Showing posts with label Purkinje cells. Show all posts
Showing posts with label Purkinje cells. Show all posts

Monday, 23 January 2017

The Purkinje-RORa-Estradiol-Neuroligin-KCC2 axis in Autism

Add testosterone/estradiol to those dysfunctional hormones

This blog is about noticing connections and making things a little simpler to understand.  Today’s post is going to be a good example; all those odd sounding things like Purkinje cells and neuroligins all fitting nicely together.

Today we see how a central hormonal dysfunction (testosterone/estradiol) can lead to an ion channel dysfunction (NKCC1/KCC2) at one end of the chain and at the other explains the absence of many Purkinje cells in the autistic cerebellum, which leads to some of the observed features of autism.

I am calling it the Purkinje-RORa-Estradiol-Neuroligin-KCC2 axis, or Purkinje-KCC2 axis for short.

We also get to see how melatonin fits in here and see why disturbed sleeping patterns should be expected in someone affected by the Purkinje- KCC2 axis.

I should point out that not everyone with autism is likely affected by the Purkinje-NKCC1 axis, but I think it will apply to a majority of those with non-regressive, multigenic, strictly defined autism (SDA).

We saw in a recent post how the enzyme aromatase acts in the so-called  testosterone – estradiol shunt.

I suggested that lack of aromatase was leading to too little estradiol which then affected neuroligin 2 (NL2) which then caused down-regulation of the KCC2 cotransporter that takes chloride out of neurons. This then caused neurons to remain in a permanent immature state.

Digging a little deeper we find recent research that shows how the control loops that balance aromatase act through RORA/RORα, RORa  (retinoic acid-related orphan receptor alpha.

The schematic illustrates a mechanism through which the observed reduction in RORA in autistic brain may lead to increased testosterone levels through downregulation of aromatase. Through AR, testosterone negatively modulates RORA, whereas estrogen upregulates RORA through ER.

androgen receptor = AR

estrogen receptor = ER

RORα (retinoic acid-related orphan receptor alpha.)

RORα certainly has a long full name. Retinoic acid is a metabolite of vitamin A (retinol).

RORα does some clever things.

RORα is necessary for normal circadian rhythms

ROR-alpha is expressed in a variety of cell types and is involved in regulating several aspects of development, inflammatory responses, and lymphocyte development

RORα is involved in processes that regulate metabolism, development, immunity, and circadian rhythm and so shows potential as drug targets. Synthetic ligands have a variety of potential therapeutic uses, and can be used to treat diseases such as diabetes, atherosclerosis, autoimmunity, and cancer. T0901317 and SR1001, two synthetic ligands, have been found to be RORα and RORγ inverse agonists that suppress reporter activity and have been shown to delay onset and clinical severity of multiple sclerosis and other Th17 cell-mediated autoimmune diseases. SR1078 has been discovered as a RORα and RORγ agonist that increases the expression of G6PC and FGF21, yielding the therapeutic potential to treat obesity and diabetes as well as cancer of the breast, ovaries, and prostate. SR3335 has also been discovered as a RORα inverse agonist.

RORs are also called nuclear melatonin receptors. Many people with autism take melatonin to balance circadian rhythms and fall asleep.

The reduced estrogen levels in women during menopause likely caused them not to sleep due to the effect on RORα.

So it would appear that some of what is good for menopausal women may actually be helpful for some people with autism.

Many Genes affected by RORα

Most exciting, the researchers say, is that 426 of RORA’s gene targets are listed in AutismKB, a database of autism candidates maintained by scientists at Peking University in Beijing, and 49 in SFARI Gene.

Therapeutic Effect of a Synthetic RORα/γ Agonist in an Animal Model of Autism

Autism is a developmental disorder of the nervous system associated with impaired social communication and interactions as well excessive repetitive behaviors. There are no drug therapies that directly target the pathology of this disease. The retinoic acid receptor-related orphan receptor α (RORα) is a nuclear receptor that has been demonstrated to have reduced expression in many individuals with autism spectrum disorder (ASD). Several genes that have been shown to be downregulated in individuals with ASD have also been identified as putative RORα target genes. Utilizing a synthetic RORα/γ agonist, SR1078, that we identified previously, we demonstrate that treatment of BTBR mice (a model of autism) with SR1078 results in reduced repetitive behavior. Furthermore, these mice display increased expression of ASD-associated RORα target genes in both the brains of the BTBR mice and in a human neuroblastoma cell line treated with SR1078. These data suggest that pharmacological activation of RORα may be a method for treatment of autism.

For those who like natural substances, some research from Japan.


The retinoic acid receptor-related orphan receptors α and γ (RORα and RORγ), are key regulators of helper T (Th)17 cell differentiation, which is involved in the innate immune system and autoimmune disorders. In this study, we investigated the effects of isoflavones on RORα/γ activity and the gene expression of interleukin (IL)-17, which mediates the function of Th17 cells. In doxycycline-inducible CHO stable cell lines, we found that four isoflavones, biochanin A (BA), genistein, formononetin, and daidzein, enhanced RORα- or RORγ-mediated transcriptional activity in a dose-dependent manner. In an activation assay of the Il17a promoter using Jurkat cells, these compounds enhanced the RORα- or RORγ-mediated activation of the Il17a promoter at concentrations of 1 × 10(-6)M to 1 × 10(-5)M. In mammalian two-hybrid assays, the four isoflavones enhanced the interaction between the RORα- or RORγ-ligand binding domain and the co-activator LXXLL peptide in a dose-dependent manner. In addition, these isoflavones potently enhanced Il17a mRNA expression in mouse T lymphoma EL4 cells treated with phorbol myristate acetate and ionomycin, but showed slight enhancement of Il17a gene expression in RORα/γ-knockdown EL4 cells. Immunoprecipitation and immunoblotting assays also revealed that BA enhanced the interaction between RORγt and SRC-1, which is a co-activator for nuclear receptors. Taken together, these results suggest that the isoflavones have the ability to enhance IL-17 gene expression by stabilizing the interactions between RORα/γ and co-activators. This also provides the first evidence that dietary chemicals can enhance IL-17 gene expression in immune cells.

Genistein is a common supplement.  It is a pytoestrogen and unfortunately these substances lack potency in real life.  In test tubes they have interesting properties, but they are poorly absorbed when taken orally and so unless they are modified they are likely to have no effect in the usual tiny doses used in supplements.

This is true with very many products sold as supplements.

Sometimes care is taken to improve bioavailability as with some expensive curcumin supplements, like Longvida.

Trehalose, a supplement referred to recently in comments on this blog, is another interesting natural substance that lacks bioavailablity.  Analogs of this natural substance have been produced that are much better absorbed and are now potential drugs.

Purkinje Cells

Back in 2013 I wrote a post about Purkinje cells.

          Pep up those Purkinje cells

Loss of Purkinje cells is one of the few non-disputed abnormalities in autism. 

These cells are some of the largest neurons in the human with an intricately elaborate dendritic arbor, characterized by a large number of dendritic spines. Purkinje cells are found within the Purkinje layer in the cerebellum. Purkinje cells are aligned like dominos stacked one in front of the other. Their large dendritic arbors form nearly two-dimensional layers through which parallel fibers from the deeper-layers pass. These parallel fibers make relatively weaker excitatory (glutamatergic) synapses to spines in the Purkinje cell dendrite, whereas climbing fibers originating from the inferior olivary nucleus in the medulla provide very powerful excitatory input to the proximal dendrites and cell soma. Parallel fibers pass orthogonally through the Purkinje neuron's dendritic arbor, with up to 200,000 parallel fibers[2] forming a Granule-cell-Purkinje-cell synapse with a single Purkinje cell. Each Purkinje cell receives ca 500 climbing fiber synapses, all originating from a single climbing fiber.[3] Both basket and stellate cells (found in the cerebellar molecular layer) provide inhibitory (GABAergic) input to the Purkinje cell, with basket cells synapsing on the Purkinje cell axon initial segment and stellate cells onto the dendrites.

Purkinje cells send inhibitory projections to the deep cerebellar nuclei, and constitute the sole output of all motor coordination in the cerebellar cortex.

In humans, Purkinje cells can be harmed by a variety causes: toxic exposure, e.g. to alcohol or lithium; autoimmune diseases; genetic mutations causing spinocerebellar ataxias, Unverricht-Lundborg disease, or autism; and neurodegenerative diseases that are not known to have a genetic basis, such as the cerebellar type of multiple system atrophy or sporadic ataxias.

Purkinje cells are some of the largest neurons in the human brain and the most important.

Neuronal maturation during development is a multistep process regulated by transcription factors. The transcription factor RORα (retinoic acid-related orphan receptor α) is necessary for early Purkinje cell maturation but is also expressed throughout adulthood.

The active form (T3) of thyroid hormone  controls critical aspects of cerebellar development, such as migration of postmitotic neurons and terminal dendritic differentiation of Purkinje cells. T3 action on the early Purkinje cell dendritic differentiation process is mediated by RORα.

In autism we have seen that oxidative stress may lead to low levels of T3 in the autistic brain.  We now see that low levels of RORα are also likely in autsim.

The combined effect would help explain the loss of Purkinje cells in autism.

Neuropathological studies, using a variety of techniques, have reported a decrease in Purkinje cell (PC) density in the cerebellum in autism. We have used a systematic sampling technique that significantly reduces experimenter bias and variance to estimate PC densities in the postmortem brains of eight clinically well-documented individuals with autism, and eight age- and gender-matched controls. Four cerebellar regions were analyzed: a sensorimotor area comprised of hemispheric lobules IV–VI, crus I & II of the posterior lobe, and lobule X of the flocculonodular lobe. Overall PC density was thus estimated using data from all three cerebellar lobes and was found to be lower in the cases with autism as compared to controls. These findings support the hypothesis that abnormal PC density may contribute to selected clinical features of the autism phenotype.

Estradiol – Neuroligin 2 to KCC2

We saw in a recent post how reduced levels of estradiol could lead to KCC2 underexpression via the action of neuroligin 2.


So in my grossly oversimplified world of autism, I think I have a plausible case for the Purkinje-KCC2 axis.  I think that in addressing this axis numerous other issues would also be solved ranging from sleep issues to those hundreds of other genes whose regulation is at least partly governed by RORα.

The KCC2 end of the axis can be treated by bumetanide, diamox/acetazolamide, potassium bromide and possibly by intranasal IGF-1/insulin.  

How to address the rest of the Purkinje-KCC2 axis?

·        More RORα, or just a RORα agonist.

·        More aromatase

·        Genistein may help, but you would need it by the bucket load, due to bioavailability issues

·        Estrogen receptor agonists

·        Exogenous estradiol

The simplest is the last one and really should be trialed on adult males with autism.  The dose would need to be much lower than the feminizing dose, so 0.2mg would seem a good starting dose for such a study.

Due to the feedback loops somethings may work short term, but not long term. 

Wednesday, 28 August 2013

Potassium may play an important role in Autistic Behaviours

This is not the kind of post that I expected to be writing.  How can the effect of something so simple as a mineral, not have been noticed by others and researched in depth?

Potassium (K+),  is one of several electrolytes that occur in humans, the others being sodium(Na+), calcium (Ca2+), magnesium (Mg2+), chloride (Cl), hydrogen phosphate (HPO42−), and hydrogen carbonate (HCO3)

Electrolyte balance or homeostasis is regulated by specific hormones.  These electrolytes are used to control many aspects of your body.  The concentration of each electrolyte varies across the boundary of each cell.  Electrolytes pass through the cell wall/membrane through so called ion channels.  These are like special valves that open and close based on particular pre-programmed circumstances.  When these ion channels malfunction, often due to a genetic fault, disease occurs.  Ion channel diseases have a special name - channelopathies.  This is still an emerging area of science.

In autism the brain has developed in an unusual way and although it is thankfully not a degenerative disease, the biological equilibrium it has evolved to is not the one originally intended.  There are both channelopathies and hormonal irregularities; indeed the two are interrelated.

Many hormones are interrelated and have multiple functions and therefore a change in one may have a cascading effect on others.  The same applies to the electrolytes, for example a deficiency in magnesium will trigger a deficiency in potassium.

Choride (Cl-)

I started my blog when I read about a successful clinical trial that set out to prove whether an imbalance in chloride between the extra/intra cellular fluid could cause one of the brain’s main neurotransmitters (GABA) to malfunction.  A clever Frenchman called Ben-Ari, had been researching neonatal seizures and proposed to trial the drug Bumetanide.    Bumetanide is known to block the NKCC1 cation-chloride co-transporter, and thus decreases internal chloride concentration in neurons. In turn, this concentration change makes the action of GABA more hyperpolarizing.  Do not be put off if this does not make sense to you.

The trial showed the positive effect on autistic behaviours of this long established and inexpensive drug.
A randomised controlled trial of bumetanide in the treatment of autism inchildren

Potassium (K+)

Not long after making a trial of bumetanide on Monty, then aged 9 years, I started this blog and my own research.  I soon came across a condition called Hypokalemic Sensory Overstimulation.  In this condition, the subject becomes overwhelmed by his senses of sound, light, smell etc.  After taking oral potassium, the symptoms disappear within 20 minutes.  It is claimed that this is also a characteristic of ADHD (attention deficit hyperactivity disorder).  Well sensory overload is pretty common in autism, and, as I have learnt, ADHD is really just a light case of autism.

There is almost no research into this condition, which is odd since it is linked to the very common ADHD condition.  The paper below  was only ever cited 3 times in other research, and only once in English.

I then did my own experiment using a small dose of K+  supplement (equivalent to one banana) to see if by any chance I could see a reduction in sensory overload.  I tested both my sons, and only in the ASD son did the potassium have any impact; and it was a marked impact.  I wrote this up in a blog post.

I kept potassium and channelopathy on my list of things to research and left it at that.

Ion Channels & Ion Channel Diseases (Channelopathies)

Ion channels are an emerging area of science all about how signals are sent throughout your body to control it.  It gets very complicated and is still far from fully understood.  So you may want to skip this part.

So far 300 types of ion channel have been identified.  The main types are:-

·         Chloride (Cl-) channels

·         Potassium channels

·         Sodium channels

·         Calcium channels

·         Proton channels

·         Non-selective cation channels

 Then there are differing ways in which the channels open and close such as:-

·         Voltage gated

·         Ligand gated

And odd ones like
·         Light gated

·         Temperature gated

·         Calcium activated potassium channels

When the ion channel and/or its gating does not work properly then a disease called a channelopathy may result.  Examples of well know disease are cystic fibrosis, various types of epilepsy and ataxia.

Puberty and Epilepsy  

As a result of the changes in hormones triggered by puberty it is therefore not surprising that around this time other changes occur in the body.  In some children with asthma, their symptoms become more mild or even appear to disappear.  In autism the hormonal changes often trigger an improvement, but may be the trigger of the onset of epilepsy.  When you consider the importance of all these electrolyte levels, and the variation of each across one each cell boundary in the body and how this is intertwined with how the neurotransmitters function, it is not surprising that a shift in Homeostatis occurs.

That shift in Homeostatis could be reflected in a mellowing of autistic characteristics.  But if you can now make some small adjustments in these levels via diet and mild drugs, why not investigate it?  You will not be able to achieve perfection, but you might be able to shift from one stable equilibrium to another one, with milder autism and no troubling side effects.

This would also imply that those children developing epilepsy during puberty might be able to treat it using the diuretic bumetanide.  By blocking the NKCC1 transporter, the level of Cl- is blocked and GABA becomes more inhibitory and thus the risk of an epileptic attack might be reduced, or perhaps eliminated.  This is surely worth some research?

K+ ion channel disease – Epilepsy & Autism

There is existing research linking potassium ion channels to both epilepsy and autism

ADHD & Ritalin

I read some research about a stimulant drug used to calm children with ADHD.  It seemed odd to use a stimulant to produce calm.  Here again potassium (K+) and sodium (Na+) levels are at the centre of argument.
Then I noted a very recent article (July 2013) reporting a study of Ritalin on children with ASD and/or ADHD.

In the world of alternative medicine there is talk of Ritalin helping in ADHD due to it altering the level of potassium:-

This inverted ratio of Na/K may be helpful in explaining why a stimulant drug like Ritalin would have a calming effect on hyperactive children and adolescents. Ritalin does, indeed, have a stimulating effect on these children, but its stimulating mechanism is neuroendocrine and biochemical, not behavioural. More specifically, its stimulating effect is on the adrenal glands and the retention of sodium in the tissues relative to potassium. It is critical for normal cellular functioning that sodium and potassium (Na and K) be in balance for the optimal operation of the Na/K pump at the cellular level. It is also critical for efficient neurotransmissions that there be a proper balance between sodium and potassium (Na and K) for neuronal conductance.

Measuring electrolyte (Ka, Na, Mg, Cl etc.) levels

I would have expected that it was easy to check the level of electrolytes and indeed to check the levels in blood is very easy.  I have done this and all was normal.  When you read further in the literature, you will realize that the level of electrolytes at the extra/intra cellular level is not so easy to measure.  A whole business has been created by people analyzing hair samples for clues as to the balance or imbalance of various minerals in the body.

Hair analysis is used in forensic toxicology and  drug testing to detect the presence of various chemicals in the body.  The method has been adopted by the complementary and alternative medicine (CAM) community to try to predict food intolerance and dietary deficiency.  It is viewed by the scientific community that much of the CAM use of the technology is not valid and potentially fraudulent,

So there is no certain way of checking the cellular level of electrolytes.  You can only measure what you eat and you can measure what is in your blood.

Other suspected electrolyte imbalances in Autism

There have been several studies regarding Magnesium and autism, but the peer reviews of these studies are highly critical of the methodologies used and conclude that nothing has been proven.

Vitamin B6-magnesium treatment for autism: the current status of the research

Calcium has also been put forward as an intervention.  One mother spent a great deal of time collecting supporting information in her paper below.

DAN doctors and Spironolactone

Having come across a “bible” of therapies proposed by DAN (Defeat Autism Now) doctors I noted the use of a potassium-sparing diuretic called Spironolactone.  For a change, there is actually a published paper setting out their case for this drug.  The case made has nothing to do with potassium, even though the intended purpose of the drug is to raise potassium levels.

Bradstreet et al wrote a paper on this in 2006.  It has been cited only 7 times up until 2013 and two of these times by the authors themselves.  This tells you that other researchers were either skeptical or just disinterested.

Potassium Supplements
There are many hundreds of types of mineral and vitamin supplements; these days many contain far more than recommended daily amount (RDA).  This is not the case with potassium.  Even though the RDA for adults is 3,500 – 4,700 mg, in the US supplements by law may not contain more than 100 mg of potassium.  In Europe potassium supplements with 500mg are common.

A typical banana contains over 400mg of potassium, which would seem to make a 100mg supplement pretty pointless.

In the US it seems that there is a perceived fear of potassium poisoning.  It is indeed the home of the lethal injection.  Potassium chloride in a very high dose will stop your heart.

You will even find people debating whether you can poison yourself with bananasIt would seem that while you can reduce your high blood pressure with bananas, it does not kill you.  One person was even eating 30 bananas a day!!

The fear of potassium though remains and it is all over the American internet.
But, in the UK, the National Health Service advises:-

You should be able to get all the potassium you need by eating a varied and balanced diet. If you take potassium supplements, do not take too much because this could be harmful.

Taking 3,700mg or less of potassium supplements a day is unlikely to cause any harm.

That amount of potassium would require 37 American supplement tablets each and every day!

In reality, a concentrated dose of potassium may indeed upset your stomach and this is why it is better to get it from a healthy mix of fruit and vegetables.  The average American apparently consumes about 1,000mg of potassium per day.

Internet Chatter

If, like me, you use Google to see what other people are up to, you will come across talk of potassium and autism.  The discussions in forums never get far, because someone starts talking about lethal injections, and then fear prevails.

Sodium Potassium Pump  (Na+/K+-ATPase)

The sodium-potassium pump was discovered in the 1957 by a Danish Scientist, who later went on to win a Nobel prize for his discovery in 1997.  Its main application has been in the understanding and treatment of heart disease, but it is now thought to be directly involved in a critical part of the brain already known to be damaged in autism. 

As I have already mentioned in my blog, the comorbidities of autism (asthma, high cholesterol etc.) mean that much of the work has already been done by others.

Those many people with hypertension (high blood pressure) are suffering due the way the sodium potassium pump works.  They eat too much sodium and far too little potassium and the end result is high blood pressure.

An author and researcher, Dr Richard Moore, has a simple explanation on his website and a link to his book showing how diet can indeed control your blood pressure.  If you check the book on Amazon you will see many very favourable comments from people who have indeed lowered their blood pressure with bananas.

His book is called: The High Blood Pressure Solution: A Scientifically Proven Program for Preventing Strokes and Heart Disease.

Sodium Potassium Pump and Autism

I was looking for evidence (other than my own) that potassium levels affect the autistic brain.  Potassium plays a key role in how most ion channels function, but I was looking for something really tangible.  I think I have found it.

In my earlier posts I introduced readers to a part of the cerebellum called the Purkinje Cell Layer (PCL).  This is a critical part of the brain and unfortunately in autism, half of the cells are dead and this then manifests itself in altered brain functioning and hence behaviour.  

As recently as 2012, scientists in England showed that the neurons in the PCL are controlled by the Sodium Potassium Pump.

The paper’s summary concludes “We propose that Na+/K+ pump activity controls the intrinsic firing mode of cerebellar Purkinje cells”
Our new friend Dr Richard Moore puts it very simply:-

For the Na-K-pump to operate normally, the diet must have a ratio of potassium to sodium ratio (the K/Na ratio, or "K Factor") that is above a threshold that is somewhere between 2 and 4. Our ancestors ate a diet with a K/Na ratio ranging between 12 and 16. However, the average American white eats a diet with a K/Na ratio of less than 1 - about 0.6 - and the average American black eats a diet with an average K/Na ratio of about only 0.38! Obviously, the American diet generally has a very deficient K/Na ratio.

A low dietary K/Na ratio causes a low K/Na ratio in each and every cell in your body. This has been known since the end of World War II when whole body radio-active counters were used to determine the amount of potassium (a small part of which is naturally radioactive) in the human body. Almost universally, to their surprise, it was found that people with hypertension have a deficient amount of potassium in their body.

By 1983, several scientists including myself had worked out the vital role of the Na-K-pump in cell function to the point where our understanding predicted that other dysfunctions, or disease states, of the body's cells would occur. Not until the mid 1990's did anyone bother to look for these other conditions. Since then, it has become well established that in the U.S., our typical diet with its low K/Na ratio is the cause of:

About 95% of the cases of high blood pressure.
At least 90% of strokes whether or not high blood pressure is involved.
Much of the osteoporosis and kidney stones.
An increased likelihood of h-pylori infection with resulting stomach ulcer and stomach cancer.
An increase in the severity of asthma.
An increased likelihood of mental decline with aging.

In addition, there is some evidence that this low K/Na ratio in the American diet contributes to insulin resistance, to obesity, and to adult diabetes

So since in autism a critical part of the brain is already damaged and has been shown to be subject to oxidative attack and neuroinflammation, it is not surprising that it is particularly susceptible to further interference.  As a result whereas, in a typical childlike Ted, aged 13 with an aversion to fruit and vegetables, can function perfectly well and additional potassium made no measurable difference to his sensory behaviour, the same was not true of Monty, aged 10 with ASD, and with a diet full of fruits and vegetables.  The additional potassium actually changed his sensory behavoiur.  Now I have a plausible explanation

Electrolytes etc.

 If you really want to go into the biology and understand intercellular/extra cellular fluid, role of hormones vasopressin and aldosterone and all about sodium and potassium balance, then take 10 minutes to carefully read the following link:-

Autism and Heart Disease, Diabetes and Cancer

On this blog I have already shown that several strategies for cardiac health also help autism.  Since in autism there is proven high cholesterol and high neuroinflammation and most likely also hypertension, it would make great sense reduce these risks regardless of the fact that those steps may likely also reduce autistic behaviours and improve functioning.

I made a study into omega 3 and conclude “eat fish”, it is cheaper than omega 3 oil, and it definitely will help cardiac health, but probably will do little to nothing for the autism.

A high potassium diet, particularly if it is based on food rather than supplements, will protect your child from heart disease later in life.  He/she is already in an at risk group.

 So there are two very good reasons increase potassium and reduce sodium in his/her diet.  If you are not aware of the health issues surrounding autism, take a look at this:-

Without intervention, adults with autism spectrum disorder appear to be at significant risk for developing diabetes, coronary heart disease, and cancer by midlife.

For a general discussion on these and other health issues, there is a well-researched paper called:


In the case of Monty, aged 10 with ASD, incremental potassium in diet and via an over the counter potassium supplement (that also contains magnesium and B vitamins) has a positive effect on autistic behaviours.  The total daily potassium ingested (1g as supplement, plus banana, orange juice, potato etc.) is still probably below the adult RDA of 3.5g, but much higher than most 10 year olds with ASD.

There is a scientific logic to show why potassium might produce beneficial effects due to better functioning of the sodium-potassium pump, particularly in the Purkinje cell layer of the brain, which is a known to be damaged in ASD.

It may also be that the magnesium, that is also present in the potassium supplement, is having a beneficial effect.  This could easily be investigated by some further research, should anyone be so inclined.

In any case, a relatively high potassium diet is well established to be very healthy and, along with strategies to lower cholesterol, will promote a healthy heart.  The literature shows that autistic people have elevated cardiac risk and so already have a good reason to be following this kind of diet;  I have just added another good reason.