Monday 11 January 2016

The GABA Switch, Altered GABAa Receptor subunit expression in Autism and Basmisanil

In today’s post I intended to dig a little deeper into the GABA switch, which appears to underlie much autism, schizophrenia, epilepsy, even Down Syndrome and, not to forget, many mood disorders.  

Once you start digging, it is rather hard to stop.

There is literature on the subject, but very little (almost none, really) looks at the big picture of what is going on.  It is the big picture that matters.

The GABA switch(es); but how many are there?

The post starts out relatively simple, but then it does get complicated, because I discovered a lot interesting avenues exist, that seem to have been completed ignored by autism research.  It seems Down Syndrome researchers are better informed.  

So if you make it to the end of this post, you will have done well.

It seems that there are tens, if not hundreds, of possible ways to repair the faulty GABA switches.  It would very much become a case for personalized medicine, correcting the precise dysfunctions, without disturbing anything else.  Each case will be slightly different.

Back to the simple part.

During very early development certain changes in the brain are expected to occur, that control how the GABA neurotransmitter functions.  If they do not all occur, some of the following may occur:-

·        Autism
·        Epilepsy
·        Mental retardation / Intellectual Disability; including the MR/ID in Down Syndrome
·        Mood disorders (anxiety, depression etc)

If later in life, after brain maturation, these same changes occur the following may occur:-

·        Schizophrenia
·        Bipolar
·        Other mood disorders (anxiety, depression etc)
·        Epilepsy

There are at least two distinct processes involved, both can be considered as part of the GABA switch. It is likely that two further process are involved, but they have not yet been adequately researched.

1.     The lowering of intracellular chloride levels

This has been very deeply documented already in this blog.  If the GABA switch has not been “flipped”, we have overexpression of a cotransporter NKCC1, which overwhelms the effect of another called KCC2 and this results in elevated intracellular chloride levels.  This then prevents GABA signaling switching from excitatory to inhibitory.  This creates an excitatory/inhibitory imbalance and neurons fire when they should not.  This disables cognitive function and creates a tendency towards pre-epilepsy and then epilepsy.

The neurons never reach their expected mature state.

2.     Change in GABAA-receptor subunit expression

We have already seen in this blog, that drugs that positively modulate the α2 and α3 subunits of GABAA receptors, like low-dose clonazepam, rescue some aspects of autism.  This is Professor Catterall’s research.  As far as I can see, he did not really explain the big picture behind this.

We saw in early posts that the composition of the GABA receptor vary over time. Remember there are 5 sub-units in each receptor.  These changes in composition of the receptor directly affect people’s mood and behavior.

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 (α1β2γ2)
In humans, the units are as follows:
·         six types of α subunits (GABRA1, GABRA2, GABRA3, GABRA4, GABRA5, GABRA6)
·         three βs (GABRB1, GABRB2, GABRB3)
·         three γs (GABRG1, GABRG2, GABRG3)
·         as well as a δ (GABRD), an ε (GABRE), a π (GABRP), and a θ (GABRQ)

What appears likely to me is that this variation in subunit expression both over time and throughout different parts of the brain is really just another part of the GABA switch.

It appears that in autism the α3 subunit is under-expressed.  In its place it appears we have the α5 subunit over-expressed.  This lowers your IQ.

This under-expression of the α3 subunit has been compensated for by Catterall by using a positive allosteric modulator.

In effect the α5 subunit is dominating, where it should not be and this presents itself as autism.

There is research showing how numerous influences can affect the sub unit expression.  Some are transitory and reversible, but others can be permanent.  It is proposed that, when permanent, the underlying change is epigenetic, when “tags” are placed on perfectly functional genes that either turn them on or off.  One example resulting in permanent change is stress. 

Repeated neonatal handling with maternal separation permanently alters hippocampal GABAA receptors and behavioral stress responses

We will focus on the reversible changes, since that is the purpose of this blog.

After some digging, I did find a nice graphic that does illustrate the first two elements of the GABA switch.  It is based on mouse research and shows the changes that should happen in the first 20 days of life. 

In the first few days NKCC1 is highly expressed, while KCC2 is weakly expressed. This results high levels of intracellular chloride.  As NKCC1 drops, less chloride enters and so intracellular chloride falls.  So depolarizing GABA (excitatory) becomes hyperpolarizing GABA (inhibitory).

In the twenty days while this is happening, the sub unit structure of the GABAA receptors is also changing.

In some autism neurons remain in their immature state, with NKCC1 highly expressed and so high levels of intracellular chloride and depolarizing GABA (excitatory).

My suggestion is that the programmed changes in sub unit expression may also fail to occur.

3.     Density of GABA A receptors ?

In the literature, when they talk about density of GABA A receptors, they are talking about either an increased or reduce number of receptors at a given location.

The density can vary over time and within different parts of the brain.


It appears that in autism, there is a reduced number of GABAA receptor, in other words lower density.

GABAA receptor downregulation in brains of subjects with autism

I am suggesting that the under-expression of GABAA receptor, which appears likely to be linked to disturbed calcium channel signaling, could be considered as the third element of the GABA switch.

In the science jargonVGCC (Voltage Gated Calcium Channel) activation is involved in GABA-induced GABAAR down-regulation”. Which begs the question, what is the effect on GABAA receptor density in humans of blocking VGCCs?

Nifedipine is specifically suggested.  My Polypill already includes verapamil, another blocker of VGCCs.

4.     GABAB subunits ?

There probably is a fourth element of the GABA switch.  If there is, it possibly relates to the GABAB receptor.  GABAB receptors are made up of just two subunits, GABAB1 and  GABAB2. This is much less well researched than GABAA, but from what little there is, it is clear that GABAB1 and GABAB2 receptor subunits expression is disturbed in some epilepsy and after TBI (traumatic brain injury).  TBI is interesting because, in many ways, that is what autism is; just it was not a physical trauma, it was a genetic/environmental trauma.

The expression of GABA(B1) and GABA(B2) receptor subunits in the cNS differs from that in peripheral tissues.

Modification of GABA(B1) and GABA(B2) receptor subunits in the somatosensory cerebral cortex and thalamus of rats with absence seizures (GAERS)

The Implication?  Repair the GABA Switch(es)

Before getting involved in the complexities of the research, we can already draw a nice simple conclusion.

Many types of autism are likely associated with a faulty GABA switch.  So if you want to treat someone’s autism, start by repairing the GABA switch(es).  Just realize there is more than one and so therapies will have to vary.

Now to the science, for those who like to go into details:-
This is a good paper that was highlighted earlier by Tyler, a reader of this blog:-

The GABAergic neurons of the thalamic reticular nucleus (nRt) provide the primary source of inhibition within the thalamus. Using physiology, pharmacology and immunohistochemistry in mice we characterized post-synaptic developmental changes in these inhibitory projection neurons. First, at postnatal day 3-5 (P3-5), inhibitory postsynaptic currents (IPSCs) decayed very slowly, followed by a biphasic developmental progression, becoming faster at P6-8, then slower again at P9-11 before stabilizing in a mature form around P12. Second, the pharmacological profile of GABAAR mediated IPSCs differed between neonatal and mature nRt neurons and this was accompanied by reciprocal changes in α3 (late) and α5 (early) subunit expression in nRt. Zolpidem, selective for α1- and α3-containing GABAARs, augmented only mature IPSCs, while clonazepam enhanced IPSCs at all stages. This effect was blocked by the α5-specific inverse agonist L-655,708 but only in immature neurons. In α3H126R mice in which α3 subunits were mutated to become BZ insensitive, IPSCs were enhanced compared to wild type animals in early development. Third, tonic GABAAR activation in nRt is age-dependent, and more prominent in immature neurons, which correlates with early expression of α5 containing GABAARs. Thus neonatal nRt neurons show relatively high expression of α5 subunits which contributes to both slow synaptic and tonic extrasynaptic inhibition. The postnatal switch in GABAAR subunits from α5 to α3 could facilitate spontaneous network activity in nRt that occurs at this developmental time point and which is proposed to play a role in early circuit development.

The following paper really covers the first GABA switch very well:-

The switch from excitatory to inhibitory GABAAR-related effects is closely related to the lowering of [Cl−]i during the course of the development. This latter mainly relies on the differential ontogenic expression of the Na+/K+/2Cl− cotransporter isoform 1 (NKCC1), which uptakes chloride ions [76–78], and the neuronal K+/Cl−cotransporter type 2 (KCC2) [79], which extrudes chloride ions [49, 80]. However, other exchangers can control the chloride gradient as the anion (Cl−–HCO3 −) exchangers, either Na+- independent (AE) or Na+-driven (NDCBE also called NDAE) [81] (NCBE) [82]. AE mediates influx of Cl− while exporting HCO3 −, these exchanges being triggered by intracellular alkalinisation. NDCBE, known as an acid extruder (extrudes H+), moves Cl− out in exchange of HCO3 −, driven by the Na+ gradient [83, 84]. NCBE also lowers [Cl−]i (and [H+]i) while importing Na+ and HCO3 − [82, 85].

This is one of Professor Catterall’s papers we looked at previously

Moreover, autistic-like behavioral impairments can be treated effectively in both BTBR and Scn1a+/− mice by enhancement of inhibitory neurotransmission with low doses of subunit-selective positive allosteric modulators of GABAA receptors containing α2 and/or α3 subunits. Together, our results support the hypothesis that reduced GABAergic inhibitory neurotransmission contributes to autism-associated behavioral and cognitive deficits and suggest that enhancement of GABAergic neurotransmission with next-generation subunit-specific pharmacological agents may be beneficial.

Subsynaptic GABAAreceptor subtypes are composed of two α, two β, and one γ subunit (Fritschy and Mohler, 1995). The action of GABA at these ionotropic receptors is increased through positive allosteric modulation by benzodiazepines, which are used to treat anxiety, insomnia, and epilepsy (Rudolph and Knoflach, 2011). In order to determine whether treatment with a benzodiazepine reverses the constitutively decreased GABAergic inhibitory signaling, we treated C57BL/6J and BTBR hippocampal slices with 0.5 μM clonazepam, a broad-acting, traditional benzodiazepine. These recordings revealed increased spontaneous IPSC amplitude (Figures 1E and 1F) and frequency (Figure S1C) in BTBR slices. In contrast, a significant increase of spontaneous IPSC amplitude (Figure S1I), but no change in IPSC frequency (Figure S1J), was observed in C57BL/6J slices. The increased GABAergic signaling after treatment with clonazepam led to a decrease in frequency of spontaneous EPSCs (Figures 1G and 1H), without change in amplitude in BTBR hippocampal slices (Figure S1D). Interestingly, the frequency of spontaneous EPSC was also decreased by clonazepam (Figure S1K), without change in amplitude (Figure S1L) in C57BL/6J slices. These data support the idea that low-dose clonazepam can reverse the underlying deficit in spontaneous GABAergic inhibitory neurotransmission in BTBR mice.

Rescue by α23-Specific Positive Allosteric Modulators of GABAA Receptors

Diversity of GABA receptor function is conferred by more than 20 different subunits, and receptors with different α subunits play distinct roles in the physiological and pharmacological actions of GABA and benzodiazepines (Fritschy and Mohler, 1995, Harmar et al., 2009, Rudolph and Knoflach, 2011,Rudolph and Möhler, 2004, Smith and Olsen, 1995). We tested the effects of subunit-selective positive allosteric modulators of GABAA receptors on social behavior in BTBR mice and C57BL/6J mice. A low dose of the α2,3-subunit-selective positive allosteric modulator L-838,417 (Löw et al., 2000, Mathiasen et al., 2008) increased social interactions in BTBR mice, with maximal effective dose of 0.05 mg/kg, and the beneficial effect was lost when the dose increased (Figures 4I and S4E). In contrast, L-838,417 did not change the social interaction behavior of C57BL/6J mice (Figure S4I). Moreover, the α1-subunit-selective positive GABAA modulator zolpidem (Mathiasen et al., 2008, Sieghart, 1995) failed to show beneficial effects in BTBR mice and actually aggravated their social interaction deficit at high doses (Figures 4J and S4F). Interestingly, a high dose of zolpidem also impaired social behavior in C57BL/6J mice (Figure S4J). Total movement tended to increase at high doses of L-838,417 (Figure S4G; not significant) but significantly decreased at 0.5 mg/kg zolpidem (Figure S4H). 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 activation of GABAA receptors with α1 subunits reducing social interaction, respectively.
Subunit-selective GABAA receptor modulators may also have an important effect on cognitive behaviors. In the context-dependent fear conditioning test, treatment with 0.05 mg/kg L-838,417 improved short-term (30 min) and long-term (24 hr) spatial memory in BTBR mice (Figure 4K), whereas 0.05 mg/kg zolpidem enhanced short-term memory but not long-term memory (Figure 4L). These data show that α2,3-subunit-containing GABAA receptors may also be important for cognitive behaviors in BTBR mice. The bell-shaped dose-response curves observed for both L-838,417 and clonazepam may explain why high-dose benzodiazepine treatment for prevention of anxiety and seizures has not been reported to improve autistic traits in ASD patients. As illustrated in Figures 4N and 4O, treatment with low doses of L-838,417 also improves social interactions in the Scn1a+/− mice, a model of Dravet syndrome with severe autistic-like behaviors (Han et al., 2012), within a narrow dose range. In contrast, similar treatment with zolpidem is not effective. Altogether, these experiments show that treatment with an α2,3-selective positive allosteric modulator of GABAA receptors is sufficient to rescue autistic-like behaviors and cognitive deficit in both a monogenic model of autism-spectrum disorder and the BTBR mouse model of idiopathic autism.

The following paper highlights disrupted subunit expression in one type of epilepsy:-

Altered thalamic GABAA-receptor subunit expression in the stargazer mouse model of absence epilepsy.



Absence seizures, also known as petit mal seizures, arise from disruptions within the cortico-thalamocortical network. Interconnected circuits within the thalamus consisting of inhibitory neurons of the reticular thalamic nucleus (RTN) and excitatory relay neurons of the ventral posterior (VP) complex, generate normal intrathalamic oscillatory activity. The degree of synchrony in this network determines whether normal (spindle) or pathologic (spike wave) oscillations occur; however, the cellular and molecular mechanisms underlying absence seizures are complex and multifactorial and currently are not fully understood. Recent experimental evidence from rodent models suggests that regional alterations in γ-aminobutyric acid (GABA)ergic inhibition may underlie hypersynchronous oscillations featured in absence seizures. The aim of the current study was to investigate whether region-specific differences in GABAA receptor (GABAAR) subunit expression occur in the VP and RTN thalamic regions in the stargazer mouse model of absence epilepsy where the primary deficit is in α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor (AMPAR) expression.


Immunofluorescence confocal microscopy and semiquantitative Western blot analysis were used to investigate region-specific changes in GABAAR subunits in the thalamus of the stargazer mouse model of absence epilepsy to determine whether changes in GABAergic inhibition could contribute to the mechanisms underlying seizures in this model of absence epilepsy.


Immunofluorescence confocal microscopy revealed that GABAAR α1 and β2 subunits are predominantly expressed in the VP, whereas α3 and β3 subunits are localized primarily in the RTN. Semiquantitative Western blot analysis of VP and RTN samples from epileptic stargazers and their nonepileptic littermates showed that GABAAR α1 and β2 subunit expression levels in the VP were significantly increased (α1: 33%, β2: 96%) in epileptic stargazers, whereas α3 and β3 subunits in the RTN were unchanged in the epileptic mice compared to nonepileptic control littermates.


These findings suggest that region-specific differences in GABAAR subunits in the thalamus of epileptic mice, specifically up-regulation of GABAARs in the thalamic relay neurons of the VP, may contribute to generation of hypersynchronous thalamocortical activity in absence seizures. Understanding region-specific differences in GABAAR subunit expression could help elucidate some of the cellular and molecular mechanisms underlying absence seizures and thereby identify targets by which drugs can modulate the frequency and severity of epileptic seizures. Ultimately, this information could be crucial for the development of more specific and effective therapeutic drugs for treatment of this form of epilepsy

This is another good paper, this time looking at how fragile X mental retardation protein (FMRP) may disrupt sub-unit expression and how this appears not only in those with Fragile-X but also in schizophrenia, mood disorders, and autism.  mGluR5 is involved not surprisingly and you may recall that mGluR5 is surprisingly also involved in GERD/GORD/reflux, which affects many people with autism.

GABA receptor subunit distribution and FMRP-mGluR5 signaling abnormalities in the cerebellum of subjects with schizophrenia, mood disorders, and autism.

Gamma-aminobutyric acid (GABA) is the main inhibitory neurotransmitter in the brain. GABAergic receptor abnormalities have been documented in several major psychiatric disorders including schizophrenia, mood disorders, and autism. Abnormal expression of mRNA and protein for multiple GABA receptors has also been observed in multiple brain regions leading to alterations in the balance between excitatory/inhibitory signaling in the brain with potential profound consequences for normal cognition and maintenance of mood and perception. Altered expression of GABAA receptor subunits has been documented in fragile X mental retardation 1 (FMR1) knockout mice, suggesting that loss of its protein product, fragile X mental retardation protein (FMRP), impacts GABAA subunit expression. Recent postmortem studies from our laboratory have shown reduced expression of FMRP in the brains of subjects with schizophrenia, bipolar disorder, major depression, and autism. FMRP acts as a translational repressor and, under normal conditions, inhibits metabotropic glutamate receptor 5 (mGluR5)-mediated signaling. In fragile X syndrome (FXS), the absence of FMRP is hypothesized to lead to unregulated mGluR5 signaling, ultimately resulting in the behavioral and intellectual impairments associated with this disorder. Our laboratory has identified changes in mGluR5 expression in autism, schizophrenia, and mood disorders. In the current review article, we discuss our postmortem data on GABA receptors, FMRP, and mGluR5 levels and compare our results with other laboratories. Finally, we discuss the interactions between these molecules and the potential for new therapeutic interventions that target these interconnected signaling systems.

The logical next step - Regulation of GABAA Receptor Subunit Expression by Pharmacological Agents

The first thing to note is that, if you are using low dose clonazepam, you are already compensating for the lack of α3 subunits. Clonazepam is a so-called positive allosteric modulator of α3.

Are there further options?  Yes there are very many, but are they safe for long term use?  Most fail this test.

For example, ethanol will down-regulate α5 expression and it appears possible that the right benzodiazepine at the right dose might also achieve this, but much mouse research is contradictory.  We are very lucky that the clonazepam dose is so tiny, it appears to have no side effects whatsoever.  At conventional doses, benzodiazepines do have problems.

Initially I looked at research that sets out very generally to look at what I am interested in.  This yielded some interesting studies, some of which we looked at already when looking at Professor Catterall and Clonazepam.

Then I decided to look at very specific things I want to modulate and I found a great deal more.  It looks like we will be able to borrow a drug being developed for Down Syndrome (Basmisanil / RG-1662).

Numerous substances do affect GABAA receptor subunit expression, but the ones available today are generally non-specific.  They will change things, but will it be for the better?

So there are two approaches:-

·        Roll the dice

Some simple substances are known to affect GABAA receptor subunit expression and some of these substances have already been associated with autism.  It is conceivable that a lit bit, more or less, of one of this might just stir things up so that the end result might be better.

·        Clever approach

The clever approach would make sure changes only affect the specific subunits that need to be modulated.  The chance of success is then very high.  The problem is that this requires waiting for a new drug, currently in phase 2 trials, to complete its approval process 

Rolling the Dice

In the following papers there numerous ideas and some of these that have already appeared in previous posts.  Those ideas include:-

·        Calcium channel blockers, like Nifedipine

They may increase GABA receptor density.

·        Zinc

We saw in research from Taiwan, that it appears there is a problem with zinc in autism and schizophrenia.  It is not a lack of zinc, rather it is in the “wrong” place.  They have a drug, Clioquinol, that can move it to the “right” place, but this drug is not regarded as safe in many countries.

Altered Homeostasis in Autism: Cl-, K+, Ca2+, and quite possibly Zn2+

It claimed that immature neurons are more sensitive to zinc than adult neurons.  Even in adults, autistic neurons remain immature.

·        BDNF

We have already seen that growth factors, including BDNF are disrupted in autism.  Some people with autism have too much BDNF and some too little.
BNDF seems to affect GABA receptor density.

·        Progesterone/Pregnenolone

We know that transdermal Progesterone/Pregnenolone helps many people with autism + anxiety.  We know, from Hardan at Stanford, that high dose of oral Pregnenolone seem to help adults with autism or schizophrenia.

We saw in earlier posts that allopregnanolone possesses biphasic, U-shaped actions at the GABAA receptor, meaning that a tiny dose can have the same effect as a large dose. Giving large doses of a female hormone to young boys does not seem a clever idea, presumably this is why the Stanford trial was on adults.

I did previously suggest small doses of oral Pregnenolone might be worthy of a clinical trial. 

Progesterone/Pregnenolone will change GABAA sub unit expression.

·        We have another mushroom-derived substance called muscimol

“The GABAAergic agonist muscimol increases KCC2 mRNA in male neurons, via activation of voltage sensitive calcium channels and calcium signaling [79, 80]; in contrast, muscimol decreases KCC2 mRNA in female SNR neurons”

So mushrooms just for the boys.

Muscimol is a selective agonist for the GABAA receptors that seems to do some clever things.

For those of you who like “natural” substance, here is a paper for you:-

It does actually mention something very interesting.  We know that ethanol would down regulate α5, but clearly you cannot “treat” a child with alcohol.  It appears that the fragrant components of whiskey, wine, sake, brandy may have the same effect as ethanol, but require only tiny concentrations.  So there may be scope for alcohol as a therapy after all.  Note that propolis is usually sold as a solution in alcohol; this widely given to children.

“Volatile components of alcoholic drinks, such as whiskey, wine, sake, brandy, and shochu potentiate GABA responses to varying degrees (Hossain et al., 2002a). Although these fragrant components are present in alcoholic drinks at low concentrations (extremely small quantities compared with ethanol), they may also modulate the mood or consciousness through the potentiation of GABAA responses after absorption into the brain, because these hydrophobic fragrant compounds are easily absorbed into the brain through the blood–brain barrier and are several thousand times as potent as ethanol in the potentiation of GABAA receptor‐mediated responses (Hossain et al., 2002a).”

The following paper is very extensive, but completely omits some extremely important possibilities I later came across.  It should be a must read for those interested in the science.

Regulation of GABAA Receptor Subunit Expression by Pharmacological Agents

F. Mechanisms Regulating GABAA Receptor Subunit Expression

The large number of GABAA receptor genes and the various types of neurons and glial cells in the brain with different patterns of subunit expression suggest a complex system regulating their transcription

Activity-dependent signaling pathways modulate the function of both transcriptional activators and repressors (West et al., 2002). Calcium is a crucial second messenger in the transduction of synaptic activity into gene expression (Carafoli et al., 2001), and it is involved in the mechanisms of GABAA receptor up- and down-regulation
 It was recently shown that the activation of protein kinase C in primary rat neocortical cultures increases transcription of α1 mRNA via phosphorylation of CREB that is bound to the GABRA1 promoter (Hu et al., 2008). In contrast, activation of protein kinase A (PKA) represses α1 mRNA transcription via inducible cAMP early repressor (ICER) that forms inactive heterodimers with CREB (Hu et al., 2008). Brain-derived neurotrophic factor (BDNF) decreases α1 transcription via activation of the Janus kinase/signal transducer and activator of transcription (STAT) pathway (Lund et al., 2008). BDNF-dependent phosphorylation of STAT3 induces the synthesis of ICER that binds with phosphorylated CREB at the GABRA1 promoter CRE site, thereby repressing transcription (Lund et al., 2008).

In cultured CGCs, BDNF induces α6 mRNA expression and enhances the expression of α1 and γ2 mRNA (Bulleit and Hsieh, 2000). These enhancements are mediated via mitogen-activated protein kinase pathway (Bulleit and Hsieh, 2000). In contrast, in cultured hippocampal pyramidal cells, BDNF reduced cell surface expression of α2, β2/3, and γ2 subunits (Brünig et al., 2001). The results suggest that BDNF affects GABAAreceptor expression in a brain region- and cell-specific manner.

Regulation of GABAA Receptor Expression By Pharmacological Agents

A. Benzodiazepines

Read the full paper!

B. Neurosteroids

5. α5 Subunit.

The expression of α5 mRNA was down-regulated by CE in the cerebral cortex (Mhatre and Ticku, 1992), whereas no effect was found by Devaud et al. (1995)(Table 14). CE did not affect cerebral cortical α5 polypeptide expression (Charlton et al., 1997). Withdrawal from CE down-regulated cortical α5 mRNA expression (Mhatre and Ticku, 1992). CE down-regulated α5 polypeptide expression in the cerebellum (Charlton et al., 1997). In the hippocampus, CE up-regulated α5 mRNA expression, although it had no effect on α5 polypeptide expression (Charlton et al., 1997). Withdrawal from CE did not affect hippocampal α5 mRNA expression (Mahmoudi et al., 1997; Petrie et al., 2001). Long-term ethanol treatment or withdrawal from it did not affect α5 mRNA expression in cultured rat hippocampal neurons (Sanna et al., 2003). The results of studies on the CE effect on α5 subunit suggest brain region-specific modulation of the expression.

Here we look at the down/up regulation of the number of receptors, rather than their substructure.  This was shown to vary in nice chart with green spots earlier in this post. In other words this about modulating receptor density.

Changes in GABA receptor (GABAAR) gene expression are detected in animal models of epilepsy, anxiety and in post-mortem schizophrenic brain, suggesting a role for GABAAR regulation in neurological disorders. Persistent (48 h) exposure of brain neurons in culture to GABA results in down-regulation of GABAAR number and uncoupling of GABA and benzodiazepine (BZD) binding sites. Given the central role of GABAARs in fast inhibitory synaptic transmission, GABAAR down-regulation and uncoupling are potentially important mechanisms of regulating neuronal excitability, yet the molecular mechanisms remain unknown. In this report we show that treatment of brain neurons in culture with tetrodotoxin, glutamate receptor antagonists, or depolarization with 25 mm K+ fails to alter GABAAR number or coupling. Changes in neuronal activity or membrane potential are therefore not sufficient to induce either GABAAR down-regulation or uncoupling. Nifedipine, a voltage-gated Ca2+ channel (VGCC) blocker, inhibits both GABA-induced increases in [Ca2+]i and GABAAR down-regulation, suggesting that VGCC activation is required for GABAAR down-regulation. Depolarization with 25 mm K+ produces a sustained increase in intracellular [Ca2+] without causing GABAAR down-regulation, suggesting that activation of VGCCs is not sufficient to produce GABAAR down-regulation. In contrast to GABAAR down-regulation, nifedipine and 25 mm K+ fail to inhibit GABA-induced uncoupling, demonstrating that GABA-induced GABAAR down-regulation and uncoupling are mediated by independent molecular events. Therefore, GABAAR activation initiates at least two distinct signal transduction pathways, one of which involves elevation of intracellular [Ca2+] through VGCCs.

Given that calcium is a ubiquitous signaling molecule, it seems reasonable that increased Ca2+ alone is not sufficient to mediate the effects of signal transduction pathways initiated by activation of a specific receptor. Studies of hippocampal neurons demonstrate that increases in [Ca2+]i by NMDA receptors or VGCCs initiate distinct signal transduction pathways (Bading et al. 1993; Xia et al. 1996). The route of Ca2+influx appears to influence which signal transduction pathway is stimulated. Compartmentalization of molecules involved in second messenger pathways may also account for the observation that Ca2+ influx in dendrites initiates signal transduction cascades distinct from those triggered in the soma (Ghosh and Greenberg 1995). Recent evidence shows that increases in [Ca2+]i initiate different signaling mechanisms depending on whether the Ca2+ increase occurs in the cytoplasm or in the nucleus (Hardingham et al. 1997).
Our results demonstrate that GABA-induced GABAAR down-regulation and uncoupling are mediated by independent molecular events, indicating that GABAAR activation leads to initiation of at least two distinct signal transduction pathways. We present evidence that VGCC activation is involved in GABA-induced GABAAR down-regulation. Understanding molecular mechanisms of GABAAR down-regulation will clarify the role of GABA-induced changes in gene expression in both normal nervous system function and in neurological disease.

This is another paper on epilepsy.  It is a very good one again talks about zinc.  It also talks about using AE3 as well as NKCC1 for therapeutic intervention.

GABAA Receptors in Normal Development and Seizures: Friends or Foes?

Developmental Changes in GABAA Receptor Structure and Pharmacology

Most studies describing developmental changes in GABAAergic signaling have been done in rats. To better understand how might these reflect changes in humans, it is generally thought that brain development in a postnatal day 8-10 (PN8-10) rat is almost equivalent to a newborn human baby. The infantile stage in rats spans from PN7-21 and is followed by the juvenile stage. Puberty onset in rats occurs at approximately P32-37, whereas adulthood is reached at 2 months [230, 342, 343]. GABA is present in the embryonic neural system from the very early days [105, 162]. In the embryonic rat neocortex, GABA is detected diffusely as early as embryonic day 10 (E10) but after E14 its presence is limited to the subplate, cortical plate, marginal and intermediate zones [105]. In parallel, GABAA receptors are expressed, even before the establishment of GABAergic synapses, to permit the autocrine and paracrine actions of GABA on brain development [164,183, 278]. Regional differences in subunit expression have been reported in rats, with α4, β1, γ1 detected in the premigratory neuroblasts of the ventricular zone [164, 183] and α2, α3, β3, γ2 at the cortical or subcortical plate [164, 183, 190]. The spatiotemporal developmental patterns of GABA / GABAA receptor expression are thought to be important in the orchestration of the normal GABA-related regulation of proliferation and migration or neural and glial progenitors [105]. The high levels of GABA in the early stages of development promote the proliferation of ventricular zone progenitors [105], whereas the subsequent decline and restriction of GABAAergic influence within the outer neocortical layers inhibits proliferation [8,105, 177], enhances migration [20], and may therefore permit further neuronal differentiation. GABAAergic signaling is also important for neuronal survival at this stage [128]. In further support of the importance of GABAAergic signaling for brain development, in utero exposure to GABAA receptor inhibitors decreases the number of parvalbumin-immuno-reactive GABAergic neurons in the striatum, by impairing the survival or differentiation of these neurons [182]. Moreover, focal application of GABAAergic agonists in the cortex of newborn rats may induce abnormal migration and heterotopias [107].
Age-related, species, and region-specific changes, gradual or transient, continue through postnatal development, adulthood and ageing for GABAA receptor subunits like α1, α2, α3, α4, α5, γ1, γ2 [138, 171,214, 255, 260, 340]. Fritschy et al. have proposed that during the early postnatal life, a gradual parallel decrease in α2 / α3 and increase in α1 expression occurs in rat brain [74, 120] (Fig. 11). Similar developmental switch from α2 / α3 to α1 subunit predominance has been observed in mouse superior colliculus [111] and visual cortex [37, 109]. Functionally, the postnatal increase in α1 has been linked to increased sensitivity to neurosteroids [214], zolpidem [111] and benzodiazepines [140], and acquisition of mature type postsynaptic IPSCs with shorter duration [29]. The latter may be important for a brain that learns to respond appropriately to novel patterns of neuronal activation. Using α1 knockout mice, Bosman et al. have elegantly shown that lack of α1 subunits leads to preservation of juvenile, long duration IPSCs and impairs spatiotemporal excitation patterns to local high frequency stimulation in the visual cortex [28, 29]. In the dentate granule cells of the rat hippocampus, the developmental switch from α5 to α1, α4, and γ2 subunits correlates with decreasing sensitivity to zinc and increase in the affinity for benzodiazepines [34, 140]. Sensitivity to zinc is important in the functional regulation of GABAAergic transmission, particularly in immature neurons. Large amounts of zinc can be stored in synaptic vesicles of nerve terminals, as in the hippocampal mossy fibers of the immature hippocampus. Stimulation-dependent zinc release in this system may therefore be useful to keep under control the excessive depolarizing effects of GABA, in a subunit-specific pattern [16, 53, 166, 285,331]. This may be less important in adult neurons, which lose their sensitivity to zinc, as GABAA receptor mediated inhibition is more efficient.
There is though regional specificity of the evolution of these changes [56]. Sex differences in GABAAreceptor subunit expression further increase the diversity. These include increased expression of α1 subunit in the female substantia nigra of infantile and juvenile rats [255] and increased γ1 expression in the male rat juvenile medial preoptic area [219]. At the cellular level, GABAA receptor trafficking also evolves. Early in development and before synaptic integration occurs, receptor complexes can be diffusely expressed at the cell membrane and can be tonically activated in the presence of GABA [61, 172, 177, 236, 311]. As the establishment and differentiation of GABAergic synapses begins, they initially occupy both extrasynaptic and synaptic sites; finally targeting and clustering at synaptic sites and dendritic processes increases with maturation and spontaneous IPSCs can be detected [1, 236, 253].
The temporal, regional, sex, and species specific variability in the expression of these subunits in the brain emphasizes that generalization across brain regions, species, genders, and ages is not possible, but one needs to specifically study each structure, age, and condition independently. To further complicate these studies, handling, caloric restriction, and even swim stress regulate GABAA receptor subunit expression, at times with a lasting effect, suggesting that epigenetic influences may be as important in shaping the GABAAreceptor related differentiation and communication patterns [122,170, 202, 238].

Repeated neonatal handling with maternal separation permanently alters hippocampal GABAA receptors and behavioral stress responses

Differential regulation of KCC2 in neurons with depolarizing or hyperpolarizing GABAAergic signaling.GABAA receptor activation and BDNF increase KCC2 in immature neurons with depolarizing GABAA ergic responses, but decrease it in neurons with hyperpolarizing ...

In addition, the intracellular concentrations of Cl-and HCO3- are regulated by anion exchangers (AE). The sodium independent electroneutral AEs exchange HCO3- for extracellular Cl-, lowering intracellular pH and increasing Cl- [112, 300, 336]. Sodium Dependent Anion (Cl- / HCO3-) Exchangers (NDAE), also called sodium-dependent Cl-/HCO3- exchangers (NDCBE or NCBE) function in the opposite direction increasing intracellular pH and lowering intracellular Cl-[87, 92, 151, 287, 288, 315, 321]. The expression of NCBE precedes KCC2 in the embryonic mouse brain and, unlike KCC2, NCBE is expressed in the peripheral nervous system and epithelial non-neuronal tissues [125].

The GABAAergic agonist muscimol increases KCC2 mRNA in male neurons, via activation of voltage sensitive calcium channels and calcium signaling [79, 80]; in contrast, muscimol decreases KCC2 mRNA in female SNR neurons with hyperpolarizing GABAAergic responses [79]. These indicate that the maturational state of a neuron, as it relates to the mode of GABAAergic signaling, is critical in defining its reaction to stimuli that tend to disturb its GABA-related developmental pathway. 

Role of sex hormones in the sexually dimorphic expression of KCC2 in rat substantia nigra.

KCC2 is a neuronal-specific potassium chloride cotransporter. The level of KCC2 expression is a factor determining whether GABA(A) receptor agonists depolarize or hyperpolarize neurons. Substantia nigra reticulata (SNR) neurons of male postnatal day 15 (PN15) rats have low KCC2 mRNA expression and respond to GABA(A) receptor activation with depolarization and activation of calcium-regulated gene expression. Female PN15 SNR neurons have high KCC2 mRNA expression and GABA(A) receptor agonists cannot activate calcium-dependent signaling processes. We investigate whether sex hormones regulate KCC2 mRNA expression in PN15 rat SNR. Using in situ hybridization, we studied the effects of acute (4 h) or prolonged (52 h) subcutaneous (s.c.) administration of testosterone (100 microg), dihydrotestosterone (180 microg) or 17beta-estradiol benzoate (5 microg) on KCC2 mRNA expression in male and female PN15 rat SNR. Different doses of estradiol (1 and 10 microg s.c., 4 h) were also acutely administered in female PN15 rats. Controls received oil injections. Separate groups of PN15 male rats were pretreated with antagonists of L-type voltage-sensitive calcium channels (L-VSCCs) [nifedipine, 100 mg/kg s.c.] or GABA(A) receptors [bicuculline, 2 mg/kg intraperitoneally (i.p.)] or their vehicles, 30 min before estradiol (5 microg s.c., 4 h). Testosterone and dihydrotestosterone upregulated KCC2 mRNA in both sexes. Estradiol downregulated KCC2 mRNA in males but not in females. Both acute and prolonged hormonal administration had similar effects. In male PN15 SNR, nifedipine and bicuculline decreased KCC2 mRNA acutely and prevented further downregulation of KCC2 mRNA by estradiol. Estradiol therefore downregulates KCC2 mRNA in male PN15 SNR, by interacting with the GABA(A) receptor and L-VSCC signaling pathway.

Stimulation of prolactin and growth hormone secretion by muscimol, a gamma-aminobutyric acid agonist.


As we have seen, the normal function of GABAA receptor-mediated inhibition is governed by several factors, including subunit composition and density of the receptors and in by the appropriate ionic gradient of chloride (Cl-) and finally the release of GABA.

From a therapeutic perspective, the options are numerous and include:-

Modify the Chloride gradient

·        Reduce NKCC1 expression and increase KCC2, thereby making mature neurons

·        Block NKCC1 using bumetanide

·        Use a KCC2 agonist to stimulate KCC2

There are no KCC2 agonists currently available, but they are being developed for the treatment of neuropathic pain.

“KCC2 represents a fresh avenue to pain medication because stimulating KCC2 normalizes endogenous pain inhibition,” said De Koninck. “In normal neurons, Cl– levels are kept very low. Therefore, the effect of KCC2 enhancers will mainly touch on troubled neurons with elevated Cl– levels.”

Besides neuropathic pain, other neurological disorders with imbalances in Cl– homeostasis, like epilepsy, migraine or anxiety, could benefit from KCC2 stimulation. De Koninck plans to study KCC2-targeted compounds in an epilepsy model in which epilepsy-related neurological changes develop before they eventually trigger epileptic episodes. The model will provide insights on the effects of KCC2 agonists on seizure-evoking hyperexcited neurons and on changes in neuronal networks.

Recall the link between fibromyalia and autism?  I suggested this is what happened to females who nearly had autism.  Genetically down regulated KCC2 would be the link.

·        Modulate AE3 or NDAE to extrude  Cl 

This is possible using carbonic anhydrase inhibitors, such as Methazolamide and Acetazolamide  (Diamox)

Modulate subunit structure

·        Change the physical sub structure of GABA receptors to increase expression of α3 and perhaps α2

·        Upregulate the existing α3 receptors using a PAM (positive allosteric modulator) such as low dose clonazepam

·        Perhaps down regulate the overexpressed α5 receptors using a negative allosteric modulator

This might sound rather farfetched, but a chance would have it there are studies that show by down regulating α5 receptors you do indeed improve cognitive function.  We need a sub unit selective inverse agonist  or a Negative Allosteric Modulator.

Perhaps more interesting is that researchers trying to reverse cognitive deficit in Down Syndrome have already focused on down regulating α5 receptors.  They even have drugs in the approval pipeline.

But Down Syndrome is not autism, you are thinking.  But recall that in mouse models, bumetanide reverses cognitive dysfunction in Down Syndrome.

So perhaps the GABA switch is key to Down Syndrome, as well as Autism, Schizophrenia?

So keep your eyes out for news from Hoffmann-La Roche regarding RG1662 / Basmisanil currently in Phase 2 trials for Down Syndrome.

Allosteric Modulation of GABAA Receptor Subtypes: Effects on Visual Recognition and Visuospatial Working Memory in Rhesus Monkeys

 in mice and rats, investigational drugs that are negative allosteric modulators (NAMs) at α5GABAARs improved performance 

L-655,708 enhances cognition in rats but is not proconvulsant at a dose selective for alpha5-containing GABAA receptors.
The in vitro and in vivo properties of L-655,708, a compound with higher affinity for GABA(A) receptors containing an alpha5 compared to an alpha1, alpha2 or alpha3 subunit have been examined further. This compound has weak partial inverse agonist efficacy at each of the four subtypes but, and consistent with the binding data, has higher functional affinity for the alpha5 subtype.
These data further support the potential of alpha5-containing GABA(A) receptors as a target for novel cognition enhancing drugs.

Another α5 NAM is MRK-016, it also demonstrates nootropic effects, but may be developed as an antidepressant.

Inverse agonists of GABAA α5

·         α5IA
·         Basmisanil (RG-1662, RO5186582): derivative of Ro4938581, negative allosteric modulator at GABAA α5, in human trials for treating cognitive deficit in Down syndrome.[3]
·         L-655,708
·         MRK-016
·         PWZ-029: moderate inverse agonist[4]
·         Pyridazines[5]
·         Ro4938581[6]
·         TB-21007[7][8]

Final thoughts

There really is a lot to digest in this post.  I really did not know where to stop; it could have just kept going. 

For example, the amount of GABA itself should start out high in very early life and then should rapidly fall; this may also have been disrupted in autism. This process appears linked to certain growth factors (bFGF/ FGF2) and perhaps physical growth itself, both of which we know are disrupted in some autism. Recently FGF2 was found to be an endogenous inhibitor of anxiety.  Anxiety is mediated through GABA subunit expression.  It turns out some Mexican doctor is injecting FGF2 into kids with autism.  Other recent research shows that FGF2 can promote remyelination.  I think that demyelination is a likely shared feature of severe autism and mitochondrial disease.  There will be a post on demyelination/ remyelination.

 FGF2 and FGFR1 signaling regulate functional recovery following cuprizone demyelination


  1. Elegant, beautiful write up. Brilliant synthesis. Thank you for making things cogent and comprehensible. It tied up several bits of information floating around in my head. I have some questions, need to read again and think things through before asking.

    Many thanks Peter.

  2. Hi peter,

    What are the specific autism symptoms that improves with Bumetanide? ...( sorry,still I didn't get get clear idea ).....It's clear that Bumetanide is good at reducing cl levels in mature neurons but what will happen to budding neurons that require GABA to be excitatory?... Also I read that hormone oxytocin produced in the baby's brain triggers mothers labor and also reduces cl levels in neurons during birth...but I never got any labor cramps ..not even one. My daughter was delivered via emergency c section 10 days after the dew date. Could that be an indication that the switch didn't flip?

    1. This comment has been removed by the author.

    2. This comment has been removed by the author.

    3. Bumetanide seems to make a child more "present", as if the fog has lifted. So if I collect Monty from school and ask "what did you have for lunch", before bumetanide, I got either no response, or a very short, often incorrect response like "pizza". With bumetanide, the response would be "I had chicken with peas and carrots and fruit salad". The school saw it as being joyful and able to learn. I think measuring IQ would be the best way to determine the effect of bumetanide in autism, not to mention Down syndrome.

      I did propose in one post than women having a c section should have the option of giving their child IV oxytocin after birth to compensate for the lack of the natural oxytocin surge during natural birth. This might be part of the reason the switch did not flip, but there is more to it.

  3. Excellent work Peter. It took me about 10 minutes to read everything which means it probably took you about 10 hours to compile it all (not including paper reading of course).

    I would comment more, but I have been spending the last several weeks nonstop reading about opioid receptors, addiction, endorphins, etc. as it pertains to a new therapy that was kind of a shot in the dark, but seems to have made a very acute and giant leap in terms of social interest and reduction in SIB.

    Actually, it is more of a borrowed and very old experimental therapy 30+ years old for morphine addiction (not naltraxone which I am sure you are familiar with in terms of autism therapies), so old that I was unable to track down the principal investigator of the research but could not find him at any of the universities he worked at because he is probably retired and/or dead) that is so simple and cheap and addresses a totally different problem domain, however, all the results from the animal studies fit in line with so many things I know that are in line with autism that it was almost eery as I kept reading through it all. You can email me if you are interested in the details because this is ongoing stuff and it is very counterintuitive (so some people might get the wrong idea because it flies in the face of what every DAN doctor will tell you), but I will just say that at school my son has had very few behaviors (as the school calls it) on every day and no behaviors on some days in the last week. Needless to say this is all very preliminary, so we will see how things are going in a month or so, but if you want to know more you have my email.

    Anyways, great post and I will comment more on it when I can digest it all and try and say something semi-intelligent in reply (-:

  4. As per natural substances that modulate the GABA receptors, L-Lysine is one particular source:

    It seems more like a benzo booster in that I have read some people use it to help get off of benzo addictions because they can achieve the same effect with much lower doses. Who knows, maybe L-Lysine can modulate clonazepram's effectiveness such that microdoses would do the job in the CNS (just speculating here as I have read nothing so far to support this point directly).

  5. I forgot to add that in the study the drug they used to test if L-Lysine was a benzo booster was

    AKA Rohypnol from which clonazepam is derived.

  6. Hi peter, I read this post again and agin ..took a print out of it and took it to my pediatrician's along with other research papers. I think we will put Nandini on Bumetanide for a week and see what happens. I know you think she doesn't have classic autism but she does have brain fog that is very disturbing ( I tried every thing from candida treatment to anti viral but nothing helped much). This GABA receptor / ion channel diseases looks plausible to me. We also discussed about low dose clonazepam but decided to go with Bumetanide first and see what happens. I forgot to discuss about her high cholesterol level and potential treatment with Atrovastatin but I will in the future. If this didn't work out for her I'm sure it will help lots other autistic kids that come to see the same pediatrician.

    1. Hi, you need to know that Bumetanide is the one autism treatment that takes a while to take effect. It is lowering the chloride concentration, by slowing the inflow. So it takes time. In my son's case this took two weeks, in others it has taken longer. So you really need to trial it for a month. A week will not be enough.

    2. My daughter has regressive autism and has responded well to bumetanide. In my experience, the cumulative effects are larger than the initial.

    3. Hi RG, can you please give more details ..what were the symptoms? What is your understanding..and what are the treatments you tried? Thank you

  7. Thank you very much for clarifying that quickly.

  8. Hi Peter,
    Peter, when I first visited your blog, I saw that you use clonazepam to raise cognition and I was puzzled because I knew that many Asperger's use it to treat anxiety/OCD, without cognitive issues. Then from this article I saw that GABAa subunits are involved in both cognition and anxiety like behaviours. Classic autism is characterised by low cognition/low anxiety and Asperger's high cognition/high anxiety but they do meet at the point of stereotypy. Stereotypy as you say means oxidative stress. What is the connection, if any, between GABAa receptors and oxidative stress?
    Also cholesterol levels affect cognition, is cholesterol relevant with GABA receptors? L-carnitine is supposed to boost memory and cognition, have you tried it to Monty? I am going to trial it, can you tell me how much carnitine you would recommend daily?
    Thank you

    1. Oxidative stress effects very many things, including hormones. So indirectly, there probably is some effect somehow on GABA.

      L-carnitine does seem to help people with some mitochondrial dysfunction, but there are several classes of mitochondrial dysfunction.

      Some people with autism respond well to L-carnitine and do so quite quickly. Monty did not show much response to it.

      In autism caused by mitochondrial disease, Dr Kelley, from Johns Hopkins, suggested 50 mg/kg/day.

    2. Speaking of mitochondrial dysfunction, the three most powerful oxidative stress therapies I know of and have used myself are:

      Nicotanimide Riboside
      Methylene Blue
      C60 Olive Oil

      A paper that came out last week:

      discusses the NAD+/NADH ratio and the best way to raise this ratio acutely is Nicotanimide Riboside. When cells are under stress, lots of NAD+ is used in all kinds of reactions (over 200 I know of) and is why oxidative stress upregulates the kynurenine pathway because its endproduct is NAD+ via quinolinic acid (an excitotoxin that tends to accumulate in the brains of Alzheimer's patients and also in autistic individuals in some autism studies). Also, if B6 (a cofactor) is insufficient then you get a buildup of quinolinic acid, rather than NAD+. Cells will do everything they can to keep NAD+ levels high enough because once that is gone the cell goes under apoptosis. There are I think probably well over about 100 or more papers on Nicotanimide Riboside now and it is actively used in research to acutely raise the NAD+/NADH ration in cells.

      Also, here is a very recent paper (December) on methylene blue rescuing in vitro cells of a form of progeria where a protein called progerin somehow downregulates an important protein called PGC1 alpha that is responsible for mitochondrial biogenesis.

      Methylene blue has been used for well over 100 years in medicine as well as in many industrial and other medical uses.

      c60 olive oil - (just fullerenes dispersed in a small amount of olive oil) is also very powerful as well when it comes to reducing oxidative stress. This is more of a new kid on the block with a lot of controversy surrounding it, but its effects are quite profound after a few weeks in myself.

      Dealing with oxidative stress in this manner seems to help my son a lot, but unfortunately this is only one small piece of the puzzle. As you mention before, if GABA interneurons are dysfunctional either from a receptor problem, or else an intracellular chloride issue, then you will have excess firing of pyrimidal neurons leading to excitotoxicity (and more oxidative stress which ends up being a vicious cycle that can eventually lead to epileptic symptoms).

      Its a big chicken and the egg issue, so I think it is best in this case to deal with the chicken and the egg at the same time. Another thing you need to consider is that sometimes parents and physicians in the past would think they are successfully treating autism by giving children megadoses of GABAergics such as Risperadol and because many of the worst behaviors went away, they think they were making progress. Unfortunately, long-term benzo use is strongly linked to dementia and mental retardation as neurons that are excessively inhibited end up eventually dying, so the goal with GABA is to use as little stimulation as possible and rather than focusing exclusively on raising GABA levels (which in some studies suggest are in excess in autistic individuals), focus more on trying to increase the number as well as density of the receptors (GABA and benzo tolerance will do the opposite).

    3. Tyler, it looks like Nicotanimide Riboside is now getting some serious attention. It does have the other effects of reducing cholesterol and increasing insulin sensitivity.

      It is not cheap. What dose are you using?

      University of Birmingham Receives Grant From Wellcome Trust To Support Human Research On ChromaDex, Inc.s NIAGEN® Nicotinamide Riboside

    4. There are people who do megadoses of the stuff (they must be megarich or simply not have kids hehe) and no adverse effects have been reported. Chromadex funded three clinical trials at different doses and found the smaller dose they trialed to be the most effective which was 250 milligrams a day in the morning (NAD+ has a bit to do with circadian rhythms though I will say I take it at night with other vitamins and have had no issues). That is what I supplement with my son. Now, in individuals under extreme mitochondrial stress perhaps a higher dose would be in order. You can google "Nicotanimide Riboside" and "Mitochondrial Myopathy" and find several studies where they discuss human dosage as well.

      When I first gave it to my son it did a lot the first week and then leveled off to a plateau which could of been because other cofactors had been depleted. Some have suggested having additional B vitamins such as folate, B12, etc. because DNA repair via PARP uses a lot of NAD:

      But really there is still a lot more to know about NR, just that its the best thing around for increasing NAD+ artificially that I am aware of.

  9. For those who like me have a hard time getting prescriptions, following the link to carbonic anhydrase inhibitors (like Diamox), I see it mentions pomegranate (Punica granatum ) as a natural source of ellagitannin, some of them "highly active carbonic anhydrase inhibitors", and refers to the following article
    A search for Punica granatum also shows several papers regarding its anti-inflammatory properties, which IMO is also interesting.
    It seems worth exploring.

    1. For those in Europe, I am told that both diamox and bumetanide are readily available in Spain at the pharmacy without prescription.

  10. I posted on another thread about the improvements (cognitive, attention, fluency of speech) in our son shortly after he developed a taste for fresh mint leaves, eating a small bowl of them daily.

    It turns out menthol also has a GABA modulating activity. Our son has also been on ceylon cinnamon caps since summer (thanks Peter for suggesting) and doing well on it, but now I suspect cinnamon and mint might be working in synergy here. Might be an option for those who cannot access prescription meds? If a child wouldn't eat fresh mint leaves then menthol oil caps might be a good enough alternative?

    "... it would be of some interest to know something more about the mechanisms of menthol's effect on GABAARs. In this regard, a recent study examined menthol's actions on GABAARs compared to sedatives (benzodiazepines) and intravenous anesthetics (barbiturates, steroids, etomidate and propofol) [38]. The study indicates that menthol exerts its actions on GABAARs via sites distinct from benzodiazepines, steroids and barbiturates, and via sites important for modulation by propofol. This result is not unexpected given the apparent structural similarities between menthol and propofol (e.g. positioning of an isopropyl group adjacent to their respective hydroxyl groups). Interestingly, propofol at clinically-relevant concentrations selectively enhances tonic currents activated by GABA at low concentrations in hippocampal neurons [6], providing additional evidence favoring the idea that menthol and propofol may modulate GABAARs with a similar molecular mechanism.

    In conclusion, the present results suggest that menthol selectively enhances tonic inhibition mediated by high-affinity, slowly desensitizing GABAARs in CA1 pyramidal neurons of rat hippocampus, leading to inhibition of in vitro neuronal excitability and in vivo network hyper-excitability of the hippocampus. Our results, therefore, reveal a novel role of menthol in the mammalian CNS and underscore the importance of tonic inhibition in controlling neuronal excitability..."

    1. I have been trying your idea but the best I can do is to let a Hall's cough drop dissolve in a drink I give my son twice a day. It is 5 milligrams and all he can tolerate (he doesn't like the taste of too much menthol I guess). I looked at a lot of options and that is the best I could come up with. I also thought of maybe trying pure menthol crystals inhaled via some sort of diffusion device (some people do this with peppermint essential oil) to achieve what you were talking about. Unfortunately, I remember one of the studies you referenced having them do like 100mg/kg for GABA subunit/benzo site effects and 200mg/kg for direct GABA receptor activation in mice. I think the general rule for mice scaling to humans is to do the milligram per kilogram conversion then divide by five due to mice having much different metabolisms so that means in a 75 pound boy (34 kilos) at the lower dose you have 3400 milligrams divided by 5 which comes out to 680 milligrams which is a crap load of menthol.

      I recently read about Lysine being able to potentiate GABA drugs so that you would need less drugs to get the same effect (the papers I found were referenced on a drug abuse website where people discuss ideas about how to get off benzo addictions which happens to be reading material I never thought I would ever be interested in before now since I have never done illicit drugs and don't plan on it either).

      I might be wrong in some of my conversions here as I am recalling these values off the top of my head for the studies where they used administration of menthol orally via drinking water. In all the other studies menthol was applied by pipette to brain slices. I was never able to find any human studies on oral menthol dosage for pharmacological purposes as the closest I could find were studies via inhalation of peppermint essential oil (which has more than just menthol in it).

    2. Tyler, if your kid likes natural fruit juice, you can try a blend of pineapple and mint. Tastes good (even for some that don't like the ingredients individually) and you probably can replace the mint leaves at least partly by some concentrated form of menthol, in order to balance taste x dosage.

  11. Peter one topic I am not sure you have covered yet that might interest you are endozepines.

    One thing I find striking is a paper cited by this paper which discusses a natural rise of endozapines during labor:

    Of course if children are delivered by caesarean section, then this may not occur. If you have GABA receptors that may be impervious to chloride due to not enough of a buildup of endozepines, then you may not have an activity trigger that causes the epigenetic changes in the pre-synaptic interneuron (via feedback mechanisms) that causes the downregulation of NKCC1 and the upregulation of KCC2.

    Maybe what is needed for the interneurons to make this change is some sort of shock to the system to get them to mature and since immature interneurons are depolarizing rather than hyperpolarizing at birth, there may be some sort of feedback mechanism necessary to switch them to a mature state.

    Just as an example, in treatment resistant depression, electroshock therapy is one of the best methods available to date even though the therapy is over 50 years old. What exactly is going on with electroshock therapy is still a mystery in how it helps people with severe depression but it is hypothesized that the acute stress somehow kicks various neurons throughout the brain to some sort of new metabolic state. Ketamine therapy is also hypothesized to work somehow along those lines as well in that it provides some sort of acute stress that changes the metabolic functioning of the brain.

    So perhaps (just a hypothesis here) that in order for immature neurons to "mature" even past the labor and delivery stage is some sort of shock to induce the epigenetic changes that lead to immature interneurons to them becoming mature interneurons so that bumetanide treatments (and similar chloride mediating drugs) would be unnecessary.

  12. Feel free to delete this post as it is just another paper on endozapines that might be very relevant to low-dose clonazepram treatment:

  13. Also, you may or may not of read this paper:

    Modulation of GABAergic transmission in development and neurodevelopmental disorders: investigating physiology and pathology to gain therapeutic perspectives

    It covers much of what you have already talked about on this blog but I also found an interesting link discussing allosteric modulation of the a2/a3 gaba receptor subunit sites via a drug that just completed its clinical trials:

    for a drug called AZD7325. For anyone living in the Los Angeles area, if they do phase III trials, it might be worth looking into.

  14. Wow. I have been researching GABA for the past few months and this is just so over my head, I have just stumbled onto your site. i have learned through my 23&me testing that my genetic variants that affect me all lead to excess glutamate. Things like glycine are excitatory for me. I have focused on my glutamate/GABA balance (don't take GABA just try to support my body to produce it). I am a classic migraine sufferer from the age of 5. Most recently my migraines were about 2/month but now that I have added bcaa's I have been migraine free for 6 months. Anyways, I had been thinking about Pregnenolone. Hope you don't mind my comment - I have just been finding it interesting learning about autism while I have found answers for myself and my migraines.

  15. Does chloride in salt increase intracellular chloride levels

    1. Within moderation it does not seem to. If you consumed a great deal of salt it probably would.

  16. Hello Peter,

    So interesting, thank you for sharing your research.

    I found this by accident and wondered what your thoughts were on it?; "Pet Remedy Natural De-Stress and Calming Plug-In Diffuser, 40 ml - Product Description: As soon as plugged in you will notice a difference as the potent de-stressing properties are slow released into the surrounding area. Pet Remedy works by enhancing production of a neuro-transmiter called GABA (Gamma Amino Butyric Acid) . GABA has an inhibitory effect (ie. calms the nerve, and is a natural chemical produced by the brain, and transmitted from nerve to nerve as a chemical calming message when the body is stressed and anxious. blend of essential oils including valerian, vetivert, basil sweet, and sage."

    kind regards

  17. I just have to... ..stand up and clap my hands on this post Peter. So far (and I am closing in on reading through all your posts) this might be your best, and most extensive. I am just wowed!! :-)

    On the top of my head I have two simple questions on this:
    1) In which (GABA modulating) category does Ponstan fall into of the above?
    2) Is I/E imbalance related to hyperactivity? I don't think you have written about the causes of hyperactivity anywhere on the blog, yet it is highly relevant for many affected to ASD. Maybe it is very obvious?

    It seems that a lot of the suggested interventions that should be good often also have the effect of hyperactivity. And so people give up thinking that they fixed the wrong thing. Though maybe they actually did fix one of the issues but need to adress another one too before seeing the positives.


    1. Ling, Ponstan does have potent effects on GABAa recepotors. This was looked into in the following post:-

      Ponstan is a positive allosteric GABAA receptor modulator, but it is sub-unit specific. So it is rather like low dose clonazepam. I think Ponstan in addition has more complex effects on GABAa receptors.

      It has effects on numerous ion channels that related to autism. Knut's point is that dysfunctional ion channels at critical moments in the brain's development cause permanent damage, so by modulating these ion channels just for a short period, you can have a permanent effect on the brain.

      I think there are many types of hyperactivity, just as there are for anxiety. The words are not very precise. I suspect that some people's hyperactivity is related to excess T3 thyroid hormone, just like too little results in lethargy. You cannot easily measure the level of T3 in the brain. It is not going to be the same as in the rest of the body. T3 in the brain is mainly produced in the brain from the pro-hormone T4 that is circulating, via an enzyme called D2. All kinds of things can lead to too much, or too little, D2.

      Psychiatrists used to give oral T3 to people with depression.

      Broccoli powder makes some people hyperactive. Broccoli contains kaempferol, which increases D2 and hence T3.

      The small polyphenolic molecule kaempferol increases cellular energy expenditure and thyroid hormone activation.

      So perhaps broccoli powder would be good for depression.

      I am sure that other types of hyperactivity relate to things other than thyroid hormones.

      My son has I/E imbalance, but has never exhibited hyperactivity. I think he could do with more D2/T3 centrally (in his brain), but other people are the opposite.

    2. Thanks for the quick and extensive response! Now I just have to bury myself in the closet and get some time to ponder all this information for a while...


    3. Ok, back from the closet, with new questions:
      1) What do you think about the idea of microdoses of Ponstan?
      2) Enzyme D2, is that prostaglandin D2 or something else? These prostaglandins keep coming up in various contexts...


    4. Ling, the active thyroid hormone is T3.

      The level of T3 sets the metabolic rate. T3 is regulated by enzymes that produce T3 from circulating T4 and other enzymes that neutralize it. These enzymes are called Deionidase 1,2 and 3

      The complication is that the body does not have a single metabolic rate. T3 is made locally, so the level in the brain is not the level in your blood.

      The effective dose of Ponstan is unknown. If it has a good effect, I would gradually reduce the dose to find the minimum effective dose. This will clearly minimize any side effects.

  18. Could you possibly tell me what a good solvent for Clonazepam would be?

    1. Crushed Clonazepam tablets do not dissolve in water, you just get a suspension. Pharmacists have special viscous syrups that can hold such substances in a uniform suspension. The "home-version" is just to use water, an old 100ml glass bottle from cough medicine, or child pain killer, and a cheap hand-held frappe mixer. Put the tablet in bottle, add water and let it slowly "dissolve", then put the mixing end of the frappe mixer inside the bottle. It will aggressively agitate the contents. Then use a syringe to measure the desired amount of the mixture. Keep in the fridge and when empty, give the bottle a good clean. Also clean the syringe. This is not perfect, but it is good enough.

  19. Would it be absolutely necessary to keep it in the fridge? Sorry I just can't come up with a reason as to why it would make a difference.

    1. You may find mold growing if you are not careful.

    2. Ok that makes more sense thank you.

  20. I tried giving my son Gaba and 5htp supplements a while ago, which aggitated him more. Does this mean he may be a Bumetanide responder?

    1. In theory GABA itself does not cross the blood brain barrier (BBB), but in some people the BBB is more permeable than it should be.

      A negative reaction to any GABAa agonist that can cross the BBB suggests a bumetanide responder. Valium is a good example, Picamillon which is otc is another.


Post a comment