Thiomuscimol GABA-A receptor agonist 5-(aminomethyl)-3-(2H)-isothiazolone Mol. Wt. 130.10; C4H13N3O2S; [62020-54-6]; White solid; mp 140°C Soluble in water, slightly soluble in ethanol. Solutions may be stored tightly sealed for several days at 4°C O \\ C---C | \\ | C-CH2-NH2 | / HN---S Br. J. Pharmacol., 87, 677 (1986); Eur. J. Med. Chem., 20, 447 (1985); J. Neurochem., 32, 1717 (1979). Research Biochemicals International (RBI), 1998 Pricing: 5mg $64, 25mg $208. ------------------------------------------------------------------------- Kurzer, J. Chem. Soc. 2288-2295 (1955) [Thiadiazoles, Part II] Kurzer, J. Chem. Soc. 2345-2352 (1956) [Thiadiazoles, Part III] Slack et al, J. Chem. Soc. 3061-3072 (1959) [Isothiazoles; A new heterocycle] Slack et al, J. Chem. Soc. 2032-2039 (1963) [Isothiazoles; Part II] Chem. Abs. 74, 125543p (1971) [Isothiazole sulfides] Chem. Abs. 74, 125521e (1971) [3-Hydroxyisoxazoles] Chem. Abs. 74, 125527m (1971) [3-Hydroxyisoxazoles] Feuer et al, J. Org. Chem. 29, 935-938 (1964) [3,5-Dialkyl Isoxazoles] Beam et al, J. Org. Chem. 35(6), 1806-1810 (1970) [New isoxazole synthesis] Loew et al, J. Med. Chem. 13(4), 738-740 (1970) [Muscimol analogues] Micetich, Can. J. Chem. 48, 3753-3755 (1970) [Isoxazoles via Isoxazolines] Hamlet et al, Tet. Lett 24, 2101-2104 (1970) [3,5-Dialkyl Isoxazoles] ------------------------------------------------------------------------- Yesterday during my exam I finally had time to read through the papers more thoroughly, and to look at a synthesis of the chain compound. I agree with you that that is the tricky part. I would be tempted to try a quick and easy synthesis just to see if it works, namely start with malonitrile (NC-CH2-CN) add H2S (probably H2S/pyridine would be fine, and use two moles H2S), then you get HN NH \\ // C--CH2--C / \ HS SH which should react with one mole of persulfate to give dithiamuscimol: HS \ C---CH // \\ N C--NH2 \ / S Sorry, I didn't mean dithiamuscimol, I meant the amine precursor to that. To get this compound would be very easy, and the reactions fairly likely to work. You could then try hydrolyzing the compound with OH-, and hope that the SH will hydrolyze and that the NH2 will stay intact. The 3- and 5- positions are vulnerable to nucleophilic attack (unlike many aromatics), because of the nitrogen. This hydrolysis would be a bit more of a gamble, though I would say the odds are in favor of it working (at least if you can get the conditions right). If this fails, you might want to take malonitrile and then form an imidate using PhCH2O-, to give a benzyl monoimidate, then doing the py/H2S reaction with the other nitrile. The thing is this gets tricky because you have to selectively produce the monoimidate, and there is no difference in reactivity between the two cyano groups in malonitrile. Also, I don't know whether an imidate would react with py/H2S, complicating the issue still further. You could solve the first problem by making the benzyl imidate from the corresponding monoamide, 2-cyanoacetamide (which I see is cheaply available). However if the second problem is real, then you still have to face that. If the imidate survives, then you get the following route to your amine: NH2-CO-CH2-CN + PhCH2OTf ===> PhCH2O-C(=NH)-CH2-CN ===> HN NH \\ // C--CH2--C / \ Ph-CH2-O SH and this should cyclize to give the desired isothiazole with persulfate. After the Sandmeyer rxn etc, you should end up with the cyano compound, and if you choose the reducing agent well you can probably both reduce off the benzyl to leave OH and also reduce -CN to CH2NH2 in one step. You are quite right about the stability of the isothiazole system. once I thought about it some more I realized that the lone pair on the ring N is tied up at 90 degrees to the pi system of the ring (as it is in pyridine). This bends it out of the way of conflicting with the sulfur lone pair which is used to provide 4 of the 6 pi electrons for aromaticity, and which is clearly vertical (if the ring is defined as horizontal). The second S lone pair is pointing away from the ring also. So whereas a system R-CH=N-S-R would be too unstable to even exist (I would guess) in an open chain, when in this ring it is in fact very stable. Since the same arguments also apply to muscimol itself, I can see why it has unusual stability for an N-O compound. ------------------------------------------------------------------------- [...] as well as a couple [of papers] that was cited in J. Chem. Soc., 3061 (1959), which look relevant. You should be aware that there is probably a much simpler way of making your target compound, but it is also true that the kind of chemistry from that time is pretty reliable and easy to run, even if it is a lot of work. I was thinking along the lines of: O CH3O // \ CN--CH2--C ==Me2SO4 or Me3BF4O==> C--CH2--CN ==H2S/Pyridine==> \ // NH2 HN CH3O SH CH3O \ / \ C--CH2--C ==K2S2O8==> C---CH // \\ // \\ HN NH N C--NH2 \ / S OR: (Simpler, but may not work - worth a try definitely) CN-CH2-CONH2 ==H2S/Pyridine==> H2N-CO-CH2-CS-NH2 ==K2S2O8==> HO \ C---CH // \\ N C-NH2 \ / S There is a good chance it would work, and would save trying to go R-OCH3 to R-OH later on. However it may that it *could* mess up the diazotization step. -------------------------------------------------------------------------- I have recently synthesized a sulfur for oxygen (thio) homolog of muscimol. Here is an outline of the reaction. I am not including stoichiometry; the knowledgeable can figure that out for themselves. Also , I assume no responsibility for anyone's assay of this new compound, which may be toxic, and whose dosage level is unknown. It is highly likely, though, that this is a new psychedelic of high potency and powerful effect, as simple thio substitutions have been highly successful, ala Shulgin and phenethylamine homologs. I use literal IUPAC style nomenclature here, and it is not totally compatible with separate heterocyclic nomenclature -- but does give more detail for the layman. Here are the structures. start with or make the chain 3-amino-1-imino-1-propanol (tautomer of 3-amino-propionamide) HO \ C-CH2-CH2-NH2 // HN bubble this with H2S for some time with a pyridine catalyst (pyridine and hydrogen sulphide is perhaps the smelliest mixture in all chemistry so use a hood). this produces 1-imino-3-thioamido-1-propanol HO NH2 \ / C--CH2--CH // \ HN SH react this with ammonium persulphate/NaOH for the cyclization to HO HO \ \ C---CH C---CH2 // \\ // \ N C-NH2 <----> N C=NH \ / \ / S S (Two isomeric resonance structures) 5-amino-2-aza-3-hydroxy-1-thio-cyclopentadiene, or "5-amino-isothiazole" react with cuprous bromide to 2-aza-5-bromo-3-hydroxy-1-thio-cyclopentadiene react with cuprous cyanide to 2-aza-5-cyano-3-hydroxy-1-thio-cyclopentadiene react with HCl in stannous chloride to give the final product HO \ C---CH // \\ N C-CH2-NH2 \ / S 2-aza-3-hydroxy-5-methylamino-1-thio-cyclopentadiene (which could also be called "5-aminomethyl-isothiazole" or "thiomuscimol"). It is a tautomer of the structure 5-(aminomethyl)-3-(2H)-isothiazolone, seen at the top of this file. This is just the outline. details can be found in published journal articles. If anyone qualified is interested, I will send a bibliography and more details to them. Again, I assume to responsibility for the misuse of this unknown , untasted compound. It is currently legal to make and possess, but I also assume no legal responsibility for misuse of it. 10/31/97 ------------------------------------------------------------------------------- 6/15/98 Update: It has now been tasted up to 500 mcg, with distinct psychedelic threshold effects. This makes sense, as animal studies give it a GABA-A binding affinity strength an order of magnitude greater than that of muscimol. I have placed articles relating to this area on five email accounts for public download. Go to www.yahoo.com Click on email. The five account names are tmusc1, tmusc2, tmusc3, tmusc4 and tmusc5. the password for all of these is fruitbat633933. please do not erase them. the 63 page total is scanned in .jpg format, as it would not OCR. so its 9 megabytes total spread out in five .zip files. dm_telvis@yahoo.com and fiaof93@rocketmail.com