Wednesday, 18 March 2015

The Role of Microglia in the Puzzle of Neuro-inflammation in Autism

Regular readers of this and similar blogs will have noticed that the human body functions in quite irrational ways.  We know why this is; we are the product of a very slow evolutionary process, rather than being a clean-sheet design like your smart phone or iPad.

As a result, nothing is ever quite as simple as it seems and at times the cleverer you are, the less likely you are to find a medical therapy effective in humans.

Such is the case with autism, inflammation and microglia.

It might seem that you can track back inflammation in autism to its “root cause”, which could appear to be those immune cells in the brain, called microglia.  We know they are “activated” in autism and we know that autism is typified by an “over-activated” immune response.

Working with the assumption that autism is a brain dysfunction, you would assume that the effective therapy should be inside the so-called blood brain barrier (BBB).

You would then just look for a potent drug that could “stabilize” the microglia/immune cells in the brain, to calm things down.  Having achieved this, you would sit back and marvel at the behavioral change and improvement in cognitive function.

This was exactly the thought process a few years ago when the US  National Institute of Mental Health (NIMH) got together with the Johns Hopkins researchers to follow up on their findings of chronic inflammation in the brains of people with autism.  Subsequent, third party, research has also confirmed that the microglial cells are “activated” in autism

Trial Description

There is a subgroup of children with autism that appear to develop typically for a period of time, and then lose skills, or regress. A recent study by Vargas and co-workers at Johns Hopkins has demonstrated that the regressive subtype of autism is associated with chronic brain neuroinflammation as exemplified by activation of microglia and astroglia and the abnormal production of inflammatory cytokines and growth factors assayed in both tissue samples (brain banks) and CS. The authors remarked that these responses were similar to those seen in some neurodegenerative disorders such as amyotrophic lateral sclerosis, and that chronic microglia activation appears to be responsible for a sustained neuroinflammatory response that facilitates the production of multiple neurotoxic mediators. Chronic neuroglial activation could be the result of an abnormal persistence of a fetal development pattern. In this scenario neuroglial activation could play a role in initiating and in maintaining the pathology. Alternatively, neuroglial activation may only be a secondary response to the initiating causal factor(s) and not a direct effector of injury. Since neuroglial activation requires the nuclear translocation of the pro-inflammatory transcription factor NF-kappa B, and since inhibitors of NF-kappa-B with good CNS penetrance are available, the role of neuroinflammation in initiating and sustaining the autistic condition can be probed.
The antibiotic minocycline is a powerful inhibitor of microglial activation, apparently through blockade of NF-kappa-B nuclear translocation. Minocycline is neuroprotective in mouse models of amyotrophic lateral sclerosis (ALS) and Huntington's disease and has been recently shown to stabilize the course of Huntington's disease in humans over a 2-year period.
To evaluate the possibility of benefit in autistic children, we propose to conduct an open-label trial of the anti-inflammatory antibiotic minocycline, an agent that reduces inflammation by blocking the nuclear translocation of the proinflammatory transcription factor NF-kappa-B. Minocycline is Food and Drug Administration (FDA)-approved for treatment of a variety of infections and has been widely used for the treatment of adolescent acne. Minocycline is currently in phase III trials for the treatment of Huntington's disease and amyotrophic lateral sclerosis.
This proposal is for an initial 6-month, single-arm, off label, open-label study (with a 3 month extension phase offered to responders) that will evaluate dose safety and efficacy of minocycline in 10 children, ages 3 to 12 years, with a primary diagnosis of autism and a history of developmental regression. The subjects will be evaluated by a diagnostic/behavioral assessment, and the extent of neuroinflammation judged by CSF cytokine/chemokine profiles before and after the 6-month treatment. Subjects will also be given 0.6 mg/kg vitamin B6 twice a day as a prophylactic for possible minocycline induced nausea and vomiting. If the results of this feasibility study are encouraging, we expect to conduct a double-blind, placebo-controlled trial of minocycline therapy.

Nothing happens fast in the world of autism and so this six month study of 10 people (who completed the actual trial) was conceived in 2006, was actually concluded in 2013.  Here is the resulting paper:-

Changes in the pre- and post-treatment profiles of BDNF in CSF and blood, HGF in CSF and CXCL8 (IL-8) in serum, suggest that minocycline may have effects in the CNS by modulating the production of neurotrophic growth factors. However, in this small group of children, no clinical improvements were observed during or after the six months of minocycline administration.

Unfortunately, this study showed that a treatment, known to effectively stabilize microglial cells, had no positive effect on autism and actually seemed in some cases to make it worse.

We can conclude from this that stabilizing the microglia will not be the “holy grail” for treating autism.  Rather, the activated microglia is just one part of a complex, and only partially understood process.

Microglia as the Immunostat 

In a recent post we saw how Rodney Johnson referred to the microglia as the “immunostat” of the body.  Like the thermostat on the wall in your home central heating system.

This is indeed an interesting analogy and might explain some of what is going on.

We saw in Johnson’s paper all the ways that the immune system outside the blood brain barrier (BBB) was able to communicate with the microglia.  We should assume that this communication works both ways; something that is usually overlooked.

In a perfectly functioning body, as in a perfectly functioning house, the immunostat/thermostat gives a good indication of the actual state/temperature, as well as the one you intended.  So if you set your room thermostat to 72 Fahrenheit / 22 Celsius  you expect the actual temperature to be 72 Fahrenheit / 22 Celsius.

However, in the real world things do not work like this.

We live in a house with very large south facing windows, a big fireplace, underfloor heating in some places and European-style hot water radiators (in the US they do have them).  So we have at least four sources of heat.  In spite of having clever German electronics to control our heating system, the thermostat in the centre of the house, by itself, is not adequate.

Something similar is happening in body and brain of people with autism, just replace temperature with inflammation.

Just as my house has multiple systems resulting in heating, the human body has numerous processes leading to “inflammation”.  Some of these inflammatory processes are interconnected and some are not.  The net result at any one time can be measured by looking at various cytokine levels, gene expression, microglial activation and numerous other things; there is no single measurable thing called “inflammation”.

There will never be a single wonder anti-inflammatory treatment.

The activated state of the microglia rather than being the ultimate target for intervention may just be a reflection of inflammation elsewhere in the body, or alternatively it may be just the result of oxidative stress in the brain.

Just like after a few years you may need to replace your wall thermostat, because it is giving false data, the clever immunostat, that may be the microglia, could have been disrupted by all that oxidative stress in the brain.  It might even be sending its proinflammatory signal in reverse, back across the BBB, to the rest of the body. Not such a crazy idea?

The future of anti-inflammatory interventions

The NIMH and Johns Hopkins would naturally be disappointed by the results of their study; but it was a study well worth doing.  Hopefully they will pursue other avenues of thought.

We already know that there are numerous ways to achieve a degree of immuno-modulatory change and that in some types of autism there can be a profound behavioral impact.

These range from simple Ibuprofen, to steroids like Prednisone; not to mention those Kv1.3 blockers and ShK-peptides.  These will likely all affect the microglia, but it is not their main mode of action.


As is often the case, there are useful insights that you can learn from a “failed” trial.

I would imagine that an autistic person with ulcerative colitis would also have activated microglia. Treating that person with minocycline should have some stabilizing influence on the microglia, but without resolving the ulcerative colitis, the pro-inflammatory signals continue to be sent around the body.

Turning down the thermostat in my house, when I have a big log fire blazing, has no effect on the temperature. 

The microglia in the brain of people with autism probably should not be activated; we really need to know why they are activated.

If you can work on the numerous processes/pathways leading to “inflammation” you would most likely also achieve some deactivation of the microglia.

Therefore we should look at things like PPAR gamma which are directly relevant to the pathology of autism, and agonists of PPAR gamma also happen to be “anti-inflammatory” and indeed, in the test tube, some can stabilize microglia.

One, far away, day they will bring those ShK-peptides to the market. 

In the meantime, my current targets are Tangeretin and Nobiletin, flavonoids found in tangerines.

For the scientists among you:-

In addition to being a PPAR gamma agonist, Tangeretin is also a known P2Y2 receptor antagonist.  Both properties are potentially useful.

PPAR gamma has been covered in this blog already.  P2 receptors are a class of Purinergic receptor.  Within the field of purinergic signalling, these receptors have been implicated in learning and memory, locomotor and feeding behavior, and sleep. 

Suramin is used in research as a broad-spectrum antagonist of P2 receptors.

It is Suramin that Robert Naviaux, at UC San Diego, has been researching as a potent autism therapy.  He has shown it effective in mouse models, but the problem is that it is not safe for long term use in humans.  Regular readers should note that, yet again, an anti-parasite drug has been found to have an effect in autism.  Parasites do not cause autism, but understanding them better would be a potential advantage.

Why Suramin, a Century-Old, Anti-Parasitic Drug May Hold the Key to Understanding Autism

Dr. Robert Naviaux's recent finding suggests reversible metabolic syndrome could be at core of autism

The full paper is below:-

In particular, P2Y11 is a regulator of immune response.  There are big gaps in the science and I have no idea if tangeretin affects P2Y11.


  1. flavanoids (ginerol and tangeritin) appear to be possibly more helpful than might be imagined. We will shortly be trailing the ginerol, low and slow

    1. I think Gingerol is a good one to try. Let us know what effects, if any, that you see.

  2. Peter,
    Thank you for this great blog post. I am very interested in minocycline (and other tetracycline) antibiotics for treatment of autism. It seems like there is a lot happening in this area. This post on sfari has references to minocycline having some therapeutic value in Angelman’s syndrome and Fragile X:
    In the trial you highlighted the researchers also provided substantial doses of B6 as you mentioned above which may have significant effects some of which are negative in some individuals.
    Personal experience with tetracyclines:
    1) My son with ASD has been on vibramycin (tetracycline) based on elevations in a titer for a tick borne disease. I think he has had significant improvement on it.
    2) I have an undiagnosed disease one symptom of which is pains in my legs and arms which I believe has a vascular component and tetracycline seems to reduce my pains.
    Your blog post, my experiences, and this other research suggest that there are some with ASD who are responders to tetracycline antibiotics. As you have noted one theory on this is that tetracyclines downregulate NF-kappa-B. I wonder if some of the therapeutic benefit has to do with its antimicrobial effects. For example in my son’s case it may be that the therapeutic affect is from vibramycin decreasing the load from a tick borne bacteria. In addition my son may also be benefiting from the antibacterial effect on some other unknown bacteria in the blood, or perhaps from decreases in dysbiotic bacteria.

  3. Did you see that Naviaux had siccess also with suramin in a genetic model (fragile x), not just the mia model? his would be worthy of a follow-up post. Also, I found this thread interesting Many thanks for this amazing blog. I'm aspergers, and are trying the Bumetanide approach (no effect) and clonazepam (small effect, possibly calming only). /Joel

    1. You may get some of the Suramin benefit from Tangeretin, which you can buy as Sytrinol. I am testing it, it is also a PPARgamma inhibitor, which Suramin is not. It does seem to help.

      If you try the GABAa drugs (Bumetanide/clonazepam) you might also try the GABAb drugs (Baclofen / Pantogam / Phenibut) these last three are supposed to reduce anxiety in Asperger's.

    2. Thank you, ill give sytrinol a test! Ive tried arbaclofen already, but couldnt stand the side effects (You wouldnt believe the amount of things Ive tried the past year)

      Have you seen the initiative


Post a comment