Wednesday 15 June 2016

Treating KCC2 Down-Regulation in Autism, Rett/Down Syndromes, Epilepsy and Neuronal Trauma ?

In this composite image, a human nerve cell derived from a patient with Rett syndrome shows significantly decreased levels of KCC2 compared to a control cell.  This will be equally true of about 50% people with classic autism, people with Down syndrome, many with TBI and many with epilepsy

In a recent post I highlighted an idea from the epilepsy research to treat a common phenomenon also found in much classic autism.  Neurons are in an immature state with too much intracellular chloride, the transporter that brings it in, called NKCC1, is over-expressed and the one that takes it out, KCC2, is under-expressed.  The net result is high levels of intracellular chloride and this leaves the brain in an over-excited state (GABA working in reverse) reducing cognitive function and with a reduced threshold to seizures.

The epilepsy research noted that increased BDNF is one factor that down regulates KCC2, which would have taken chloride out of the cells.  So it was suggested to block BDNF, or something closely related called trkB.

Unfortunately there is no easy way to this.  But I did some more digging and found various other ways to upregulate KCC2.

There is indeed a clever safe way that may achieve this and it is a therapy that I have already suggested for other reasons, intranasal insulin.

BDNF is a neurotrophin and other neurothrophins also have the ability to regulate KCC2. IGF-1 is another such neurotrophin and we even have very recent experimental data showing its effect on KCC2.

Regular readers will know that several trials with IGF-1, or analogs thereof, are underway.

I actually am rather biased against IGF-1 as a therapy, since in my son’s case the level of IGF-1 in blood is already high.  So I do not want to inject him with IGF-1 or even give him an oral analog.

However by using intranasal insulin the effect would be just within the CNS and insulin binds at the same receptors as IGF-1. So if IGF-1 upregulates KCC2 so will insulin.

We know from extensive existing trial data and direct feedback from one researcher that intranasal insulin is well tolerated and has no effect outside the CNS.

So rather to my surprise there seems to be a safe, cheap way to treat KCC2 down-regulation and this would also be applicable in epilepsy, traumatic brain injury (TBI) and any other condition involving immature neurons or neuronal trauma. 

The Science

There is a very thorough recent review paper that looks at all the ways that KCC2 expression is regulated.

The epilepsy researchers consider trkB, top left in the figure below.  But just next to it is IGFR which can be activated by both insulin and IGF-1.

In Rett syndrome they are already using IGF-1 to modulate KCC2.  The research is done at Penn State.

As you can see in the figure the mechanism for IGF-1 and insulin is not the same as BNDF/trkb, but Penn State have already shown that IGF-1 works in vitro.

We saw in early posts regarding intranasal insulin that this was a safe way to deliver insulin to the brain without effects in the rest of the body.

So we know it is safe and in theory it should achieve the same thing that the Penn State researchers are trying to achieve.

Signaling pathways controlling KCC2 function. The regulation of KCC2 activity is mediated by many proteins including kinases and phosphatases. It affects either the steady state protein expression at the plasma membrane or the KCC2 protein recycling. All the different pathways are explained and discussed in the main text. The schematic drawings of KCC2 as well as other membrane molecules do not reflect their oligomeric structure. GRFα2, GDNF family receptor α2; BDNF, Brain-derived neurotrophic factor; TrKB, Tropomyosin receptor kinase B; Insulin, Insulin-like growth factor 1 (IGF-1); IGFR, Insulin-like growth factor 1 receptor; mGluR1, Group I metabotropic glutamate receptor; 5-HT-2A, 5-hydroxytryptamine (5-HT) type 2A receptor; mAChR, Muscarinic acetylcholine receptor; NMDAR, N-methyl-D-aspartate receptor; mZnR, Metabotropic zinc-sensing receptor (mZnR); GPR39, G-protein-coupled receptor (GPR39); ERK-1,2, Extracellular signal-regulated kinases 1, 2; PKC, Protein kinase C; Src-TK, cytosolic Scr tyrosine kinase; WNKs1–4, with-no-lysine [K] kinase 1–4; SPAK, Ste20p-related proline/alanine-rich kinase; OSR1, oxidative stress-responsive kinase -1; Tph, Tyrosine phosphatase; PP1, protein phosphatase 1; Egr4, Early growth response transcription factor 4; USF 1/2, Upstream stimulating factor 1, 2.

The Penn State research on using IGF-1 to increase KCC2 in Rett Syndrome

The researchers also showed that treating diseased nerve cells with insulin-like growth factor 1 (IGF1) elevated the level of KCC2 and corrected the function of the GABA neurotransmitter. IGF1 is a molecule that has been shown to alleviate symptoms in a mouse model of Rett Syndrome and is the subject of an ongoing phase-2 clinical trial for the treatment of the disease in humans.
"The finding that IGF1 can rescue the impaired KCC2 level in Rett neurons is important not only because it provides an explanation for the action of IGF1," said Xin Tang, a graduate student in Chen's Lab and the first-listed author of the paper, "but also because it opens the possibility of finding more small molecules that can act on KCC2 to treat Rett syndrome and other autism spectrum disorders."

More Melatonin?

As Agnieszka pointed out in the previous post it appears that extremely high doses of melatonin can increase KCC2 in traumatic brain injury (TBI). In this example BDNF was increased by the therapy, so I think TBI may be a specific case.  In most autism BDNF starts out elevated and in epilepsy, seizures are known to increase BDNF and that process is seen as down regulating KCC2 expression.  So in much autism and epilepsy you want less BDNF.

Melatonin attenuates neuronal apoptosis through up-regulation of K+ -Cl- cotransporter KCC2 expression following traumatic brain injury in rats

Compared with the vehicle group, melatonin treatment altered the down-regulation of KCC2 expression in both mRNA and protein levels after TBI. Also, melatonin treatment increased the protein levels of brain-derived neurotrophic factor (BDNF) and phosphorylated extracellular signal-regulated kinase (p-ERK). Simultaneously, melatonin administration ameliorated cortical neuronal apoptosis, reduced brain edema, and attenuated neurological deficits after TBI. In conclusion, our findings suggested that melatonin restores KCC2 expression, inhibits neuronal apoptosis and attenuates secondary brain injury after TBI, partially through activation of BDNF/ERK pathway.

More Science

There is plenty more science on this subject.

It is suggested that in addition to IGF-1/insulin it may be necessary to involve Protein tyrosine kinase (PTK).

Protein tyrosine kinase (PTK) phosphorylation is considered a key biochemical event in numerous cellular processes, including proliferation, growth, and differentiation, and has also been implicated in synaptogenesis. Protein tyrosine kinases are subdivided into the cytosolic nonreceptor family and the transmembrane growth factor receptor family, which includes receptors for insulin and insulin-like growth factor (IGF-1). The maturation of postsynaptic inhibition may require both a cytoplasmic PTK, which increases GABAA receptor-mediated currents, and insulin, which was shown to induce a rapid translocation of GABAA receptors from intracellular compartments to the plasma membrane. KCC2 is also known to have a C-terminal PTK consensus site. Therefore, the maturation of postsynaptic inhibition may, in addition to other mechanisms, also involve the effects of PTK and insulin acting on KCC2.


I would infer from all this science that intranasal insulin is likely to increase KCC2 expression in the brain, certainly worthy of investigation.

Protein tyrosine kinase (PTK) phosphorylation is considered a key biochemical event in numerous cellular processes.  This might be a limiting factor on the effectiveness of insulin in raising KCC2.  This would then add yet more complexity.

Protein kinases are enzymes that add a phosphate(PO4) group to a protein, and can modulate its function.  A protein kinase inhibitor is a type of enzyme inhibitor that blocks the action of one or more protein kinases.

Abnormal protein tyrosine kinases (PTKs) cause many human leukaemias, so there is research into PTK inhibitors (PTK-Is).

As we know from Abha Chauhan’s mammoth book, oxidative stress controls the activities of PTK.


  1. With respect to insulin signaling and the very giant elephant in the room nobody wants to discuss with respect to autism, a paper that more or less beats a dead horse came out in the last couple days which suggests that poor maternal line nutrition can negatively impact mitochrondia for up to 3 generations (at least in mice):

    Every time these obesity and poor lifestyle issues are critically brought up on discussion boards I have read, the first reply is "so you are gonna blame it all on the mom now". What this shows is that your great grandma could of been eating cookies and bacon like crazy and you could have genetically dysfunctional mitochondria because of her past behavior. The effect of course is stronger when it comes to the grandmother and mother, but this likely explains why women around the world, and especially in the United States where 70% of women are overweight, over 40% of women are classified as obese, and 10% are classified as morbidly obese. Dysfunctional mitochondria lead to all kinds of problems and of course might as well be considered a standard symptom of autism, but it also seems to lead to more obesity.

    Part of the rise in autism is obviously due to increased diagnosis and maybe even overdiagnosis, but the single biggest correlation with autism in the last 30 years and its rise is poor maternal health and the obesity that ensues.

    But even if you do everything right as a soon to be mother and don't happen to be obese, the sins of your own mother can negatively impact the mitochrondial health of any children you may have (that is if you don't get metabolic syndrome and PCOS, and are therefore unable to conceive a child).

    Nevertheless, prevention at this point won't help my children, but perhaps high levels of IGF-1 in those with autism is a reaction to decreased or diseased receptors, including the ones responsible for helping to regulate KCC2. Maybe this is one of many other reasons why ketogenic diets help some with autism and many with epilepsy as they restrict insulin levels in the body, thereby allowing the receptors that bind to insulin to upregulate over time as a homeostatic response. Maybe what is needed is a drug that binds to the appropriate receptor (while ignoring the corresponding autoreceptors) to normalize GABAergic plasticity, and also does not effect the rest of the body.

    Maybe trehalose would be a start as among many things it does, it upregulates insulin receptors:

  2. I have question for you, Peter, and perhaps Tyler can chime in: what's your understanding of IGF-1 administration and mtor? The little information I've been able to gather is difficult to interpret. I have low IGF-1, but have yet to try to raise it. However, I'm wary that increased IGF-1 would also lead to mtor increase, and that may not be a problem in Rett's, but most likely in most other forms of autism. I don't believe Buxbaum et al's study on increlex for idiopathic autism has been published yet. Best, Joel

    1. Joel, I do not claim to be an expert. All these pathways have multiple mechanisms involved and so you cannot really predict the result. Once you have the result, it is easy to explain it. If you have autism and low IGF-1 you have more scope to use IGF-1 as a therapy. However using IGF-1 and GH is known to produce growth effects in the body and there is no way to know what these will be and you may not find out what they were for 10-20 years. People used GH to keep weight low and have good muscle tone, 20 years on they need new hip and knee joints.

      I think intranasal insulin is very much safer and should give the same potential benefits.

  3. Peter, you say that high levels of BDNF is a major problem in some autism because this has been linked to epilepsy and as I read, in some OCD and PTSD.
    Decreased BDNF is also a problem connected with many serious central nervous system disorders, since it is needed for survival and protection.
    How can you measure BDNF to see if you need to upregulate or downregulate? I suppose it's an important biomarker of someone's condition and treatment.
    Do you think flavonoids can have an impact on BDNF? If so in what way?

  4. Petra, high BDNF may just be a characteristic of someone's autism. In this graphic

    autism was split into two groups; either hyper or hypo active pro growth signalling pathways.

    So if you have high BDNF it may not just be a case of how to lower it. If you have very low BDNF of NGF, it might indeed be wise to change its status, because it may be damaging.

    There are commercial blood tests to measure BDNF, but apparently they do give quite varying results. Should be good enough to see if you are very high or very low.

    Low BDNF can be improved by exercise and a good diet.

    If you have autism and high BDNF, you probably also have high IGF-1 as well.

    My conclusion was that there is no clever way to down-regulate BDNF, that is why I looked at insulin as a different way to upregulate KCC2.

    1. Well, with respect to drugs I am not sure about good ways to lower BDNF, especially since it causes different effects in different cells in the brain, sometimes even the opposite. What BDNF does in the hippocampus is different than what BDNF does in the cerebellum as well as the prefrontal cortex, and that is just discussing neurons (and there are so many types too).

      Nevertheless, I have read a lot about this with respect to something else I have been working on and I will say that BDNF seems to be activated via excitatory receptors and I know that various patterns of stimulation with rTMS (transcranial magnetic stimulation) are believed to effect excitatory or inhibitory receptors differently. A pattern of stimulation called iTBS is thought to downregulate GABA receptors (increasing LTP [long term potentiation] excitation which is important for learning), while a separate pattern of stimulation called cTBS is thought to wear out or disrupt excitatory receptors (NMDA specifically) causing inhibition or LTD (long term depression) like effects.

      There is also a protocol I have been very intrigued by lately in a recent paper that involved 6000 pulses at 20hz which is a lot. With interstimulus intervals it takes an hour. What this protocol has been shown to do is increase the cortical silent period (CSP) which is thought to be a measure of true inhibition via GABAb receptors in the central nervous system (the other protocols merely increase transient excitation or depression). It would take a long time to explain this all, but the point is if you block chronically block excitatory receptors or else wear them out (as is thought to happen in clinical depression from overuse), you will decrease BDNF indirectly.

      So for example in rTMS protocols for depression (the disease), they use high frequency stimulation protocols in the hope of upregulating BDNF, whereas in many rTMS autism studies they have pretty much all used low frequency stimulation (increases LTD) or else cTBS (which is similar). The stuff I have been working on is in this realm but may be far superior to what you can do with rTMS and other forms of artificial stimulation because it can target almost all areas of the brain (even the deep subcortical areas such as the basal ganglia and thalamus) and done so non-invasively.

  5. You should read this on MET gene mutation and its impact on tyrosine kinase efficacy:

    Quite fascinating

  6. Thought this was interesting and shows there maybe multiple approaches to KCC2 activation/upregulation:

    Activation of 5-HT2A receptors upregulates the function of the neuronal K-Cl cotransporter KCC2

    5ht2a activation is obviously caused by hallucogenics such as LSD and psilocybin, but also by agmatine. MDMA also activates 5ht2a.


Post a comment