Benzodiazepine Binding Site

[Hammilton]

It's been a while since I started a thread here, and I thought this might be useful to other members. It's just started so far, as I've only gone through one paper at this point. I plan on doing on article per day until I have exhausted by GABA-A receptor stuff. Hopefully someone interested in receptor modelling will be interested in reading this.

Because of its nature, there's a good chance that some of what is written here will be changed as I progress, but this will be updated regularly until I'm done, so it shouldn't matter much. There are some things you should know before you read this or try to do anything with it, but only one is coming to mind right now, and that's to let you know that when you see g2-X## (or sometimes just g2X##) it is referring to gamma2, the subunit looked at in this first study. The letter refers to the amino acid, and the number which repetition of that amino acid it refers to. Often you will see g2-X##C. This refers to a specific amino acid (as mentioned earlier) that has been replaced with a cysteine genetically.

So, g2-A52 refers to the 52nd alanine in the gamma 2 receptor protein. g2-A52C would then refer to the gamma 2 receptor protein that has had its 52nd alanine replaced with a cysteine.

After I have the posted all of the mini-documents here, I will combine it into one document, I'm thinking a spreadsheet, to make it really easy to see which amino acid in which subunit is involved in forming the benzodiazepine binding site, and how it seems to be involved. Sorry about the formatting, Word to vBulletin, I guess.

Details of the BZD Binding Site
• To date, six residues in the g2 subunit have been shown to affect ligand discrimination at the BZD site: g2-F77 (Buhr et al., 1997; Sigel et al., 1998), g2-A79 and g2-T81 (Kucken et al., 2000), g2-M130 (Buhr and Sigel, 1997; Wingrove et al., 1997), and g2-M57 and g2-Y58 (Buhr and Sigel, 1997; Kucken et al., 2000).
• We demonstrate that the polypeptide backbone surrounding g2-F77 is a b-strand, that g2-A79 and g2-T81 line the BZD binding pocket
• For g2-D75C- and g2-I76C-containing receptors, FLZM EC50 values were increased 19- and 10-fold, respectively (Table 1; Fig. 2B). In other words, when g2-D75 and g2-I76 are mutated to cystiene, flurazepam affinity is increased markedly.
• FLZM was unable to potentiate IGABA in a1b2g2-F77C receptors even at high concentrations (.10 mM). In other words, when g2-F77 becomes a cysteine, benzodiazepine binding is severely hindered, indicating that it lines the BZD binding site, and is involved in binding.
• Other ligands, such as zolpidem, and CL 218-872 were tested on GABA-A a1g2F77C receptors expressed by human kidney, they were completely ineffective, suggesting that g2F77 is especially important.
• Explanation of experiment for part of the binding site discovery, especially g2-A79 and g2-T81
 “Although Ro 15-1788 is a BZD antagonist that competitively blocks the binding of FLZM, it does not enhance or inhibit IGABA. Thus if the rate at which a MTS reagent reacts with an introduced cysteine is slowed by both FLZM and Ro 15-1788, then it is likely that both compounds are blocking the MTS reaction sterically and that the introduced cysteine is positioned in the BZD binding site. MTS reaction rates were measured by examining the decrease in FLZM potentiation of IGABA after repeated exposure to MTSEA (a1b2g2-D75C), MTSEA-biotin (a1b2g2-A79C), or MTSEA-biotin-CAP (a1b2g2-T81C).
 Flumazenil inhibited g2-T81C sulfhydryl formation, but flurazepam did not, indicating that g2-T81 may be part of an extended binding pocket for flumazenil at least, and possibly antagonists in general.

• Amino acids lining the BZD binding site: g2-F77, g2-T81 (see directly above) g2-A79
• Amino acids from g2-T73 to g2-T81 form a beta-strand
• AA g2-A79 has been suggested to participate with pi-pi stacking with the 5’-phenyl. Czajkowski says: “Although it is unlikely that FLZM chemically interacts with this alanine, the small size of the methyl group at this position may be important in creating an open volumetric space to accommodate BZD ligands of different sizes.”
• Neither flurazepam of flumazenil bind near g2-D75 or g2-T73, but g2-D75 is thought to be important in maintaining the binding site structure, because replacing it with a cysteine results in decreased flurazepam sensitivity.

 

[MurphyClox]

I've seen and read Hammilton's efforts at the dark side and it's really a shame that I can't contribute much though. My pharmacology knowledge is quite basic here...

Anyway, I'd like to help out (not just Hammi but everyone else interested in the topic, too) with providing an important resource:

"The GABA Receptors"
edited by S. J. Enna & H. Möhler
Humana Press
Download-Link

I think this should be the basic literature for everybody who wishes to participate in this discussion, as it clears out many questions. Reading is education

Cheers! Murphy

 

[permastoned]

One thing I have always pondered regarding the BZD binding site: is there any endogenous ligand or is it just chance that the benzos happen to bind to a site on the receptor and cause a conformational change?

I know that there are some neurosteroids that bind but I was under the impression that they bind to different sites than the benzos. It seems strange that a site would be there without an endogenous ligand. Darwin, eat your heart out! I'm gonna go slit my wrist now for using that turn of phrase.

 

[Hammilton]

Well, the difference between the "benzo site" and the site that the neurosteroids bind is not really as big as you think. I'm going to get the numbers wrong, but the neurosteroids in question bind to alpha5 containing (6? 4? both? My memory is terrible and I'm running on 2 hours sleep in 36 hours) GABA-A receptors. Where an alpha5 subunit butts up against a gamma2 is where you find the binding site for the steroids.

benzos bind to the site were alpha1 butts up against gamma2. I don't want to go too far, but I would not be surprised if the binding site in a1y2 receptors is coincidence and an artifact of the existence of the a5y2 binding site.

They may all be GABA-A receptors, but there were many different types of GABA-A receptors.

 

[Deleted member 81238]

Benzodiazepines cause permanently bad brain changes/damages. DO NOT USE!

 

[MurphyClox]

Benzodiazepines cause permanently bad brain changes/damages. DO NOT USE!

Could we please refrain from such unconstructive statements, especially if they aren't backed up by any viable references and lack the overall context? This is ADD, not 'The Lounge'!!!

Stay on topic, for f*** sake!

- Murphy

 

[PsychedelicPeptide]

Hammilton do you know how much sequence similarity exists between the nicotinic acethylcholine receptors and the GABA receptors? I suggest you get both of their amino acid sequences lined up next to each other for comparison. A lot of structural data (including an entire receptor) exists for the nAchRs already, and it would be the most logical starting point to model a GABAR off of.

http://www.expasy.ch/tools/

This website has a bevy of proteomics tools. Important for your purposes here are the BLAST search (to find similar sequences) and the CLUSTALW multiple sequence alignment function. Both of these are in this page of links.

Also you should go search through the PDB (http://www.rcsb.org/pdb/home/home.do) for what existing structures related to the entire Cys-loop LGIC family (of which GABA is a member), if you are not in the position to even try obtaining your own laboratory data, you need to make the best use of what already exists for your purposes. This means finding structures in the family that might have some homology with the GABAR and thus be relevant to showing you what structure a particular segment of it may exist in.

Some structures I found for you (these can be opened up in accelrys but accelrys gives you a lot of power and can be overwhelming for simple structure viewing at first, I suggest getting whatever latest version of pymol, jmol or something similar that doesn't let you do everything like accelrys does and viewing in there first -- pymol is the best but if you have windows vista [any other win works though] their free older software is not compatible with it, never versions are but you have to be a student or pay to get access -- http://pymol.sourceforge.net/):
http://www.rcsb.org/pdb/explore.do?structureId=1OED
http://www.rcsb.org/pdb/explore.do?structureId=2BG9
http://www.rcsb.org/pdb/explore.do?structureId=2QC1
http://www.rcsb.org/pdb/explore.do?structureId=1C5D

These were just from searching for the nAchR. GABAR itself has some segments in there, and I'm sure there's some other bits and pieces of other Cys-loop receptors you can find in there.

Also important to check out are any papers that these structures are published with, and especially any reviews papers you can get on Cys-loop family LGIC structure. There's sure to be at least a few since the nAchR structure publication was a big deal back in 02-03 (I remember presenting a review paper along with the Nature paper myself to a class around then =P).

Do some literature searches at pubmed.org for the latest in reviews on Cys-loop LGIC family structure and function. For your purposes I think the most important thing would be finding papers with good diagrams/models/drawings in them. Don't worry so much about something like allosteric regulation, mechanism of ion selectivity or even really the specifics you listed above about binding site mutations and their effects. First you need to get your grounds set with the available structural data and then look for homology in sequence, combined with what is present already in the literature of course. Think model first, function later. I don't know how much that helps, but this is where I would start.

 

[Hammilton]

Doing all of that and more.

It seems the papers I've been reading break in two directions, either creating a model of the whole receptor, and others just of the binding site(s). When I can't sleep, going through and marking off which amino acids are directly involved binding and line the pocket is easy and doesn't take much thought. The other stuff is interesting and useful to me, but this should provide a shortcut for others doing similar things.

IIRC, nAchR and GABAR have something like 22% homology. That's not all that great, but about average, I think.

 

[vecktor]

One thing I have always pondered regarding the BZD binding site: is there any endogenous ligand or is it just chance that the benzos happen to bind to a site on the receptor and cause a conformational change?

I know that there are some neurosteroids that bind but I was under the impression that they bind to different sites than the benzos. It seems strange that a site would be there without an endogenous ligand. Darwin, eat your heart out! I'm gonna go slit my wrist now for using that turn of phrase.

my guess is there is a natural enogenous ligand of the true BZD site is probably a carboline or a isoquinoline (cyclised tryptamine or PEA respectively) of some sort as a lot of carblines have affinity and these creatures are found in vivo.
sites don't have to have endogenous ligands per se, as far as I am aware the PCP high affinity/ ion channel site in NDMA doesn't.

 

[PsychedelicPeptide]

Just make sure you pay attention to what kind of chemistry or technique people use to identify the "binding sites." Some techniques can produce artifacts and results can often be misinterpreted especially if the people do not know entirely what they are doing or jump to conclusions which can be easy to do if techniques are not performed with enough precision. If you have any doubts about anything bring up the paper or technique and I can double-check how valid the experiment looks if you want. This is all very important and given how you may be dealing with neuroscience-trained PIs that have divulged into purely biochemical/chemical means later in their career to study these sorts of interactions, this increases room for errors. Hopefully not though, a lot depends on the system.

I recently came across a paper about a receptor model of a particular GPCR where the group used constraints for the ligand binding residues out of a few scattered papers. One paper they referenced for a specific residue on the ligand to bind to a specific site on the receptor was horribly misinterpreted as I noted upon examining their reference on my own, where the original paper did not even do work around the critical ligand residue the modelers used the constraint on (they clearly do not know much about the ligand-receptor system and probably only did the model because they have the computer know-how to make it). I wouldn't want to publish something like this that's for sure. But it just shows how much caution you need to exhibit when looking at this science where people are trying to make delicate structural models of things they cannot see. Their model looks nice though, just wrong.

 

[nabollocks]

Hammilton: Can you shed any light on these article? Particularly which GABA sites are effected and how?

Benzodiazepine/GABA(A) receptors are involved in magnesium-induced anxiolytic-like behavior in mice.

Full Abstract
Behavioral studies have suggested an involvement of the glutamate pathway in the mechanism of action of anxiolytic drugs, including the NMDA receptor complex. It was shown that magnesium, an NMDA receptor inhibitor, exhibited anxiolytic-like activity in the elevated plus-maze test in mice. The purpose of the present study was to examine interaction between magnesium and benzodiazepine/GABA(A) receptors in producing anxiolytic-like activity. We examined behavior of mice treated with magnesium and benzodiazepine/GABA(A) receptor ligands, in the elevated plus maze. The anxiolytic-like effect of magnesium (20 mg/kg) was antagonized by flumazenil (10 mg/kg) (benzodiazepine receptor antagonist) while combined treatment with the non-effective doses of magnesium (10 mg/kg) and benzodiazepines (diazepam (0.5 mg/kg) or chlordiazepoxide (2 mg/kg)) produced synergistic interaction (increased time in open arms and number of open arm entries) in this test. The obtained data indicate that benzodiazepine receptors are involved in the anxiolytic-like effects of magnesium.

http://www.find-health-articles.com...ors-involved-magnesium-induced-anxiolytic.htm

And this one:

1998: Hantouche E G; Guelfi J D; Comet D
[alpha-beta L-aspartate magnesium in treatment of chronic benzodiazepine abuse: controlled and double-blind study versus placebo]
L'Encéphale 1998;24(5):469-79.
OBJECTIVE: To evaluate the efficacy of alpha-beta L-Aspartate Magnesium (Asp Mg) in discontinuation of long-term benzodiazepine use and to search for a predictive model of success for BZD cessation. METHOD: Using a double-blind procedure, 144 patients selected as chronic users of one of 3 BZD lorazepam, alprazolam or bromazepam (duration of use > 6 months; regular dose > or = 3 mg lorazepam equivalent) and with clinical remission (score on Hamilton-Anxiety < 14; Raskin-Depression < 6) had entered a controlled study (versus placebo) and were randomized in two parallel groups. The trial was conducted on 3 consecutive phases (co-administration of Asp Mg or placebo with BZD during 1 month; gradual taper of BZD during 1 month; follow-up during a third month after complete BZD discontinuation, with urinary BZD control on d75 and d90). RESULTS: The intent-to-treat analysis showed at the endpoint an overall rate of 80% of "BZD discontinuation" and of 35.4% of "BZD cessation without withdrawal" in the total population (no significant intergroup differences were observed on these rates). However, there were some tendencies to positive differences between Asp Mg versus placebo on the following: 1) prolonged delay of BZD use if reintake (30 days vs 20 days, p [log-rank] = 0.5); 2) reduction of withdrawal intensity: 11% of important difficulties during BZD cessation versus 23% with placebo (p = 0.2) and on Benzodiazepine Withdrawal Symptoms Questionnaire (BWSQ) (final score 4.0 vs 4.8, p = 0.10); 3) lower modification of anxiety during BZD tapering and discontinuation (rate of increase on HAM-A between d30-d90 of 6% vs 23% in placebo group, p = 0.10). Moreover, 3 predictive factors of "success" (BZD cessation without withdrawal phenomenon) were identified by uni- and multivariate analysis with logistic regression: chronicity of anxiety disorder (p = 0.04) and amplitude of BWSQ change during tapering phase (p < 0.0001) as negative factors; and initial score of Speilberger Anxiety Inventory "Anxiety-Trait" (p = 0.002) as positive factor. A predictive model is constructed according to these 3 parameters. Further clinical trials are needed to explore the benefits of alpha-beta L-Aspartate Magnesium in different criteria of prescription (dosage, duration of treatment, repetitive cures...).

http://www.biomedexperts.com/Abstra...buse_controlled_and_double-blind_study_versus

ideas?

 

[Hammilton]

Sorry, I just moved so I have to go to my parents house to use the internet and watch TV, which is getting a little old.

Unfortunately these are behavioral tests, not binding assays. Flumazenil is anxiogenic, so it's posssible that the anxiogenic nature cancels out the anti-anxiety effects of magnesium. In the second, I imagine that it's the magnesium ion that's responsible.

I'm zapped, so perhaps someone else can comment on this.

I doubt magnesium is going to become a drug of abuse though.

 

[nabollocks]

Ha.. certainly not going to become a drug of abuse. But definately a great tool in your goodies box.

I just recently stopped Alprazolam with the help of magnesium among others.

I found this article and was just wondering where exactly at the GABA(A) receptor it was bound, and saw your recent studies and thought I would see if you had come across anything similar.

i.e. I wanted to know if I was holding myself back by using magnesium.

Ham: Good to see the move has not re-isomerized you

 

[Hammilton]

Actually, moving was never what isomerized me, that was just coincidence. Did it actually happen when I moved last? I don't remember. It was just an interesting catch 22,

Forgotten Password
Closed Email (didn't access my hush account for 30 days, and wasn't going to spend money on it)

So, can't have it resent because I can't access the email

Can't change the email because I forgot my password

NOTE: I just wanted to say that I haven't forgotten this. I have a few more pages worth of stuff that would be useful time savers for anyone else working on this sort of thing, but I don't have internet access at my house, so I'm visiting my parents to post stuff. It was supposed to be hooked up last week but my carpet hasn't even arrived, so I'm not holding my breath. My house is all hardwood floors except my living room that I'm putting new carpet in (at 20USD/yard^2!) and so I've been living with plywood for a living room floor until they finally show up, which is taking forever. The previous idiots apparently allowed their cat or dog to use a corner of the room as a urinal, because I had to replace a bunch of wood and paint over what couldn't be replaced. Disgusting, who the hell could live like that?

I have so much work to do. I pulled out stumps and cut down brush last week, changed the shower stuff, pulled out a big cabinet that was in the way and superfluous, painted the baby's room, started refinishing the most trafficked floors, new door for the basement, planed the doors that were sticking like crazy, hauled away brush, brought in new refrige and oven, new kitchen sink, and dealt with water in the basement. I'm right across from the Wisconsin River, and 5 inches of rain one night meant 1 inch in my basement. I just hope my sauna doesn't get ruined.

Still, even with the tons of work to be done, at least I'm not on the third floor of a crappy apartment. It's nice to have a seperate dining room and a study, and a kitchen that >1 person can fit in.

I'll try to update this tommorow.

 

[nabollocks]

Keep up the good work Ham.

In Australia, this is about the time we ask a few mates around for a bbq Beer and a piece of meat goes along way in Australia.

 

[kakti]

Bumpity bump

 

[Hammilton]

I went through and compared all of the different Cys-loop family LGIC subunits to eachother, accidentally lost it, and just finished it again. It's not as complete as before, but I'll get it up as soon as I remember.