Monday 29 September 2014

Mounting Evidence Regarding Autism, Neurofibromatosis and PAK1

When I google “autism” and “PAK1”, I keep seeing my own posts come up.  This is beginning to be a regular occurrence, when I research an idea.  Google “verapamil autism”, “clonazepam autism” “bumetanide autism” and even “NAC autism”, the same thing happens.

So it is nice to have some further studies that also show the possible importance of PAK1 in treating autism.  This time it is from the University of Indiana and more precisely, Anantha Shekhar, Professor of Psychiatry at the School of Medicine.

We have the study’s abstract and the more people-friendly press release.


Children with neurofibromatosis type 1 (NF1) are increasingly recognized as having a high prevalence of social difficulties and autism spectrum disorders (ASDs). We demonstrated a selective social learning deficit in mice with deletion of a single Nf1 allele (Nf1+/−), along with greater activation of the mitogen-activated protein kinase pathway in neurons from the amygdala and frontal cortex, structures that are relevant to social behaviors. The Nf1+/− mice showed aberrant amygdala glutamate and GABA neurotransmission, deficits in long-term potentiation and specific disruptions in the expression of two proteins that are associated with glutamate and GABA neurotransmission: a disintegrin and metalloprotease domain 22 (Adam22) and heat shock protein 70 (Hsp70), respectively. All of these amygdala disruptions were normalized by the additional deletion of the p21 protein-activated kinase (Pak1) gene. We also rescued the social behavior deficits in Nf1+/− mice with pharmacological blockade of Pak1 directly in the amygdala. These findings provide insights and therapeutic targets for patients with NF1 and ASDs.

Here is the very informative and readable press release.


INDIANAPOLIS -- Blocking a single gene that is active in the brain could provide a means to lessen behavioral problems among children with a common genetic disease, many of whom are also diagnosed with an autism disorder, according to researchers at the Indiana University School of Medicine.
The genetic disorder, neurofibromatosis type 1, is one of the most common single-gene diseases, affecting about 1 in 3,000 children worldwide. Symptoms can range from café-au-lait spots on the skin to tumors that are disfiguring or that can press dangerously against internal organs.
"Physicians are increasingly recognizing that many children with the disorder have social and behavioral difficulties, and as many as one in five cases of autism may be associated with the same biochemical defects seen in neurofibromatosis type 1," said Anantha Shekhar, M.D., Ph.D., Raymond E. Houk Professor of Psychiatry at the IU School of Medicine.
The researchers used a mouse model of neurofibromatosis, examining both behavioral differences from normal mice and biochemical differences in the animals' brains, particularly in the amygdala, a brain structure associated with social behavior and emotional regulation.
Reporting their work in the journal Nature Neuroscience, the researchers found that the neurofibromatosis model mice had problems with long term social learning -- remembering important social cues involving interactions with other mice. Tests also showed that neurochemical pathways between structures of the brain involved with social behavior were disrupted by the neurofibromatosis mutation.
However, blocking the activity of another gene -- called Pak1, which is involved with those neurochemical pathways -- improved the social behaviors of the mice. Mice bred to have both the neurofibromatosis mutation and the deletion of the Pak1 gene engaged in social behavior similar to normal mice. In addition, mice with the neurofibromatosis mutation that were injected with a compound known to block Pak1 gene activity had normal social behavior restored.
"These findings could lead to novel approaches to treating behavioral problems that are seen in NF1 patients and some patients with autism spectrum disorders," said D. Wade Clapp, M.D., Richard L. Schreiner Professor of Pediatrics at the IU School of Medicine.


The researchers from Indiana are suggesting that 20% of people with autism may have the same dysfunction as the very much rarer condition of neurofibromatosis type 1.  Those 20% are likely to benefit from treatments shown to be effective in NF-1.

How do you know whether you are in the 20%?  A little genetic testing might tell you, or maybe not (see below).

In the absence of such testing, you could possibly deduce something from looking at the comorbidities.

It might seem odd that NF-1, a rare disorder affecting 1 in 3,000 children could share its underpinnings with 20% of children with autism, which would roughly equate to 6 in 3,000 children.

This reminds me of a question I raised earlier:-

In that post it became clear that you can have a partial dysfunction of a “rare” genetic disorder.  I wonder if that partial dysfunction will show up on today’s genetic tests.


The comorbidities of autism that most intrigue me are asthma, allergies and ulcerative colitis.  I have a suspicion that they are all linked by mast cell degranulation and further, that what is underlying autism is promoting mast cells to degranulate.

A recent study showed how PAK1 is involved in modulating mast cell degranulation:-


And another one:- 


Fortunately, the effects of PAK1-deficiency on the immune system have a very encouraging up-side. As demonstrated by otherwise relatively healthy PAK1-/- mice, Pak1 is critical for disassembly of cortical F-actin upon allergen stimulation, and PAK1 deficiency prevents the release of pro-inflammatory molecules from the granules of mast cells during the IgE-associated allergic responses

I have already shown the effectiveness of Verapamil as a therapy for autism and mast cell degranulation.  I suspect that a further improvement may follow with a potent PAK1 inhibitor.

I think the Indiana research also points in the same direction.

There is also the issue of malformed dendritic spines, which will be fully addressed in a later post.  This appears in autism and schizophrenia and may explain much of why autistic brains function differently to other peoples.  It is thought that this malformation is also linked to PAK1.

So while treating mast cell degranulation will help some people’s autism, you could also go one step backwards up the chain and address the signal that was prompting them to degranulate.  This same signal may trigger an unrelated damaging cascade of events elsewhere in the brain.

Which PAK1 inhibitor?

In earlier post we saw that the choices of PAK1 inhibitor are:-

1.     Experimental drugs still under development by Afraxis, the MIT spin-off  

2.     Ivermectin, an old anti-parasite drug, used with some success by fringe alternative doctors in the US.  At least one reader of this blog is a fan of Ivermectin for autism.
3.     Certain types of Propolis, like the one containing CAPE (Caffeic Acid Phenethyl Ester) that comes from New Zealand
The question remains whether the Propolis is potent enough to have the same effect as Ivermectin.  In the NF-1 and NF-2 community, opinion is split as to whether Propolis can shrink existing tumours.  This issue of stopping new tumours developing, versus shrinking existing ones does seem to crop up quite often in cancer research as well.  Drugs are, not surprisingly, most effective when used very early on.

Ivermectin cannot be used long term continuously, since it is toxic.  It can be used “on and off” for decades as an anti-parasite therapy.

Crossing the Blood Brain Barrier

Once question arose in an earlier post as to how Ivermectin could be effective in autism, since it does not readily cross the blood brain barrier.  According to the experts it does not have to, see below:-


11. Expert opinion: Is PAK1 a suitable target for therapy?

As discussed above, there is growing evidence that PAKs are involved in the phenomena that are clinically significant for various cardio-vascular disorders, but the specificity of PAK1 involvement is still uncertain. Studies indicate that even closely related PAKs (e.g. PAK1 and PAK2) have non-identical sets of substrates. The issue is further complicated because of the multiple and sometimes opposing roles of PAKs in these processes and certainly merits further investigation.

The reports on the involvement of PAK1 in various diseases of the brain indicate that both up- and down-regulation of this enzyme may be associated with pathological changes. This, along with the uncertainty about the relative contribution of other isoforms, clouds the prospect of targeting PAK1 for therapeutic intervention in these conditions. Furthermore, these observations necessitate a close attention to the affects that any anti-PAK therapy targeted at other organs might have on the nervous system, including the cognitive functions and the memory. In this regard, failure of an anti-PAK1 agent to penetrate the blood-brain barrier may not be a detriment to its therapeutic utility. Similarly complicated is the question of PAK1 targeting in infections: while it may partially attenuate certain viruses, it would also negatively impact some functions of the immune system. In fact, the recent report of PAK1-deficient animals having IgE-mediated responses to allergens may indicate that, at least, for such acute life-threatening conditions as anaphylaxis the benefits of suppressing PAK1 may outweigh the risks.

My PAK-1 inhibitor Trial

I am practicing what I preach, so to speak.  Only once the pollen allergy season is well and truly over, will I trial my PAK-1 inhibitor.  I want a genuine result, free from external effects, like degranulating mast cells.

Since Ivermectin is known to react with other drugs in my PolyPill, I will be using the Propolis from New Zealand. 

Friday 26 September 2014

Autism Drugs - Horses for Courses and Safety over Assured Efficacy?

Only a few months goes by without there being an uplifting report in the media of some breakthrough drug for autism.  These reports usually relate to research on mice.

So where are the resulting approved drugs for use on humans?

There still are no drugs approved for the core symptoms of autism.  It is quite likely that in spite of all the ongoing research, the situation will not change anytime soon.

I was reading about yet another potential wonder treatment, based on research into a very old drug called Suramin.   This rather toxic drug has been shown to be effective in a particular mouse model of autism call MIA (Maternal Immune Activation).  There is some doubt as to whether the researchers have got the method of action correct, but nobody doubts the positive effect it had on some mice.

Today’s post does not look at the science of Suramin, which is, by the way, another anti-parasite drug like Ivermectin, which I looked at earlier.  The subject of this post is much more down to earth and practical.

There is a problem with all Autism Clinical Trials

It is not just me that thinks something is amiss with Autism Clinical Trials, first read what the head of Medical Research at Autism Speaks has to say.  He is talking about this in the context of Naviaux’s recent trials of Suramin on “autistic” mice:-

Paul Wang, Head of Medical Research, Autism Speaks :-

Hedging bets: “Animal models of autism, such as the maternal immune activation (MIA) model studied here by Naviaux and his colleagues, are the best tools that researchers have for examining the cellular and molecular pathophysiology of autism and for testing experimental treatments before they can be advanced to human trials.

“But, of course, none of the models can be considered valid until treatment effects in them are proven to be predictive of effects in people. In the case of the MIA mouse, the authors here candidly hedge their bets by calling it a model of both autism and schizophrenia. Meanwhile, the field of autism research wisely hedges its own bets by studying multiple treatments of the MIA mouse, including probiotics as well as antipurinergic therapy.”

Precedent lacking: “Although milestones in the initial stage of testing basic research findings for translational research continue to accumulate — from mGluR5-targeted rescue of the FMR1 knockout mouse to suramin reversal of social deficits in the MIA mouse — we appear to be making little headway on the hurdles of clinical trials. From arbaclofen to oxytocin to Trichuris suis ova, clinical trial results have been tepid at best. This should not be surprising. We have no successful precedent to guide the design of clinical trials in autism.
“How should we quantitate clinical improvement — or deterioration? How long must treatment be provided before effects are evident? At what age will each treatment be most effective: 6 years? 16 years? 6 months? Which individuals will benefit most from each treatment: those with more severe or more mild symptoms? Those with regression or not? Those with or without comorbidities? Results in Phelan-McDermid syndrome (presented by Joseph Buxbaum at the 2014 International Meeting for Autism Research) represent a rare but preliminary exception to the frustrations of clinical trials.”

Clearing the hurdles: “As basic research continues to generate more candidate treatments for autism, we need to work harder on clinical trials. Most especially, we need to identify measures of improvement that emerge early, potentially within a few weeks of treatment initiation and well before the broad functional improvement that the U.S. Food and Drug Administration is likely to require for drug approval.”

Multiple mouse models, suggests multiple human types of autism

The fact that researchers have created multiple types of mutant mice that mimic autistic behaviour does rather suggest that numerous distinct dysfunctions in humans might also result in autistic behaviour. 

In fact it is now a widely held belief, in the scientific community, that there are numerous sub-types of autism, each with its own biological dysfunction(s).

Clinical trials doomed to fail?

Since no effort is made to stratify the autistic population by sub-type, clinical trials are likely doomed to fail.  They usually just require that participants fall into the vague autism behavioral category of DSMIV, or now DSM V.

While a trial drug may indeed have a positive effect in one sub-type of autism, it may have no effect, or worse still a negative effect in other subtypes.  This is exactly what happed with Arbaclofen, and Roche pulled the plug on that one.

Horses for Courses

Perhaps a more pragmatic approach is required.  “Horses for courses”, was suggested to me the other day by that prolific autism science blogger from Sunderland.

Just accept that one Alzheimer’s drug may work for Fragile X, but be totally in-effective in broader autism.  Or maybe it only works in some people with Fragile-X?

This sound fine, but what if you do not know which “course” your horse (child) is running on?

Science may indeed have the answer in the form of something called micro RNA analysis, which is a way of looking for a large number of known genetic dysfunctions quickly and therefore relatively cheaply.  It just needs a blood sample. It is available to autism researchers today.

In the meantime we are left with that reliable old workhorse called trial and error, which does seem to work, if you do your homework.

Safety over Assured Efficacy

While clinical trials may not be able to guarantee which drugs are helpful in autism, they can tell us which are safe to use.  Fortunately many of the interesting drugs for autism are existing ones that have been in use for decades, but for other conditions.

One interesting point I noticed in the autism trials of Alzheimer’s drugs was that the drugs were very well tolerated.  Not surprisingly, older patients claim to have far more frequent side effects, since they likely have multiple ailments and may attribute their various ills to the new drug.

So what is required to treat autism is a range of drugs that are known to be safe for long term use; and then some indication of effectiveness in some people with autism.

Last year, when reading the very detailed critique of most recent clinical trials into autism, produced  by the UK’s National Institute for Health and Care Excellence (NICE), it was clear that they are looking for a level of success in clinical trials that will likely never materialize.  This was a 700 page document produced in advance of the final 40 page report.  Only the 40 page report seems to be available now.

A “one size fits all” approach will fail, because “autism” is a vague behavioral diagnosis and not a precise biological one.

Any particular drug might be effective in only 10% of what psychiatrists rather arbitrarily define as “autism”, but if your child is in that 10%, you would be delighted.

The logical way forward is blocked for most people, since they cannot access even very safe prescription drugs.  This is of course for the “greater good” of society and avoids doctors worrying about getting prosecuted for malpractice.

Monday 22 September 2014

Back to School and “Learning Years”

School for Monty, aged 11 with ASD, did start a couple of weeks ago but then a nasty virus swept through school, sending him back home again.

To recap, Monty attends a very small mainstream international school with his own assistant. The school uses the English system. To get the equivalent US grade, you subtract one from the English year.  He comes home after lunch and then has one-to-one, ABA-inspired, home schooling for another three hours.    In school holidays he has eight hours a day of ABA-inspired one-to-one home program.  This has been going on for seven years so far.

Following all these years of ABA, schooling at home and 20 months of his PolyPill he is now able to learn at school, follow the rules and interact with staff and other children.  He now initiates play with the other kids.

When his assistant leaves at 2pm, the teachers now want him to stay by himself for afternoon classes like art and physical education.  This is quite a change, until quite recently the teachers did not want him there if his assistant was unable to be at school, or got delayed in traffic.

The clever move turned out to be holding him back two years, a while back; so that he is now in a group of 8 year olds.  This makes sense for many reasons; most importantly, he is at the academic level of classmates.  Since he did not speak a word until he was three and half years old and for most of 2012 he was raging and regressing, it also makes sense.  In “learning years” he is, at best, a seven year old.

Until a couple of years ago, all learning (speaking, reading, writing, numeracy) was acquired at home; school was just for practice and socialization.

Socialization is the main point of inclusion, but even that needs a lot of managing.  Socialization without any learning does not seem a clever choice.

The Wider World

In some countries there is a very developed system of Special Education, with the US being far ahead, partly because it diagnoses so many kids to have a special need.

Most other countries now seem to have adopted elements of what is seen as best practice, like having an IEP (Individual Educational Plan) and some interpretation of “inclusion”.  Unless the IEP is well thought out, it is just another stack of paper.  If inclusion is not accompanied by plenty of training and supervision, the results will not be good.

Given the resources for 1:1 education, much can be achieved, but this is rarely going to be possible; only very expensive private schools or home schooling can provide this.

In a large inclusive classroom, I do not see how children with classic autism can make any academic progress, except with the help of a very good 1:1 assistant (but when is there 1:1 time in a noisy inclusive classroom?).  In many inclusive schools, the teachers have had no special training, and quite often, neither has the 1:1 assistant.

Parents often make great efforts to avoid their child going to special education, due to the perceived stigma.  Readers from the US may find this odd, but in most of the world autism remains hidden.  People turn down free intensive early years support, preferring the child to be with typical children.

I see plenty of parents writing commenting things like, “I wish the school would teach my child to read and write”.  Without individual tuition at school and/or home it is easy to see how such kids will not get far at all.

From what appears in the media, most people are not happy with schooling for classic autism.  If you want better, you will have to take on much of the job yourself.

There are plenty of good ideas you can use.

Extended School Year and Duration

In some countries kids with autism have an extended school year, i.e. very short holidays.  This seems a very good idea for both the kids and the parents.  It means that the learning year is more like 11 months long, rather than the typical 9 months.

In most developed countries school finishes when you are 18.  In the US special education in high school continues to 22.  That is quite a big difference, which brings me on to the next point.

Final Academic Level with Classic Autism

I was interested to see what range of academic levels is typical for people with classic autism to achieve when they finish their school education.  It is very hard to find this anywhere and I only found one range, which was between 2nd grade and 6th grade, on leaving “high school”, using the US system.  This seems plausible.

It is clear that many special schools are really focused on living skills rather than academics. 

If you manage to progress academically all the way through school, then it must have been a case of High Functioning Autism or Asperger’s. 

What Monty did

Monty, now aged 11 with ASD, started out un-able to learn in the conventional sense, like most kids with classic autism.

Using an ABA-inspired home program, he did gradually start to learn.  He went to school for socialization and fun.

We have no external agencies, Education Authorities etc. involved in Monty’s education.  We have a nice, responsive, mainstream private school, which has always tried to help, although they have no special needs resources or knowledge.  The class sizes are tiny; this year there are 13 in the group. 

From the age of about 10, things changed sufficiently for school to be about learning.  By that stage he had acquired the academic skills of a typical 7-8 year old, based almost entirely on his supplemental 1:1 tuition.

The home program continues and will be needed for years to come.

Monty has three school years left in Primary before moving on to Secondary/High School.  Primary school is a nice place to be if you have ASD, the same may not be true for Secondary School. 

In the UK system, Secondary school starts when you are 11 years old.  In other countries it starts much later; where we live Secondary school is normally from 14 to 18 years old.

Summertime is no longer developmentally lost, due to the odd effect of allergy and some key neurological autism issues have been identified and treated; more are likely to follow.

I am optimistic that we will see three years of uninterrupted development, twelve months a year.  Every calendar year should be a “learning year”.

Thursday 18 September 2014

GABA A Receptors in Autism – How and Why to Modulate Them

This post will get complicated, since it will look at many aspects of the GABA A receptor, rather than just a small fraction that usually appear in the individual pieces of the scientific literature. 

It was prompted by comments I have received from regular readers, regarding Bumetanide, Clonazepam, epilepsy and whether there might be alternatives with the same effect.  So it is really intended to answer some complex issues. 

There are some new interesting facts/observations that may be of wider interest, just skip the parts that too involved.

Regarding today’s picture, most readers of this blog are female and by the way, while the US is the most common location by far,  a surprisingly high number of page views come from France, Hong Kong, South Africa and Poland.


We have seen that GABA is one of the brain's most important neurotransmitters and we know that various forms of GABA dysfunction are associated with autism, epilepsy and indeed schizophrenia.

One recurring aspect in the research is the so-called excitatory-inhibitory balance of GABA.

The way the brain is understood to function assumes that GABA should be inhibitory and NMDA should be excitatory.

What makes GABA inhibitory is the level of the electrolyte chloride within the cells.  If the level is “wrong”, then GABA may be excitatory and the fine balance required with the NMDA receptor is lost.  The brain then cannot function as intended.

Source: Sage Therapeutics, a company that is developing new drugs that target GABAA and NMDA receptors

To understand what is going wrong in autism and how to treat it we need to take a detailed look at the GABAA receptor and all the anion transport mechanism associated with it.  Most research looks at either the receptor OR the transporters and exchangers.

Anion Transport Mechanisms of the GABAA receptor

You will either need to be a doctor, scientist or very committed to keep reading here.

We know that level of chloride within the cells is critical to whether GABA behaves as excitatory or inhibitory.  This has all been established in the laboratory.

The usual target in autism is the NKCC1 transporter that lets chloride INTO cells, but as you can see in the two figures below, there are other ways to affect the concentration of chloride.  

·        The KCC2 transporter lets chloride out of the cells

·        The sodium dependent anion exchanger (NDAE) lets chloride out of the cells

·        The sodium independent anion exchanger 3 (AE3), lets chloride in.  It extrudes intracellular HCO3- in exchange for extracellular Cl-.

All this does actually matter since we will be able to link it back to a known genetic dysfunction and it would suggest alternative therapeutic avenues.  We can also see how epilepsy fits into the picture.

NKCC1 in Autism

Without doubt, the transporter that controls the flow of chloride into the brain is the expert field of Ben Ari.

His recent summary paper is below:-

He showed that by reducing the level of chloride in the autistic brain using the common diuretic Bumetanide, a marked improvement in many peoples’ autism could be achieved

This post is really about expanding more on what he does not tell us.

KCC2 in Autism

In typical people, very early in life the KCC2 transporter develops and as a result level of chloride falls inside the cells, since the purpose of the transporter is to extrude chloride.

It appears that in autism this mechanism has been disrupted.  The existing science can show us what has gone wrong.

The following study shows that KCC2 is itself regulated by neuroligin-2 (NL2), a cell adhesion molecule specifically localized at GABAergic synapses.

It gets more interesting because the scientists looking for genetic causes of autism have already identified the gene that encodes NL2, which they call NLGN2 (neuroligin 2) as being associated with autism and schizophrenia (adult onset autism).




GABAA receptors are ligand-gated Cl- channels, and the intracellular Cl- concentration governs whether GABA function is excitatory or inhibitory. During early brain development, GABA undergoes functional switch from excitation to inhibition: GABA depolarizes immature neurons but hyperpolarizes mature neurons due to a developmental decrease of intracellular Cl- concentration. This GABA functional switch is mainly mediated by the up-regulation of KCC2, a potassium-chloride cotransporter that pumps Cl- outside neurons. However, the upstream factor that regulates KCC2 expression is unclear.


We report here that KCC2 is unexpectedly regulated by neuroligin-2 (NL2), a cell adhesion molecule specifically localized at GABAergic synapses. The expression of NL2 precedes that of KCC2 in early postnatal development. Upon knockdown of NL2, the expression level of KCC2 is significantly decreased, and GABA functional switch is significantly delayed during early development. Overexpression of shRNA-proof NL2 rescues both KCC2 reduction and delayed GABA functional switch induced by NL2 shRNAs. Moreover, NL2 appears to be required to maintain GABA inhibitory function even in mature neurons, because knockdown NL2 reverses GABA action to excitatory. Gramicidin-perforated patch clamp recordings confirm that NL2 directly regulates the GABA equilibrium potential. We further demonstrate that knockdown of NL2 decreases dendritic spines through down-regulating KCC2.


Our data suggest that in addition to its conventional role as a cell adhesion molecule to regulate GABAergic synaptogenesis, NL2 also regulates KCC2 to modulate GABA functional switch and even glutamatergic synapses. Therefore, NL2 may serve as a master regulator in balancing excitation and inhibition in the brain.

KCC2 in Peripheral nerve injury (PNI)

Autism is not the only diagnosis associated with reduced function of the KCC2 transporter; Peripheral nerve injury (PNI) is another.

In this condition researchers sought to counter the failure of KCC2 to remove chloride from within the cell by increasing the flow chloride through the Cl-/HCO3- anion exchanger known as AE3.


Peripheral nerve injury (PNI) negatively influences spinal gamma-aminobutyric acid (GABA)ergic networks via a reduction in the neuron-specific potassium-chloride (K(+)-Cl(-)) cotransporter (KCC2). This process has been linked to the emergence of neuropathic allodynia. In vivo pharmacologic and modeling studies show that a loss of KCC2 function results in a decrease in the efficacy of GABAA-mediated spinal inhibition. One potential strategy to mitigate this effect entails inhibition of carbonic anhydrase activity to reduce HCO3(-)-dependent depolarization via GABAA receptors when KCC2 function is compromised. We have tested this hypothesis here. Our results show that, similarly to when KCC2 is pharmacologically blocked, PNI causes a loss of analgesic effect for neurosteroid GABAA allosteric modulators at maximally effective doses in naïve mice in the tail-flick test. Remarkably, inhibition of carbonic anhydrase activity with intrathecal acetazolamide rapidly restores an analgesic effect for these compounds, suggesting an important role of carbonic anhydrase activity in regulating GABAA-mediated analgesia after PNI. Moreover, spinal acetazolamide administration leads to a profound reduction in the mouse formalin pain test, indicating that spinal carbonic anhydrase inhibition produces analgesia when primary afferent activity is driven by chemical mediators. Finally, we demonstrate that systemic administration of acetazolamide to rats with PNI produces an antiallodynic effect by itself and an enhancement of the peak analgesic effect with a change in the shape of the dose-response curve of the α1-sparing benzodiazepine L-838,417. Thus, carbonic anhydrase inhibition mitigates the negative effects of loss of KCC2 function after nerve injury in multiple species and through multiple administration routes resulting in an enhancement of analgesic effects for several GABAA allosteric modulators. We suggest that carbonic anhydrase inhibitors, many of which are clinically available, might be advantageously employed for the treatment of pathologic pain states.


Using behavioral pharmacology techniques, we show that spinal GABAA-mediated analgesia can be augmented, especially following nerve injury, via inhibition of carbonic anhydrases. Carbonic anhydrase inhibition alone also produces analgesia, suggesting these enzymes might be targeted for the treatment of pain

Treatment of neuropathic pain is a major clinical challenge that has been met with minimal success. After peripheral nerve injury, a decrease in the expression of the K–Cl cotransporter KCC2, a major neuronal Cl extruder, leads to pathologic alterations in GABAA and glycine receptor function in the spinal cord. The down-regulation of KCC2 is expected to cause a reduction in Cl extrusion capacity in dorsal horn neurons, which, together with the depolarizing efflux of HCO−3 anions via GABAA channels, would result in a decrease in the efficacy of GABAA-mediated inhibition. Carbonic anhydrases (CA) facilitate intracellular HCO−3 generation and hence, we hypothesized that inhibition of CAs would enhance the efficacy of GABAergic inhibition in the context of neuropathic pain. Despite the decrease in KCC2 expression, spinal administration of benzodiazepines has been shown to be anti-allodynic in neuropathic conditions. Thus, we also hypothesized that spinal inhibition of CAs might enhance the anti-allodynic effects of spinally administered benzodiazepines. Here, we show that inhibition of spinal CA activity with acetazolamide (ACT) reduces neuropathic allodynia. Moreover, we demonstrate that spinal co-administration of ACT and midazolam (MZL) act synergistically to reduce neuropathic allodynia after peripheral nerve injury. These findings indicate that the combined use of CA inhibitors and benzodiazepines may be effective in the clinical management of neuropathic pain in humans.

In conclusion, the major finding of the present work is that ACT and MZL act synergistically to inhibit neuropathic allodynia. In light of the available in vitro data reviewed above, a parsimonious way to explain this synergism is that CA inhibition blocks an HCO−3 -dependent positive shift in the Er of GABA and/or glycine-mediated currents and the consequent tonic excitatory drive mediated by extrasynaptic GABAA receptors, while preserving shunting inhibition that is augmented by benzodiazepine actions at postsynaptic GABAA receptors. Obviously, further work is needed at the in vitro level in order to directly examine the cellular and synaptic basis of the ACT-MZL synergism and clinical studies are required to determine the safety of intrathecally applied CA inhibitors in humans. Since MZL and ACT, as well as several other inhibitors of CA [37], are clinically approved, we propose that their use in combination opens up a novel approach for the treatment of chronic neuropathic pain

Midazolam and Acetazolamide

The therapeutic as well as adverse effects of midazolam are due to its effects on the GABAA receptors; midazolam does not activate GABAA receptors directly but, as with other benzodiazepines, it enhances the effect of the neurotransmitter GABA on the GABAA receptors (↑ frequency of Cl− channel opening) resulting in neural inhibition. Almost all of the properties can be explained by the actions of benzodiazepines on GABAA receptors. This results in the following pharmacological properties being produced: sedation, hypnotic, anxiolytic, anterograde amnesia, muscle relaxation and anti-convulsant.

Acetazolamide, usually sold under the trade name Diamox in some countries.  Acetazolamide is a diuretic, and it is available as a (cheap) generic drug.

In epilepsy, the main use of acetazolamide is in menstrual-related epilepsy and as an adjunct in refractory epilepsy.

Acetazolamide is not an immediate cure for acute mountain sickness; rather, it speeds up part of the acclimatization process which in turn helps to relieve symptoms.  I am pretty sure, many years ago, it was Diamox that I took with me when crossing the Himalayas from Nepal into Tibet.  We did not have any problems with mountain sickness.

If periodic paralysis above rings some bells it should.  Two forms already mentioned in this blog are Hypokalemic periodic paralysis and Andersen Tawil syndrome.  We even referred to a paper suggesting the use of Bumetanide.

Acetazolamide is a carbonic anhydrase inhibitor, hence causing the accumulation of carbonic acid Carbonic anhydrase is an enzyme found in red blood cells that catalyses the following reaction:

hence lowering blood pH, by means of the following reaction that carbonic acid undergoes

Anion exchanger 3 (AE3) in autism

Anion exchange protein 3 is a membrane transport protein that in humans is encoded by the SLC4A3 gene. It exchanges chloride for bicarbonate ions.  It increases chloride concentration within the cell.  AE3 is an anion exchanger that is primarily expressed in the brain and heart

Its activity is sensitive to pH. AE3 mutations have been linked to seizures

Bicarbonate (HCO3-) transport mechanisms are the principal regulators of pH in animal cells. Such transport also plays a vital role in acid-base movements in the stomach, pancreas, intestine, kidney, reproductive organs and the central nervous system.


Chloride influx through GABA-gated Cl channels, the principal mechanism for inhibiting neural activity in the brain, requires a Cl gradient established in part by K+–Cl cotransporters (KCCs). We screened for Caenorhabditis elegans mutants defective for inhibitory neurotransmission and identified mutations in ABTS-1, a Na+-driven Cl–HCO3 exchanger that extrudes chloride from cells, like KCC-2, but also alkalinizes them. While animals lacking ABTS-1 or the K+–Cl cotransporter KCC-2 display only mild behavioural defects, animals lacking both Cl extruders are paralyzed. This is apparently due to severe disruption of the cellular Cl gradient such that Cl flow through GABA-gated channels is reversed and excites rather than inhibits cells. Neuronal expression of both transporters is upregulated during synapse development, and ABTS-1 expression further increases in KCC-2 mutants, suggesting regulation of these transporters is coordinated to control the cellular Cl gradient. Our results show that Na+-driven Cl–HCO3 exchangers function with KCCs in generating the cellular chloride gradient and suggest a mechanism for the close tie between pH and excitability in the brain.


During early development, γ-aminobutyric acid (GABA) depolarizes and excites neurons, contrary to its typical function in the mature nervous system. As a result, developing networks are hyperexcitable and experience a spontaneous network activity that is important for several aspects of development. GABA is depolarizing because chloride is accumulated beyond its passive distribution in these developing cells. Identifying all of the transporters that accumulate chloride in immature neurons has been elusive and it is unknown whether chloride levels are different at synaptic and extrasynaptic locations. We have therefore assessed intracellular chloride levels specifically at synaptic locations in embryonic motoneurons by measuring the GABAergic reversal potential (EGABA) for GABAA miniature postsynaptic currents. When whole cell patch solutions contained 17–52 mM chloride, we found that synaptic EGABA was around −30 mV. Because of the low HCO3 permeability of the GABAA receptor, this value of EGABA corresponds to approximately 50 mM intracellular chloride. It is likely that synaptic chloride is maintained at levels higher than the patch solution by chloride accumulators. We show that the Na+-K+-2Cl cotransporter, NKCC1, is clearly involved in the accumulation of chloride in motoneurons because blocking this transporter hyperpolarized EGABA and reduced nerve potentials evoked by local application of a GABAA agonist. However, chloride accumulation following NKCC1 block was still clearly present. We find physiological evidence of chloride accumulation that is dependent on HCO3 and sensitive to an anion exchanger blocker. These results suggest that the anion exchanger, AE3, is also likely to contribute to chloride accumulation in embryonic motoneurons.

Sodium dependent anion exchanger (NDAE)

Not much has been written about these exchangers, outside of very technical literature.

Sodium-coupled anion exchange is activated by intracellular acidification (Schwiening and Boron, 1994), suggesting that regulation of the chloride gradient by NDAEs may be closely linked to the regulation of cellular pH. As prolonged neuronal activity can cause neuronal acidification by efflux of bicarbonate through GABAA receptors (Kaila and Voipio, 1987), sodium-coupled anion exchange may help to maintain a hyperpolarizing chloride reversal potential and thus promote the inhibitory action of GABA. Thus activation of sodium-coupled anion exchange by acidosis may also contribute to seizure termination by promoting a more negative chloride reversal potential and thus promoting the inhibitory effects of GABA.

The GABAA receptor  (background is cut and paste from Wikipedia)

In order for GABAA receptors to be sensitive to the action of benzodiazepines they need to contain an α and a γ subunit, between which the benzodiazepine binds. Once bound, the benzodiazepine locks the GABAA receptor into a conformation where the neurotransmitter GABA has much higher affinity for the GABAA receptor, increasing the frequency of opening of the associated chloride ion channel and hyperpolarizing the membrane. This potentiates the inhibitory effect of the available GABA leading to sedative and anxiolytic effects.

Structure and function

Schematic diagram of a GABAA receptor protein ((α1)2(β2)2(γ2)) which illustrates the five combined subunits that form the protein, the chloride (Cl-) ion channel pore, the two GABA active binding sites at the α1 and β2 interfaces, and the benzodiazepine (BDZ) allosteric binding site

The receptor is a pentameric transmembrane receptor that consists of five subunits arranged around a central pore. Each subunit comprises four transmembrane domains with both the N- and C-terminus located extracellularly. The receptor sits in the membrane of its neuron, usually localized at a synapse, postsynaptically. However, some isoforms may be found extrasynaptically. The ligand GABA is the endogenous compound that causes this receptor to open; once bound to GABA, the protein receptor changes conformation within the membrane, opening the pore in order to allow chloride anions (Cl) to pass down an electrochemical gradient. Because the reversal potential for chloride in most neurons is close to or more negative than the resting membrane potential, activation of GABAA receptors tends to stabilize or hyperpolarise the resting potential, and can make it more difficult for excitatory neurotransmitters to depolarize the neuron and generate an action potential. The net effect is typically inhibitory, reducing the activity of the neuron. The GABAA channel opens quickly and thus contributes to the early part of the inhibitory post-synaptic potential (IPSP).


GABAA receptors are members of the large "Cys-loop" super-family of evolutionarily related and structurally similar ligand-gated ion channels that also includes nicotinic acetylcholine receptors, glycine receptors, and the 5HT3 receptor. There are numerous subunit isoforms for the GABAA receptor, which determine the receptor's agonist affinity, chance of opening, conductance, and other properties.
In humans, the units are as follows:
There are three ρ units (GABRR1, GABRR2, GABRR3), however these do not coassemble with the classical GABAA units listed above,[18] but rather homooligomerize to form GABAA-ρ receptors (formerly classified as GABAC receptors but now this nomenclature has been deprecated[19] ).
Five subunits can combine in different ways to form GABAA channels. The minimal requirement to produce a GABA-gated ion channel is the inclusion of both α and β subunits, but the most common type in the brain is a pentamer comprising two α's, two β's, and a γ (α2β2γ)
The receptor binds two GABA molecules, at the interface between an α and a β subunit

The important subunits for this post are:-
Very little is written about this subunit.

While the effect of editing on protein function is unknown, the developmental increase in editing does correspond to changes in function of the GABAA receptor. GABA binding leads to chloride channel activation, resulting in rapid increase in concentration of the ion. Initially, the receptor is an excitatory receptor, mediating depolarisation (efflux of Cl- ions) in immature neurons before changing to an inhibitory receptor, mediating hyperpolarization(influx of Cl- ions) later on. GABAA converts to an inhibitory receptor from an excitatory receptor by the upregulation of KCC2 cotransporter. This decreases the concentration of Cl- ion within cells. Therefore, the GABAA subunits are involved in determining the nature of the receptor in response to GABA ligand. These changes suggest that editing of the subunit is important in the developing brain by regulating the Cl- permeability of the channel during development. The unedited receptor is activated faster and deactivates slower than the edited receptor.

Editing of the I/M site is developmentally regulated

A switch in the GABA response from excitatory to inhibitory post-synaptic potentials occurs during early development where an efflux of chloride ions takes place in immature neurons, while there is an influx of chloride ions in mature neurons (Ben-Ari 2002). GABA switches from being excitatory to inhibitory by an up-regulation of the cotransporter KCC2 that decreases the chloride concentration in the cell. However, if GABA itself promotes the expression of KCC2 is still under debate (Ganguly et al. 2001; Ludwig et al. 2003; Titz et al. 2003). Further, the α subunits are critical elements in determining the nature of the GABAA receptor response to GABA (Böhme et al. 2004). The α3 mRNA (Gabra-3) is present at high levels in several forebrain regions at birth with a major decline after post-natal day 12 (P12), when the expression of α1 is going up (Laurie et al. 1992). The change from α3 to α1 may cause the switch in GABA behavior from excitatory to inhibitory post-synaptic potentials during development.
GABAA receptors respond to anxiolytic drugs such as benzodiazepines and are thus important drug targets. The benzodiazepine binding site is located at the interface of the α and γ2 subunits (Cromer et al. 2002). Antagonists that bind to this site enhance the effect of GABA by increasing the frequency of GABA-induced channel opening events. Post-transcriptional modifications of the α3 subunit, such as the I/M editing described here, could be important in determining the mechanistic features that are responsible for the diversity of GABAA receptors and the variability in sensitivity to drugs


A number of ligands have been found to bind to various sites on the GABAA receptor complex and modulate it besides GABA itself.


  • Agonists: bind to the main receptor site (the site where GABA normally binds, also referred to as the "active" or "orthosteric" site) and activate it, resulting in increased Cl conductance.
  • Antagonists: bind to the main receptor site but do not activate it. Though they have no effect on their own, antagonists compete with GABA for binding and thereby inhibit its action, resulting in decreased Cl conductance.
  • Positive allosteric modulators: bind to allosteric sites on the receptor complex and affect it in a positive manner, causing increased efficiency of the main site and therefore an indirect increase in Cl conductance.
  • Negative allosteric modulators: bind to an allosteric site on the receptor complex and affect it in a negative manner, causing decreased efficiency of the main site and therefore an indirect decrease in Cl conductance.
  • Open channel blockers: prolong ligand-receptor occupancy, activation kinetics and Cl ion flux in a subunit configuration-dependent and sensitization-state dependent manner.
  • Non-competitive channel blockers: bind to or near the central pore of the receptor complex and directly block Cl- conductance through the ion channel.

The GABAA receptor include a site where benzodiazepine can bind.  These are drugs that include like valium. Binding at this site increase the effect of GABA.  Since this receptor is meant to be inhibitory, giving valium should make it strong inhibitory, ie calming. 

It was noted that in autism the effect of valium was often the reversed, instead of calming it further increased anxiety.

The Valium is working just fine, it is magnifying the effect the effect of GABA, the problem is that the receptor is functioning as excitatory, the Valium is making it over-excitatory.  Now we come to the reason why.

We know that the excitatory-inhibitory balance is set by the chloride concentration within the cells.  We also know that exact mechanism that determines this level.


BTBR mice have reduced spontaneous GABAergic inhibitory transmission
Nonsedating doses of benzodiazepines improved autism-related deficits in BTBR mice
Impairment of GABAergic transmission reduced social interaction in wild-type mice
Behavioral rescue by low-dose benzodiazepine is GABAA receptor α2,3-subunit specific


Autism spectrum disorder (ASD) may arise from increased ratio of excitatory to inhibitory neurotransmission in the brain. Many pharmacological treatments have been tested in ASD, but only limited success has been achieved. Here we report that BTBR T+ Itpr3tf/J (BTBR) mice, a model of idiopathic autism, have reduced spontaneous GABAergic neurotransmission. Treatment with low nonsedating/nonanxiolytic doses of benzodiazepines, which increase inhibitory neurotransmission through positive allosteric modulation of postsynaptic GABAA receptors, improved deficits in social interaction, repetitive behavior, and spatial learning. Moreover, negative allosteric modulation of GABAA receptors impaired social behavior in C57BL/6J and 129SvJ wild-type mice, suggesting that reduced inhibitory neurotransmission may contribute to social and cognitive deficits. The dramatic behavioral improvement after low-dose benzodiazepine treatment was subunit specific—the α2,3-subunit-selective positive allosteric modulator L-838,417 was effective, but the α1-subunit-selective drug zolpidem exacerbated social deficits. Impaired GABAergic neurotransmission may contribute to ASD, and α2,3-subunit-selective positive GABAA receptor modulation may be an effective treatment.
These results indicate that different subtypes of GABAA receptors may have opposite roles in social behavior, with activation of GABAA receptors containing α2,3 subunits favoring and of GABAA receptors with α1  subunits reducing social interaction, respectively.

Because of their broad availability and safety, benzodiazepines and other positive allosteric modulators of GABAA receptors administered at low nonsedating, nonanxiolytic doses that do not induce tolerance deserve consideration as a near-term strategy to improve the core social interaction deficits and repetitive behaviors in ASD.

These results are most consistent with the hypotheses that reduced inhibitory neurotransmission is sufficient to induce autistic-like behaviors in mice and that enhanced inhibitory neurotransmission can reverse autistic-like behaviors.


I have received various comments about epilepsy.  Epilepsy has many variants, just like autism.  Epilepsy is often comorbid with autism.  GABA dysfunction is known to be closely involved in some types of autism and some types of epilepsy.

It is known that Bumetanide has very different effects in different types of epilepsy.
The question that naturally arises is whether you can give Bumetanide to someone who has autism and epilepsy and if you cannot, is there an alternative with the same desired effect?

Well it appears that any method that changes chloride levels is likely to affect epilepsy.  It appears that all three methods (NKCC1, KCC2 and AE3) would likely have the same impact on epilepsy.

But would it be a good effect or a bad effect?

Would it interact with any existing anti-epilepsy drugs?

I suspect that Bumetanide might be an effective anti-epileptic in people with autism and that other GABA related drugs might no longer be needed.  Quite likely the effect of Bumetanide and the anti-epileptic targeting GABA might be too much.  So the blog reader that pointed out that the bumetanide clinical trial excluded children with epilepsy has highlighted an important point.

While epilepsy is not fully understood and there are various variants, it would seem plausible that the epilepsy common in core classic autism and early regressive autism is the same type and that it is linked to the same excitatory/inhibitory dysfunction.

You may be wonder if other diuretics have anti-epileptic properties. Here is a paper by a Neurologist from Denver on the subject:-

Why is there an excitatory/inhibitory dysfunction in Autism?

People are writing entire books on the GABA excitatory/inhibitory balance.  I was curious as to why this dysfunction exists at all in autism. 

We learnt from Ben-Ari in earlier posts all about this switch from excitatory to inhibitory that is supposed to occur very early on in life, we now have two reasons why this may fail to happen in autism:-

1.     Editing modifies the GABAA receptor subunit α3.  The change from α3 to α1 may cause the switch in GABA behavior from excitatory to inhibitory post-synaptic potentials during development.  This change appears not to occur in some types of autism.  We see from the Clonazepam research that α3 and  α1 have opposite effects in autism.   In autism, activation of GABAA receptors containing α2,3 subunits favours social interaction  and activation of α1  subunits reduces social interaction.


2.     The GABA functional switch is mainly mediated by the up-regulation of KCC2, a potassium-chloride cotransporter that pumps Cl- outside neurons.  NL2 also regulates KCC2 to modulate GABA functional switch. Therefore, NL2 may serve as a master regulator in balancing excitation and inhibition in the brain.  The gene that encodes NL2 is called NLGN2 (neuroligin 2).  Dysfunction in gene NLGN2 is known to occur in both autism and schizophrenia (adult onset autism).


We came full circle back to Bumetanide and Clonazepam as most likely the safest and most effective therapy to adjust the E/I (excitatory/inhibitory) balance in autism. KCCI agonists do not seem to exist.  The bicarbonate exchanger agonist Acetazolamide/Diamox is another common diuretic and I see no reason why it would not also be effective, but we would then affect bicarbonate levels.  Since these ions play a role in controlling pH levels, I think we might risk seeing some unintended effects.  We know that Bumetanide is safe in long term use.  We know that all diuretics that change chloride level within the cell and will affect epilepsy; so it is a case of “better the devil you know”. 

I finally understood exactly why tiny dose of Clonazepam are effective and how this fits in with the changes the Bumetanide has produced.  Thankfully, such tiny doses are free of the typical side effects expected from benzodiazepines.  One tablet lasts 10 days.

It also answers somebody else’s question about starting with Clonazepam before the Bumetanide.  If you did that you might well make things much worse, you would magnify the unwanted excess brain cell firing.  Once you added bumetanide things would then reverse and brain cell firing would be inhibited.

I rather like the parallel with neuropathic pain, the other condition we looked at with reduced KCC2 transporter function, the researchers there proposed the combination of a diuretic (Acetazolamide) to lower cellular chloride (via exchanger AE3) and a benzodiazepine (Midazolam) as a positive allosteric modulator.  This is extremely similar to Ben Ari’s bumetanide (diuretic affecting transporter NKCC1) plus Catterall’s tiny doses of clonazepam (benzodiazepine) as a positive allosteric modulator.

As for epilepsy and bumetanide, we know that bumetanide has different effects on different types of autism. It seems plausible that people with autism might tend to have the same type of epilepsy.  In any case Monty, aged 11 with ASD, does not have epilepsy/seizures and I suspect taking bumetanide has decreased the chance he ever will.  Of course I cannot prove this, it is just conjecture.