms of schizophrenia spectrum disorders. The reduction is more pronounced in those with the more altered immune function. Inflammation may constitute a potential new target for antipsychotic drug development”.  A similar clinical trial, Celecoxib as adjunctive therapy in schizophrenia: a double-blind, randomized and placebo-controlled trial , another gold standard trial, which also had findings in the same vein, “Although both protocols significantly decreased the score of the positive, negative and general psychopathological symptoms over the trial period, the combination of risperidone and celecoxib showed a significant superiority over risperidone alone in the treatment of positive symptoms, general psychopathology symptoms as well as PANSS total scores.” [Celecoxib is a cox-2 inhibitor; i.e., anti-inflammatory, i.e., immunomodulatory]

What about bi-polar disorder?  More of the same, including, The activation of monocyte and T cell networks in patients with bipolar disorder, or Elevation of cerebrospinal fluid interleukin-1ß in bipolar disorder, which reports, in part, “Our findings show an altered brain cytokine profile associated with the manifestation of recent manic/hypomanic episodes in patients with bipolar disorder. Although the causality remains to be established, these findings may suggest a pathophysiological role for IL-1ß in bipolar disorder.”. These studies were published in April and March, 2011, respectively. 

Brain tissue from persons with bi-polar disorder also showed increased levels of excitotoxicity and neuroinflammation in Increased excitotoxicity and neuroinflammatory markers in postmortem frontal cortex from bipolar disorder patients (full version), and authors report differential cytokine profiles depending on state of mania, depression, or remission in Comparison of cytokine levels in depressed, manic and euthymic patients with bipolar disorder.

Another disorder based solely around behavior, Tourette syndrome, has increasingly unsurprising findings.  Polymorphisms of interleukin 1 gene IL1RN are associated with Tourette syndrome reports “The odds ratio for developing Tourette syndrome in individuals with the IL1RN( *)1 allele, compared with IL1RN( *)2, was 7.65.” (!!!) , and Elevated expression of MCP-1, IL-2 and PTPR-N in basal ganglia of Tourette syndrome cases is yet another example of observations of CNS based immune participation in a disorder that is diagnosed by behavior.

There are also some reviews that perform a cross talk of sorts between disorders; i.e., The mononuclear phagocyte system and its cytokine inflammatory networks in schizophrenia and bipolar disorder, or Immune system to brain signaling: Neuropsychopharmacological implications, published in May 2011, which has this abstract:

There has been an explosion in our knowledge of the pathways and mechanisms by which the immune system can influence the brain and behavior. In the context of inflammation, pro-inflammatory cytokines can access the central nervous system and interact with a cytokine network in the brain to influence virtually every aspect of brain function relevant to behavior including neurotransmitter metabolism, neuroendocrine function, synaptic plasticity, and neurocircuits that regulate mood, motor activity, motivation, anxiety and alarm. Behavioral consequences of these effects of the immune system on the brain include depression, anxiety, fatigue, psychomotor slowing, anorexia, cognitive dysfunction and sleep impairment; symptoms that overlap with those which characterize neuropsychiatric disorders, especially depression. Pathways that appear to be especially important in immune system effects on the brain include the cytokine signaling molecules, p38 mitogen-activated protein kinase and nuclear factor kappa B; indoleamine 2,3 dioxygenase and its downstream metabolites, kynurenine, quinolinic acid and kynurenic acid; the neurotransmitters, serotonin, dopamine and glutamate; and neurocircuits involving the basal ganglia and anterior cingulate cortex. A series of vulnerability factors including aging and obesity as well as chronic stress also appears to interact with immune to brain signaling to exacerbate immunologic contributions to neuropsychiatric disease. The elucidation of the mechanisms by which the immune system influences behavior yields a host of targets for potential therapeutic development as well as informing strategies for the prevention of neuropsychiatric disease in at risk populations.

All of the conditions above, depression, schizophrenia, bi-polar, and tourettes are diagnosed behaviorally; it is only in the last few years that the medical dimension of these disorders were even understood to exist.   None of the studies that I referenced above are more than five years old; the idea that behavioral disorders were so closely entangled with the immune system is very, very new.  It should be noted that I intentionally left out disorders that also have reams of evidence of immune participation, but which are more degenerative in nature; i.e., Alzheimer’s, ALS, Parkinson’s.  When discussing autism, I also left out studies involving aberrant presence of auto-antibodies, of which there are many.

One of the things that I have learned in trying to refine my thought processes during my time on the Internet is that rarely does a single study tell us much about a condition; but the converse also holds true, if we have many studies with different methodologies or measurement end points, but they all reach similar conclusions, then the likely-hood that the findings are accurate is much, much greater.  All of the studies I have listed above tell us something similar; that the immune system is clearly, unmistakably playing a part in a lot of conditions classically considered neurological and diagnosed behaviorally.  It isn’t enough to nitpick flaws in a single one of the studies in order for ‘correlation does not equal causation’ to make meaningful headway into the implications of these studies; instead, all of the studies above, and lots more, have to be wrong in the same way if we would like to return to a place where we can keep our heads in the sand, hoping for coincidences and bleating out catchphrases in the face of clinical findings.  That isn’t going to happen.  Given this reality, we should not and cannot ignore the growing evidence of immune abnormalities in the autism population, no matter how inconvenient following that trail of evidence might become. 

-pD

There is a lot of over simplification in discussions about autism on the Internet, sometimes I don’t think the people that use them really understand that their points are founded on primitive facsimiles of reality, but other times, I’m pretty sure they do know.  That second group are the ones that really leave me in a confused rage; smart enough to know better (or have had the difference explained to them previously), but continue to rely on utilization of grade school quality parameters to govern complicated and entangled systems.  It seems I’m often wrong when I wonder about the reason people do things (doh!), but when someone otherwise sufficiently knowledgeable relies on the crutch of simplicity because they think it bolsters their argument, I do tend to trust their motives before I consider human fallibility.  It reminds me a lot of politicians, especially Republicans.  [sorry] 

That being said, one of the big simplifications you used to see a lot during the thimerosal wars was this gem:

“The poison in the dose.”

I googled this a bit.  This phrase is attributed to Paracelsus, who Wikipedia tells me is considered ‘the father of toxicology’.  He apparently wrote this:

All things are poison, and nothing is without poison; only the dose permits something not to be poisonous.

Good stuff.  By the way, Paracelsus, who no doubt was pretty smart in his day, was born over six hundred goddamn years ago and the primary observation metrics available to Paracelsus was whether or not something died or not.  Sure, oxygen is deadly in sufficient concentrations, as is water, salt, and everything else, so if we want to have a discussion that allows only for endpoints of livingness or death, the parameters laid out by him are good boundaries.  However, if we would like our conversations to allow for somewhat more subtle changes associated with environmental exposures, something a dispassionate evaluation of the data dictates, we may need to find ways to have conversations that allow for endpoints other than death, and we will need to acknowledge that we have lots of evidence to suggest that there are inputs other than dose that are occasionally meaningful, no matter how this might affect our ability to take comfort in one study or the other.  Even worse, we have actual, real empirical data to suggest there are times when there is an inverse dose relationship.

One of my pubmed alerts somewhat tangential to autism sent me the abstract for Differential mRNA expression of neuroimmunemarkers in the hippocampus of infant mice following toluene exposure during brain developmental period.  It’s a doozy:

Toluene, a volatile organic compound with a wide range of industrial applications, can exert neurotoxic and immunotoxic effects. However, the effects of toluene exposure on developmental immunotoxicity in the brain have not yet been characterized. To investigate the susceptible window to toluene exposure during development and the effects of fetal and neonatal toluene exposure on the neuroimmune markers, gestational day (GD) 14 pregnant mice, postnatal day (PND) 2 and PND 8 male offspring were exposed to filtered air (control; 0 ppm), or 5 or 50 ppm toluene for 6 h per day for five consecutive days. The neuroimmune markers in the hippocampus of PND 21 were examined using a real-time RT-PCR and immunohistochemical analysis. Mice exposed to 50 ppm toluene on PND 2–6 showed significantly increased levels of nerve growth factor (NGF) and tumor necrosis factor (TNF)- mRNAs. In contrast, NGF and brain-derived neurotrophic factor (BDNF) and proinflammatory cytokines TNF-, CCL3, NF-kB, toll-like receptor (TLR)-4, astrocyte marker glial fibrillary acidic protein (GFAP), and microglia marker ionized calcium binding adapter molecule (Iba)-1 mRNAs were increased significantly in mice exposed to 5 ppm toluene on PND 8–12. These results indicate that low-level toluene exposure during the late postnatal period (PND 8–12) might induce neuroinflammatory mediators via TLR4-dependent NF-?B pathway in the hippocampus of PND 21 male mice. Among the three developmental phases, PND 8–12 seems to be most sensitive to toluene exposure. This is the first study to show developmental phase- and dose-specific changes in neuroimmune markers in infant mice following toluene exposure.

Essentially the authors took a bunch of mice exposed them to different amounts of airborne toluene at different days before and after birth, then looked for a variety of changes in immune system markers and neurotrophic factors in the hippocampus.  Toluene was certainly capable of tinkering around with lots of systems that we know are skewed in the autism population.  Curiously, what they found was that there were time dependent changes that had just as much of an impact than dose of toluene; and in fact, much, much lower doses of toluene were capable of causing more robust changes if the exposure occurred during critical developmental windows. 

The authors state that the timeframe of exposures in this study, postnatal days 2 -6 and postnatal days 8 – 12 roughly map to the early and late third trimester of human fetal development, respectively.  I’ve seen similar equivalencies in other papers, some with earlier and later timeframes, but certainly these timeframes are generally within the range that other papers have used.  Consistent with the theme of this post, I’d just say that rat to human is difficult, and rat to human specific brain area and developmental timeframe equivalency is even more difficult.

The authors speculate that the difference in effect may be related to what was happening, developmentally within the brain at the time of toluene exposure that made the impact.  

During this period, hippocampus undergoes an increase in size and a change in excitatory neurotransmission to allow for adult-like synaptic plasticity by the end of the second postnatal week (Dumas, 2005). This transiently heightened level of brain plasticity is shaped byenvironmental factors which have profound effects on this brain growth spurt (Goodlett et al., 1989). Furthermore, during this period, the immune system undergoes maturation to immunocompetence (Dietert et al., 2000).

There are also some stuff about why the hippocampus is a particularly promising target for investigation into effects of toluene exposure. 

Here are a couple of graphs of their findings:

Check that shit out!  During some very specific developmental timeframes, a decreased exposure resulted in increased physiological effect, not only that, the more affected animals received ten times less agent.   Less poison, more effect.  The exact opposite of what Paracelsus predicts.  [Sorry for the formatting/stupid wordpress!]

Saliently towards autism, these graphs just happen to show some measurements that have great functional overlap with findings from autism.  These graphs are for CCL3, an immune bugler of sorts, a chemokine, an agent responsible for attracting components of the immune response, one numeral down for CCL2, aka MCP-1, which we’ve also seen increased in the in autism brains, iba1, a marker for microglial activation, NGF and BDNF, neurotrophic factors that have a variety of signaling and maintenance processes in the CNS, and we have much data implicating altered BDNF levels in autism. 

Not only did the authors observe an inverted dose relationship, some of the measurements found that the time dependencies are also reversed from what you might expect in that later exposure was worse than earlier exposure.  Environmental exposures do not necessarily follow the linear timelines you might expect.

The idea of an inverted, or skewed dose relationship has actually been explored for some time.  For example, The frequency of U-shaped dose responses in the toxicological literature   

Hormesis has been defined as a dose-response relationship in which there is a stimulatory response at low doses, but an inhibitory response at high doses, resulting in a U- or inverted U-shaped dose response. To assess the proportion of studies satisfying criteria for evidence of hormesis, a database was created from published toxicological literature using rigorous a priori entry and evaluative criteria. One percent (195 out of 20,285) of the published articles contained 668 dose-response relationships that met the entry criteria. Subsequent application of evaluative criteria revealed that 245 (37% of 668) dose-response relationships from 86 articles (0.4% of 20,285) satisfied requirements for evidence of hormesis. Quantitative evaluation of false-positive and false-negative responses indicated that the data were not very susceptible to such influences. A complementary analysis of all dose responses assessed by hypothesis testing or distributional analyses, where the units of comparison were treatment doses below the NOAEL, revealed that of 1089 doses below the NOAEL, 213 (19.5%) satisfied statistical significance or distributional data evaluative criteria for hormesis, 869 (80%) did not differ from the control, and 7 (0.6%) displayed evidence of false-positive values. The 32.5-fold (19.5% vs 0.6%) greater occurrence of hormetic responses than a response of similar magnitude in the opposite (negative) direction strongly supports the nonrandom nature of hormetic responses. This study, which provides the first documentation of a data-derived frequency of hormetic responses in the toxicologically oriented literature, indicates that when the study design satisfies a priori criteria (i.e., a well-defined NOAEL, > or = 2 doses below the NOAEL, and the end point measured has the capacity to display either stimulatory or inhibitory responses), hormesis is frequently encountered and is broadly represented according to agent, model, and end point. These findings have broad-based implications for study design, risk assessment methods, and the establishment of optimal drug doses and suggest important evolutionarily adaptive strategies for dose-response relationships.

We have other examples from the synthetic world that may be of interest to autism.  For example, in Developmental Exposure to Polychlorinated Biphenyls Interferes with Experience-Dependent Dendritic Plasticity and Ryanodine Receptor Expression in Weanling Rats the authors report an inverted dose relationship regarding exposure to PCBs and dendrite growth.

Developmental A1254 exposure significantly enhanced dendritic growth in cerebellar Purkinje cells and neocortical pyramidal neurons among P31 rats not trained in the Morris water maze, which is consistent with our previous observations that similar exposures accelerated dendritic growth in Purkinje cells and hippocampal CA1 pyramidal neurons between P21 and P60 (Lein et al. 2007). In Purkinje cells, this effect was observed among animals in the 1 mg but not 6 mg/kg/day A1254 group, whereas in neocortical neurons, responses were comparable between A1254 groups. The reason for the different dose–response relationship in different brain regions is not known. Possibilities include regional differences in RyR regulation (Berridge 2006; De Crescenzo et al. 2006; Hertle and Yeckel 2007) or differential upregulation of cytochrome P450 enzymes by AhR ligands in the cerebellum versus neo-cortex (Iba et al. 2003), which could result in regional differences in PCB toxicodynamics and toxicokinetics, respectively.

What about situations where we have evidence for an environmental factors already associated with autism?  Neuroinflammation and behavioral abnormalities after neonatal terbutaline treatment in rats: implications for autism found that terbutaline administration at postnatal day 2 -5 resulted in chronically activated microglia and behavioral abnormalities in rodents, but the same dose in days 11 – 14 resulted in no such effect.  Same dose, different time, different outcome.

There is more, lots more, but how many times must a rule fail primitive logical tests before we acknowledge that it’s utility in complex discussions is extremely limited?  This absolutely is not meant as an expose meant to reignite discussions about thimerosal, but rather, to illustrate the dangers of trying to understand complicated rules by leveraging simplistic heuristics.  There’s a lot of that in the autism discussion landscape; it is a dangerous concoction of hubris and faith to think that we can have gain meaningful insight into our shared mystery by applying very simple rules. 

I haven’t seen the ‘poison is in the dose’ canard used for a while now.  Good riddance and long live models that are not exceedingly simple to invalidate.

-          pD

Whatever your take on the predominant cause of autism(s), your thoughts on the appropriateness, or inappropriateness of research allocations in the autism realm, one thing can’t be denied; collectively, a lot of researcher time and dollars have been poured into autism research.  We’ve learned some important things, some cool things, some confusing things, some obscure things, and some useless things.   But even with all of the resources we have applied towards understanding autism, one of the most curious unknowns is also one obvious to the most rudimentary observations, the persistently skewed male to female ratio, with a finding of three or four males to every female with autism.   It is one of the most vexing questions, almost taunting us with seeming obviousness, but consistently elusive.  It isn’t just autism, lots of other neurological conditions are similarly tilted, and in a bazillion animal models it seems an unfortunate fact that being born male simple predisposes you (or rat you) to a variety of things you’d rather not have. 

The ideas I have seen floated most frequently to explain this observation involve the effect of prenatal testosterone and the associated ‘extreme male brain’ theory, a loss of genetic backups to compensate for mixups, and synergistic effects of testosterone on chemicals, notably, mercury; explanations which I generally like a little, a little less, and almost not at all, respectively.  An idea I’ve floated a couple of times, but that seemed even less accepted (or, more likely, completely unnoticed), was that estrogen might be acting in a protective manner; as it is known to exhibit attenuate the effects of neuroinflammation and oxidative stress.  Indeed, it is starting to look like estrogen receptors are expressed in a large variety of situations salient to CNS processes.  

A few weeks ago, the same group that has published some immensely cool studies on epigenetics and brain proteins, genetic expression differentials within twin siblings with autism, and circadian rhythm alterations comes a paper which may give us insight into this question, so that is pretty exciting.  Even cooler, it invokes a negative feedback loop in a complicated system and is built upon the foundation of several earlier studies on a protein implicated in lots of things we know are awry in autism.  That study is Sex Hormones in Autism: Androgens and Estrogens Differentially and Reciprocally Regulate RORA, a Novel Candidate Gene for Autism (full paper).  Here is the abstract:

Autism, a pervasive neurodevelopmental disorder manifested by deficits in social behavior and interpersonal communication, and by stereotyped, repetitive behaviors, is inexplicably biased towards males by a ratio of ~4:1, with no clear understanding of whether or how the sex hormones may play a role in autism susceptibility. Here, we show that male and female hormones differentially regulate the expression of a novel autism candidate gene, retinoic acid-related orphan receptor-alpha (RORA) in a neuronal cell line, SH-SY5Y. In addition, we demonstrate that RORA transcriptionally regulates aromatase, an enzyme that converts testosterone to estrogen. We further show that aromatase protein is significantly reduced in the frontal cortex of autistic subjects relative to sex- and age-matched controls, and is strongly correlated with RORA protein levels in the brain. These results indicate that RORA has the potential to be under both negative and positive feedback regulation by male and female hormones, respectively, through one of its transcriptional targets, aromatase, and further suggest a mechanism for introducing sex bias in autism.

The press release and google news cycle for this paper seemed to have been well ahead of the pubmed robot; Kev had a post on this study a few weeks before it hit pubmed with a postdate.  I generally skip out on the interest story/vaccine fairytale story/vaccine nightmare story/lost child nightmare story that is the google news autism feed, but in this case, it harbored a story on a paper that I was actually interested in.  In any situation, the paper landed in pubmed this morning, and is available in full via PLOS, so great stuff is available to us all.

The paper starts with some of the backstory, the ‘inexplicable’ male predominance in autism, some of the theories on why this might be the case, and most importantly, details on previous findings by this set of researchers on reduced levels of RORA in the CNS of people with autism, a protein with a great number of functions of interest to the autism community.

Together, these results link molecular changes in RORA in peripheral cells to molecular pathology in the brain of autistic individuals. These findings are particularly relevant to ASD as RORA is involved in several key processes negatively impacted in autism, including Purkinje cell differentiation, cerebellar development, protection of neurons against oxidative stress, suppression of inflammation, and regulation of circadian rhythm. Behavioral studies on the RORA-deficient staggerer (RORA+/sg) mouse, primarily used as a model to study ataxia and dystonia[13], further show that RORA is also associated with restricted behaviors reminiscent of ASD, such as perseverative tendencies, limited maze patrolling, anomalous object exploration as well as deficits in spatial learning.

It’s tough to find a protein with a greater key word hitlist for our population of interest than Purkinje cell differentiation, cerebellar development, protection from oxidative stress, suppression of inflammation, and regulation of the circadian rhythm.  In fact, I’d be shocked to find a protein touching so many fracture points that wasn’t found altered in the autism population; it makes too much sense within the framework of an entangled system and what we already know about the physiology of autism.   Remember that the previous paper found decreased RORA in the brain of people with autism; i.e., less of a protein that protects from oxidative stress, supports Purkinje cell development, and suppresses the inflammatory response.  A relative lack of RORA makes a depressingly good amount of sense. 

That being said, what makes the current paper so interesting is that they found the RORA is differentially, and inversely modulated by female and male hormones (i.e., testosterone and estrogen).  But even more insidiously, one of the downstream products regulated by RORA, aromatase, participates in the cleavage of testosterone to estrogen; the authors essentially describe a negative feedback loop.  It turns out, not only is RORA decreased in the CNS of autism, but so too is aromatase. 

We also show that one of the transcriptional targets of RORA is aromatase, which is a crucial enzyme in the biosynthesis of estrogen from testosterone. It is noteworthy that both RORA and aromatase proteins are decreased in the frontal cortex of autistic subjects, and that the level of aromatase protein is strongly correlated with the level of RORA protein in the brain tissues. We therefore propose that the reduction of RORA observed in autism is exacerbated by a negative feedback mechanism involving decreased aromatase level, which further causes accumulation of its substrate, testosterone, and reduction of its product, estradiol. Testosterone and estradiol respectively exhibit negative and positive feedback regulation of RORA expression as illustrated in Fig. 5, which summarizes the principal findings of this study. Thus, a deficiency in RORA in autistic brain is expected to be further aggravated by increased levels of testosterone due to suppression of aromatase, a transcriptional target of RORA.

This is pretty neat; it shows how simply being male can lead to the downregulation of a system with tendrils attached to a great number downstream processes we know to be disturbed in autism. 

I particularly liked that this paper established a chain of learning more, something I think we can all agree is a great idea.  Some of the people on this study have been plugging away with some interesting ideas for a while, all of which, I believe, are ancestors of these findings.  They had two really neat papers on genetic expression in autism twins with differential degrees of autism severity, both of which used genomic bioinformatic tools to understand which the genetic pathways were affected.  This is actually rather brilliant; they essentially leveraged the genetic uniqueness of the twins to gain more insight into which processes were being affected in autism by seeing which genes were differentially expressed in identical twins that manifested differently, using genetics to learn about what is happening a layer above the genome.  Next, the original RORA paper began to probe the mechanism by which the previously observed expression was achieved, they found that a particular protein, RORA, was overmethylated and consequently at depressed levels.  Another bioinformatic approach told them that RORA was a particularly attractive candidate for further evaluation based on its descendant interactions, and the association between RORA, aromatase, and sexual hormones appeared.  Beautiful.

All that coolness not withstanding, some of the articles I saw on this lacked the caution and nuance we ought to see with these kinds of findings; the paper was pretty clear that previous CNS studies hadn’t shown decreased RORA in all of their samples, just most of them.   This doesn’t answer all of our questions about the male dominance of autism, but we do know more than when this study was published, and that is pretty cool.   Hooray for knowledge.

-    pD

I have decidedly mixed feelings on the genetic side of autism research; clearly genetics plays a part, but it does appear that autism has largely mirrored other complicated conditions in that what we thought we were getting when we cracked the genetic code has, for all practical purposes, failed to materialize.  To what extent our genetic makeup really plays a part in autism more than any other condition that is currently mystifying us, I don’t think we can say with much certainty; unless you want to count some.

To my mind, one particularly bright spot in the gene realm is the associations of the MET-C allele and an increased risk of an autism diagnosis.  At first glance, MET doesn’t seem like a big deal; lots of people have the MET-C mutation, in fact, nearly half of everyone has it.   But people with autism have it just a little more frequently, an observation that has been replicated many times.  But what is exciting is not only that the MET-C findings are robust, but they can also affect a lot of implicated systems in autism in biologically relevant ways.  From an ideological standpoint, the fissure in the autism community about research priorities regarding genetics versus environment, the MET-C studies are a superb example of just how much useful knowledge there is by starting at the genome and working upwards, and finding once we get there that the reality involves lots more than just genes.  There is something for everyone!

Getting to the big picture where we can appreciate the beautiful complexity takes a little bit of digging, but it’s worth the effort. 

Every now and again you’ll see a period piece about the forties, fifties or sixties, and you’ll get a glimpse of the female operator, someone who would take a call and literally connect two parties together; the gatekeeper. The operator’s actions were binary; either she connected the lines and the call went through, or she didn’t, and nothing happened.  Of course, one operator couldn’t connect you to any other phone, but participated in groupings of phones with some logical or functional structure.  Ultimately, the operators were the enabler of communication, physically putting two entities into contact to perform whatever business they had with each other. 

Within our bodies, tyrosine kinases  are enzymes responsible transferring phosphate to proteins; a chemical exchange critical towards a great number of cellular functions, and in a sense, the tyrosine kinases act as cellular operators, helping implement a physical swap of chemicals that ultimately set in motion a great number of processes.  Some very rudimentary cellular functions are initiated by the tyrosine kinases; for example, cell division, which is why mutated kinases can lead to the generation of tumors; i.e., the signaling for cell division gets turned on, and never gets turned off.  Inhibiting tyrosine kinases is the mechanism of action for some drugs that target cancer.  

The MET gene is responsible for creating the MET receptor tyrosine kinase.  This particular receptor is involved in lots of processes that are of great interest to autism; the MET receptor is expressed heavily during embryogenesis in the brain, has immune modulating capacities, and is associated with wound healing, and is particularly implicated in repair of the gastro-intestinal track. 

Kinases don’t just fire away, shuttling phosphates around any old time, they must be activated by a triggering molecule, or a ligand.  There is only one known ligand for the MET receptor; hepatocyte growth factor, or HGF (also sometimes referred to as HGF/SF, or hepatocyte growth factor/scatter factor).  We’ll get to why we bother worrying about HGF a little later on, but it is important to keep in mind that without HGF, the functions affected by the MET-C receptor, early brain development, immune modulating, and wound repair cannot be achieved. 

So what about autism, and why is it a beautiful illustration of complexity?  Walking our way through the MET findings in autism is a rewarding task; it is one of the few instances I’ve seen where the glimpses of relevance gleaned from straight genetic studies have been incrementally built upon to achieve a much grander understanding of autism.  This is the kind of thing that I think a lot of people who dismiss the utility of genetic studies are missing; genetics are only the first piece of the puzzle, it doesn’t only implicate genes, it tells us about the processes and the proteins disturbed in autism; and with that knowledge, we can perform targeted analysis for environmental participants.

The first clues about MET involvement with autism came in 2006, when A genetic variant that disrupts MET transcription is associated with autism (full paper) was published.  The abstract is longish, but here is a snipet:

MET signaling participates in neocortical and cerebellar growth and maturation, immune function, and gastrointestinal repair, consistent with reported medical complications in some children with autism. Here, we show genetic association (P = 0.0005) of a common C allele in the promoter region of the MET gene in 204 autism families. The allelic association at this MET variant was confirmed in a replication sample of 539 autism families (P = 0.001) and in the combined sample (P = 0.000005). Multiplex families, in which more than one child has autism, exhibited the strongest allelic association (P = 0.000007).

I appreciate the pleiotropic nature of what we are seeing here, a gene that is involved with brain growth and maturation, immune function, and GI repair.  The association in ‘multiplex’ (i.e., families with more than one child with autism) was very, very strong.  Even still, this was a pretty short paper, and it was all genetics.  Coolness factor:  3.

Neater studies were on the horizon shortly thereafter, a year later, some of the same group looked for expression of MET in post mortem brain tissue and found significantly decreased levels of MET protein in Disruption of cerebral cortex MET signaling in autism spectrum disorder. 

MET protein levels were significantly decreased in ASD cases compared with control subjects. This was accompanied in ASD brains by increased messenger RNA expression for proteins involved in regulating MET signaling activity. Analyses of coexpression of MET and HGF demonstrated a positive correlation in control subjects that was disrupted in ASD cases.

This is a nice follow up; lots of times a genetic study might suggest a hit, but we really don’t even know how such a genetic change might manifest physiologically, like having a jigsaw puzzle of solid black and finding two pieces that fit together.  In those instances, we can’t really go looking for different levels of the protein, so there you are.  In this case, the authors found an allele worth investigating, and then went looking to see if relevant proteins were altered in the population, and in the CNS no less!  Not only that, but they also looked at the initiating end of the process, the ligand, HGF, and found abnormalities.  Good stuff.  Unfortunately, I haven’t found myself a copy of this paper yet, but the fact that other proteins in the pathway were altered is another line of evidence that something is amiss.  I’ve begun to appreciate the fact that I have spent a long time under appreciating the interconnectedness of biological systems; you aren’t going to have a disturbance in one system without altering the way upstream, and downstream processes are working; so  the fact that we see other proteins, those related to MET functions, modified, makes beautiful sense.  Coolness factor: 5.

Likely because of the mixed findings of skewed proteins in the MET pathway (?), the next study in line is, Genetic Evidence Implicating Multiple Genes in the MET Receptor Tyrosine Kinase Pathway in Autism Spectrum Disorder (full paper available).  Here’s the abstract:

A functional promoter variant of the gene encoding the MET receptor tyrosine kinase alters SP1 and SUB1 transcription factor binding, and is associated with autism spectrum disorder (ASD). Recent analyses of postmortem cerebral cortex from ASD patients revealed altered expression of MET protein and three transcripts encoding proteins that regulate MET signaling, hepatocyte growth factor (HGF), urokinase plasminogen activator receptor (PLAUR) and plasminogen activator inhibitor-1 (SERPINE1). To address potential risk conferred by multiple genes in the MET signaling pathway, we screened all exons and 5′ promoter regions for variants in the five genes encoding proteins that regulate MET expression and activity. Identified variants were genotyped in 664 families (2,712 individuals including 1,228 with ASD) and 312 unrelated controls. Replicating our initial findings, family-based association test (FBAT) analyses demonstrated that the MET promoter variant rs1858830 C allele was associated with ASD in 101 new families (P=0.033). Two other genes in the MET signaling pathway also may confer risk. A haplotype of the SERPINE1 gene exhibited significant association. In addition, the PLAUR promoter variant rs344781 T allele was associated with ASD by both FBAT (P=0.006) and case-control analyses (P=0.007). The PLAUR promoter rs344781 relative risk was 1.93 (95% Confidence Interval [CI]: 1.12−3.31) for genotype TT and 2.42 (95% CI: 1.38−4.25) for genotype CT compared to genotype CC. Gene-gene interaction analyses suggested a significant interaction between MET and PLAUR. These data further support our hypothesis that genetic susceptibility impacting multiple components of the MET signaling pathway contributes to ASD risk.

We’ve got two new genes added to the mix, PLAUR and SERPINE.  The juicy part here is that the authors didn’t look for these variants at random, but performed a targeted search; they knew that the proteins encoded by these genes interact with either MET receptor function or HGF, and they also had found altered expression of these genes in the CNS study.  From the Introduction:

The hepatocyte growth factor (HGF) gene encodes the activating ligand for the MET receptor. HGF is translated as an inactive precursor protein that requires cleavage for efficient binding to the MET receptor [Lokker et al 1992]. The activating cleavage of HGF is achieved most efficiently by the enzyme plasminogen activator (urokinase-type; uPA; gene symbol: PLAU) under conditions in which uPA binds to its receptor, the urokinase plasminogen activator receptor (uPAR; gene symbol: PLAUR). Activating cleavage of HGF can be suppressed by the plasminogen activator inhibitor-1 (PAI-1; gene symbol: SERPINE1). Together, these proteins regulate the activity of MET receptor tyrosine kinase signaling, and our recent microarray analyses of postmortem temporal lobe of individuals with ASD indicate that disrupted MET signaling may be common to ASD pathophysiology [Campbell et al 2007]. For example, we found that there is increased expression of the HGF, PLAUR and SERPINE1 transcripts in ASD in postmortem cerebral cortex. The observation of disrupted expression suggests a general dysfunction of MET signaling in the cerebral cortex of individuals with ASD.

The proteins encoded by PLAUR and SERPINE were also found increased in the expression study; a finding further supported by the genetic study here.  The really grand slice here is that the SERPINE protein suppresses cleavage of HGF; essentially another way MET function can be affected, from a disturbance upstream of HGF binding.   In other words, more SERPINE (possibly as a result of a ‘promoter allele’) would result in less MET receptor activation because the SERPINE interferes with the cleavage of HGF, and thus, another pathway to reduced MET activation.  In a finding that seems 20/20 with hindsight, a functional promoter of the SERPINE gene was found to increase autism risk; i.e., if you have more SERPINE, you get less functional HGF, and therefore less triggering of the MET receptor.  This is cool and begins a portrait of the complexity; it shows how the effect of reduced MET functionality can come from multiple drivers; the reduced MET allele, or, the promoter SERPINE allele, and what’s more, having both is an even bigger risk; the authors are describing a synergy of low penetrance genes.

From the discussion section of the paper:

Beyond genetic susceptibility, the functional integrity of the MET signaling system also is sensitive to environmental factors. This concept is supported by bioinformatics analyses that identified PLAUR, SERPINE1 and HGF as genes active in immune response regulation, sensitive to environmental exposures, and within chromosomal regions previously implicated in ASD linkage studies [Herbert et al 2006]. Moreover, a recent cell biological study shows that chemically diverse toxicants reduce the expression of MET in oligodendrocyte progenitor cells, a result that is interpreted as the convergence of toxicant effects on oxidative status and the MET-regulating Fyn/c-Cbl pathway

Here are links to the Hebert paper, Autism and environmental genomics, and the Li paper, Chemically Diverse Toxicants Converge on Fyn and c-Cbl to Disrupt Precursor Cell Function.  What is neat here is that we are starting to be able to see a pathway of genes, and resultant proteins, that can effect disparate systems.   I believe that there is a subset of acupuncture, acupressure that relies on more knuckles than needles, and while the science on accu* based therapies isn’t very good, it does occur to me that in a sense, our lattice work of HGF-PLAUR-SERPINE proteins that participate in the MET-C process are pressure points in a delicate system, push a little bit and things will bend down the line accordingly.  It also exemplifies why I am offended by highly negative attitudes on genetic studies held by people who believe in a non trivial, environmentally mediated increase in the rates of autism; we are approaching a nearly impossible to overturn reality that genes we know to be associated with autism are particularly sensitive to interference from environmental agents, and participate in immune function.  That is important information.  Coolness factor 8.  First glimpse of beauty factor: 10.

Next up we have Dynamic gene and protein expression patterns of the autism-associated Met receptor tyrosine kinase in the developing mouse forebrain (full paper). 

The establishment of appropriate neural circuitry depends upon the coordination of multiple developmental events across space and time. These events include proliferation, migration, differentiation, and survival – all of which can be mediated by hepatocyte growth factor (HGF) signaling through the Met receptor tyrosine kinase. We previously found a functional promoter variant of the MET gene to be associated with autism spectrum disorder, suggesting that forebrain circuits governing social and emotional function may be especially vulnerable to developmental disruptions in HGF/Met signaling. However, little is known about the spatiotemporal distribution of Met expression in the forebrain during the development of such circuits. To advance our understanding of the neurodevelopmental influences of Met activation, we employed complementary Western blotting, in situ hybridization and immunohistochemistry to comprehensively map Met transcript and protein expression throughout perinatal and postnatal development of the mouse forebrain. Our studies reveal complex and dynamic spatiotemporal patterns of expression during this period. Spatially, Met transcript is localized primarily to specific populations of projection neurons within the neocortex and in structures of the limbic system, including the amygdala, hippocampus and septum. Met protein appears to be principally located in axon tracts. Temporally, peak expression of transcript and protein occurs during the second postnatal week. This period is characterized by extensive neurite outgrowth and synaptogenesis, supporting a role for the receptor in these processes. Collectively, these data suggest that Met signaling may be necessary for the appropriate wiring of forebrain circuits with particular relevance to social and emotional dimensions of behavior.

Coooooool.   Here we touch on the complexity of brain formation, all the little things that need to go exactly right, and how MET might play a role in that incredibly complicated dance.  Even better, a mouse model is used to gain an understanding of where and when peak expression of MET proteins occur, a period of significant changes to neural structures and the formation of synapses, the physical structures that enable thought.   This is a dense paper, too dense to get deeply into blockquoting for this posting, but there are some parts that deserve notice, namely, documentation of spatially localized MET expression in brain areas associated with social behaviors and some fine grained information on the specific parts of synapse formation that utilize MET.    Coolness factor: 8.  Complexity Factor: 12.

Here is a paper that a lot of people that play skeptics on the Internet ought to hate, Distinct genetic risk based on association of MET in families with co-occurring autism and gastrointestinal conditions.  (full paper)

In the entire 214-family sample, the MET rs1858830 C allele was associated with both autism spectrum disorder and gastrointestinal conditions. Stratification by the presence of gastrointestinal conditions revealed that the MET C allele was associated with both autism spectrum disorder and gastrointestinal conditions in 118 families containing at least 1 child with co-occurring autism spectrum disorder and gastrointestinal conditions. In contrast, there was no association of the MET polymorphism with autism spectrum disorder in the 96 families lacking a child with co-occurring autism spectrum disorder and gastrointestinal conditions. chi(2) analyses of MET rs1858830 genotypes indicated over-representation of the C allele in individuals with co-occurring autism spectrum disorder and gastrointestinal conditions compared with non-autism spectrum disorder siblings, parents, and unrelated controls.

There is a lot of caution in this paper, but the nice part is that there are biologically plausible mechanisms by which a reduction in MET could snowball into problems in the gastro-intestinal track.

In the gastrointestinal system, MET signaling modulates intestinal epithelial cell proliferation, and thus acts as a critical factor in intestinal wound healing. For example, activation of MET signaling via application of exogenous hepatocyte growth factor has been shown to reduce the effects of experimentally induced colitis, inflammatory bowel disease, and diarrhea.

Pushing on the other end of the balloon, increasing MET signaling, has been shown to help GI problems; no less than evidence that a genetic change associated with autism has biologically plausible mechanisms by which GI problems would be more prevalent. In fact, unless our findings of MET alleles are in error, or our clinical findings of the effects of HGF are spurious, it is absolutely expected. There is also a section with the startlingly simple, and simultaneously great idea of why findings like these might be useful markers for phenotypic categorization in studies in the future; i.e., to discern the prevalence of GI problems in autism, it might, for example, make sense to design that study to take presence or absence of MET alleles into consideration.  Nice.  Coolness Factor: 7.  Insidiousness factor: 9.

Here’s another one that found associations with MET and social behavior, and GI disturbances again.  Association of MET with social and communication phenotypes in individuals with autism spectrum disorder

Autism is a complex neurodevelopmental disorder diagnosed by impairments in social interaction, communication, and behavioral flexibility. Autism is highly heritable, but it is not known whether a genetic risk factor contributes to all three core domains of the disorder or autism results from the confluence of multiple genetic risk factors for each domain. We and others reported previously association of variants in the gene encoding the MET receptor tyrosine kinase in five independent samples. We further described enriched association of the MET promoter variant rs1858830 C allele in families with co-occurring autism and gastrointestinal conditions. To test the contribution of this functional MET promoter variant to the domains of autism, we analyzed its association with quantitative scores derived from three instruments used to diagnose and describe autism phenotypes: the Autism Diagnostic Interview-Revised (ADI-R), the Autism Diagnostic Observation Schedule (ADOS), and both the parent and the teacher report forms of the Social Responsiveness Scale (SRS). In 748 individuals from 367 families, the transmission of the MET C allele from parent to child was consistently associated with both social and communication phenotypes of autism. Stratification by gastrointestinal conditions revealed a similar pattern of association with both social and communication phenotypes in 242 individuals with autism from 118 families with co-occurring gastrointestinal conditions, but a lack of association with any domain in 181 individuals from 96 families with ASD and no co-occurring gastrointestinal condition. These data indicate that the MET C allele influences at least two of the three domains of the autism triad.

Really sort of plain, but very nice to see the GI component validated in another data set.  Coolness factor 5.

Then a few months ago, Prenatal polycyclic aromatic hydrocarbon exposure leads to behavioral deficits and downregulation of receptor tyrosine kinase, MET was released, an uber cool showcase of the autism bigfoot, the often regaled, only very rarely documented, gene/environment interaction. 

Gene by environment interactions (G × E) are thought to underlie neurodevelopmental disorder, etiology, neurodegenerative disorders, including the multiple forms of autism spectrum disorder. However, there is limited biological information, indicating an interaction between specific genes and environmental components. The present study focuses on a major component of airborne pollutants, polycyclic aromatic hydrocarbons (PAHs), such as benzo(a)pyrene [B(a)P], which negatively impacts cognitive development in children who have been exposed in utero. In our study, prenatal exposure of Cpr(lox/lox) timed-pregnant dams to B(a)P (0, 150, 300, and 600 μg/kg body weight via oral gavage) on embryonic day (E14-E17) consistent with our susceptibility-exposure paradigm was combined with the analysis of a replicated autism risk gene, the receptor tyrosine kinase, Met. The results demonstrate a dose-dependent increase in B(a)P metabolite generation in B(a)P-exposed Cpr(lox/lox) offspring. Additionally, a sustained persistence of hydroxy metabolites during the onset of synapse formation was noted, corresponding to the peak of Met expression. Prenatal B(a)P exposure also downregulated Met RNA and protein levels and dysregulated normal temporal patterns of expression during synaptogenesis. Consistent with these data, transcriptional cell-based assays demonstrated that B(a)P exposure directly reduces human MET promoter activity. Furthermore, a functional readout of in utero B(a)P exposure showed a robust reduction in novel object discrimination in B(a)P-exposed Cpr(lox/lox) offspring. These results confirm the notion that common pollutants, such as the PAH B(a)P, can have a direct negative impact on the regulated developmental expression of an autism risk gene with associated negative behavioral learning and memory outcomes.

Oh snap.  A common pollutant (well, common in the last few decades anyways), is shown to interact with MET in a dose dependent fashion to reduce protein expression in the brain during embryonic development and cause ‘a robust reduction in novel object discrimination’. (Ouch)  This is an example of just what we mentioned above, referenced Herbert, concerning the possibility of MET as a gene sensitive to ‘environmental exposures’.  Indeed.  From the discussions section:

The results from the present study demonstrate that the transcription and developmental expression patterns of a replicated ASD risk gene, MET, are highly sensitive to a common PAH pollutant. In utero exposure to B(a)P produces an oxidative milieu of B(a)P metabolites in offspring during a key postnatal period of synapse development, providing evidence that environmental exposure creates a sustained cerebral cortical burden that likely contributes to an increased oxidative load. Oxidative stressors in the form of metabolites would be expected to negatively impact gene expression (Kerzee and Ramos 2000) and, more specifically, receptor tyrosine kinase function, including Met (Li et al. 2007). These data suggest that B(a)P-induced exposure would impact the expression of key neurodevelopmental genes, including Met. Additionally, the predominance of the 3-OH and 9-OH metabolites places a sustained burden in the brain because of the potential for further oxidization to form B(a)P quinones (McCallister et al. 2008, Hood et al. 2000, Brown et al. 2007) which undergo redox cycling to generate reactive oxygen species (Kerzee and Ramos 2000, Bolton et al., 2000).

And

In conclusion, specific developmental events such as glutamatergic excitatory synapse formation and maturation may be particularly vulnerable to G x E effects that impact regulatory and signaling proteins involved in this process. While we do not suggest that the current study reflects specific defects related to a complex clinical condition such as the ASDs, current molecular, behavioral and functional imaging data are converging on the concept that the ASDs are a manifestation of altered local and long-distance cortical connectivity (Geschwind et al. 2007, Bill and Geschwind 2009, Geschwind and Levitt, 2007, Levitt and Campbell 2009). Also, Met and other related signaling components of this receptor tyrosine kinase pathway have been implicated in both syndromic and idiopathic disorders where the ASDs are diagnosed at a high rate. In combination with risk alleles in key genes, the in utero exposure to PAHs such as B(a)P, which results in both a reduction in absolute levels and the mistiming of peak Met expression, could drive the system toward a pathophysiological threshold that neither genetic risk nor environmental factors could produce individually. The present study focused on the neocortex, but given the highly restricted spatial and temporal expression of Met in mouse limbic circuits associated with social-emotional development and cognition (Judson et al. 2009), it is likely that perturbations occur throughout these key circuits, including in the hippocampus.

Really cool stuff; particularly the finding that developmental, in utero exposure was capable of driving abnormal protein expression well after birth. This is the best of both sides of the genetics versus environment conundrum; the kind of finding that sheds light on how environmental pollutants could be participating in increasing the number of children with autism by interacting with genetically susceptible children.  But what I love about this is that it is the death knell of the fairytale of a static rate, or near static rate of autism, just having the genes or the exposure isn’t enough; instead, the interaction of alleles and timed exposure ‘could drive the system toward a pathophysiolical threshold that neither genetic risk nor environmental factors could produce individually’.  I think there are some more findings coming from this group soon that might be exciting, or terrifying, depending on how you see it.  (or both).  Coolness factor: 99.   

So what have we learned and just how cool is it?

1)      The MET receptor enables some types of cellular signaling that have relevance to the autism community including synapse formation, immune modulation, and gastro intestinal function.  The ligand, or trigger of the MET receptor is HGF.

2)      Certain alleles of the MET gene that result in decreased expression are more common in children with autism than people without autism.

3)      Consistent with findings of increased prevalence of MET alleles, MET protein expression was found to be decreased in brain tissue from people with autism.  Other, related proteins, HGF, PLAUR, and SERPINE were also found to be disturbed.

4)      Following up on the differential findings of SERPINE and PLAUR, genetic studies found gene to gene interactions between the MET allele and alleles involved with production of SERPINE and PLAUR. Some of the proteins in question are known to be particularly vulnerable to environmental interference.

5)      Animal models tell us that MET is heavily expressed in many areas of the mammalian brain during prenatal and postnatal development, and we gain insight into the spatial and temporal expression of MET during the intricate dance of brain formation.

6)      Two studies add evidence that the one function of decreased MET expression, GI disturbances, are indeed found with greater consistency within children with autism and the MET allele.  This should be a relatively unsurprising finding considering what we know about MET and children with autism.

7)      Finally, a portrait of genetic / environmental interactions capable of disturbing physiology and behavior in ways consistent with findings in autism is rendered using an agent that is the product of the automobile age and already associated with decreased cognitive skills for groups with the highest gestational exposure.

It should be noted that this is just a slice of the MET papers out there in the autism realm; they all shared one or more authors, I picked them because they seem to show a nice progression of knowledge, and incremental approach towards learning more.   There is a lof more to learn, in particular, I think that the immune modulating effects of reduced expression would be an interesting subject, but one that will have to wait for another posting. 

-  pD

Hot on the heels of Mitochondrial Dysfunction in Autism, another study on mitochondrial function in the autism population was just released, this time giving us insight into what is happening inside the gated community behind the blood brain barrier.  How potentially inconvenient.  Brain region-specific deficit in mitochondrial electron transport chain complexes in children with autism came out the other day; I’ve yet to receive a full copy (one has been promised to my real world email), but the abstract is juicy enough to warrant a small posting.

Mitochondria play important roles in generation of free radicals, ATP formation, and in apoptosis. We studied the levels of mitochondrial electron transport chain (ETC) complexes, i.e., complexes I, II, III, IV, and V, in brain tissue samples from the cerebellum and the frontal, parietal, occipital, and temporal cortices of subjects with autism and age-matched control subjects. The subjects were divided into two groups according to their ages: Group A (children, ages 4-10 years) and Group B (adults, ages 14-39 years). In Group A, we observed significantly lower levels of complexes III and V in the cerebellum (p < 0.05), of complex I in the frontal cortex (p < 0.05), and of complexes II (p < 0.01), III (p < 0.01), and V (p < 0.05) in the temporal cortex of children with autism as compared to age-matched control subjects, while none of the five ETC complexes was affected in the parietal and occipital cortices in subjects with autism. In the cerebellum and temporal cortex, no overlap was observed in the levels of these ETC complexes between subjects with autism and control subjects. In the frontal cortex of Group A, a lower level of ETC complexes was observed in a subset of autism cases, i.e., 60% (3/5) for complexes I, II, and V, and 40% (2/5) for complexes III and IV. A striking observation was that the levels of ETC complexes were similar in adult subjects with autism and control subjects (Group B). A significant increase in the levels of lipid hydroperoxides, an oxidative stress marker, was also observed in the cerebellum and temporal cortex in the children with autism. These results suggest that the expression of ETC complexes is decreased in the cerebellum and the frontal and temporal regions of the brain in children with autism, which may lead to abnormal energy metabolism and oxidative stress. The deficits observed in the levels of ETC complexes in children with autism may readjust to normal levels by adulthood. (my emphasis)

A few things immediately jump out at me.  Firstly, the Chauhan’s are authors of this paper, who have been around the autism / oxidative stress block since the get go, as authors of the very nice Oxidative stress in autism: increased lipid peroxidation and reduced serum levels of ceruloplasmin and transferrin–the antioxidant proteins, a really nice paper that was one of the first I saw that broke the autism groups into classic and regressive phenotypes with findings of increased oxidative stress in the latter.

Secondly, one of the biggest concerns with Mitochondrial Dysfunction in Autism when it was released a few weeks ago was, whether or not the findings taken from lymphocytes, cells outside of the brain, could be reliably used as proxies for what is happening within the CNS.  Based on the findings in Brain region-specific deficit in mitochondrial electron transport chain complexes in children with autism it would seem that, at least in children, there is an increased frequency of mitochondrial problems in the brain.  Of course, if we acknowledge the reality of the interconnectedness of immune activation, oxidative stress, mitochondrial impairment and what we already know about the CNS in autism, these findings shouldn’t really be all that surprising.  None the less, it is nice to have some direct evidence of this.

Unfortunately, we still don’t know what is causing the problems with mitochondria function in the brain; it is possible, though exceedingly unlikely that all of the participants in this study also had a diagnosable electron chain disorder (I haven’t gotten a full copy of the paper yet).  I think it is possible that there is a feedback loop in place involving the immune response, oxidative stress, and mitochondria that for some reason our children’s physiology cannot shake loose from. 

The very small sample size of the children in this study, five, is an unfortunate reality for nearly all brain based studies in the autism world.  Though I’ve yet to read the full paper, my prediction is that it is liberally peppered with cautious language regarding interpreting the findings widely without further confirmation.  That is probably pretty good thinking.

But, if we look closely, and we taken notice of the where of mitochondrial problems in the autism group was observed, we may have evidence of participatory processes.  Specifically, Chauhan found decreased electron chain transport measurements in the cerebellum, frontal cortex, and temporal cortex.

In Group A, we observed significantly lower levels of complexes III and V in the cerebellum (p < 0.05), of complex I in the frontal cortex (p < 0.05), and of complexes II (p < 0.01), III (p < 0.01), and V (p < 0.05) in the temporal cortex of children with autism as compared to age-matched control subjects, while none of the five ETC complexes was affected in the parietal and occipital cortices in subjects with autism.

(my emphasis)

There have been a few other studies (that I know of) that have looked for brain region specific abnormalities that might be of interest to u.  Brain Region-Specific Changes in Oxidative Stress and Neurotrophin Levels in Autism Spectrum Disorders (ASD), which found increased markers of oxidative stress in the cerebellum:

Consistent with our earlier report, we found an increase in NT-3 levels in the cerebellar hemisphere in both autistic cases. We also detected an increase in NT-3 level in the dorsolateral prefrontal cortex (BA46) in the older autistic case and in the Wernicke’s area and cingulate gyrus in the younger case. These preliminary results reveal, for the first time, brain region-specific changes in oxidative stress marker 3-NT and neurotrophin-3 levels in ASD.

(My emphasis)

Interesting note: the ‘Wernicke’s area’ of the brain plays a large part in language skills, and in fact, damage to the Wernicke’s area can cause a type of aphasia. 

The number of studies that tie together oxidative stress and mitochondrial function are many and numerous to the point of cumbersomeness, I have a short list of them on a previous post about mitochondria function in autism, here. 

Two of the really nice neuroimmune studies in the autism realm, Neuroglial Activation and Neuroinflammation in the Brain of Patients with Autism, and Immune Transcriptome Alterations In the Temporal Cortex of Subjects With Autism both provide evidence of an ongoing immune response in some of the specific areas of the CNS where Chauhan found impaired mitochondrial function, the cerebellum and the temporal cortex.

From Vargas:

We demonstrate an active neuroinflammatory process in the cerebral cortex, white matter, and notably

in cerebellum of autistic patients.

And

The neuroglial activation in the autism brain tissues was particularly striking in the cerebellum, and the changes were associated with upregulation of selective cytokines in this and other regions of the brain.

From Garbett:

Expression profiling of the superior temporal gyrus of six autistic subjects and matched controls revealed increased transcript levels of many immune system related genes. We also noticed changes in transcripts related to cell communication, differentiation, cell cycle regulation and chaperone systems.

Detangling if these findings are related, and if so, the direction of causality is for another series of studies to discern.  Calls towards the possibility that relationships like this are spurious are common, but I hate to invoke coincidences for no good reason other than coincidences do occur.  My suspicion is that the immune findings and impaired mitochondria findings are related, but a cautious suspicion is all that is warranted at this time.  I do believe that the relationship between immune activation and mitochondria function is being evaluated now; though I do not know if it is being addressed directly in the CNS, which would be ideal.

Curiously from my perspective, however, is the finding that young adults and adults with ASD in Chauhan did not exhibit decreased electron chain function.  The original microglia paper from Vargas, Neuroglial Activation and Neuroinflammation in the Brain of Patients with Autism found extensive evidence of an ongoing immune response in the CNS of people with autism into adulthood.  From the standpoint of a theory wherein an immune response were driving the mitochondrial impairment due to increased oxidative stress, the findings in Chauhan of normal mitochondria function are contradictory to what was found in Vargas.  (?) 

A few other thoughts occurred to me as I considered the age differences found in Chauhan.  If mitochondrial dysfunction is part of the pathogenic force driving behaviors associated with autism, it is possible that a decrease as adulthood is reached conforms with a general improvement in adaptation many people seem to report.  Alternatively, if we are actually observing a true increase in the number of people with behaviors that can be classified as autistic, that is, the number of children with autism is a new phenomena, the age findings in Chauhan could be artifacts of different underlying causes of autism in the adults versus the children.  I’m a big believer in a wide range of physiological roads to the end point of autistic behaviors, so such a situation doesn’t really bother me conceptually, though it is very, very problematic to put to any kind of designed experiment. 

Lastly, for a while now I’ve been putting some thought towards something that’s really been bugging me about the neuroimmune findings in autism when put in context with other ‘classic’ neurological diseases that also exhibit a strong immune component; i.e., Alzheimer’s or Parkinson’s, both of which have strong immune findings as well, but are more strikingly degenerative in nature when compared to autism.  Generally you talk about a child with autism gradually getting better, or in some cases reaching a plateau; but very rarely (or never) is there the steady and unforgiving decrease in function that you see in diseases like Alzheimer’s.  I’m struggling with this reality and how our findings fit in.  I’m not sure how, or if, the age differences in Chauhan are meaningful towards this apparent paradox, but my pattern recognition unit sure is trying to tell me something, I just can’t tell if it’s sending me on (another) snipe hunt or not.

When the entire paper lands in my inbox, I may write another post about it.  I’m interested in seeing if any other blogs pick up on this paper or not and what their take on it is.  I’m still sort of in the dark on the machinations of the press cycle as it relates to autism news, but this paper doesn’t seem to have gotten the press release treatment that Mitochondrial Dysfunction in Autism did, even though its findings are just as interesting. 

-          pD