Alcohol abuse and alcoholism are major health problems. It is likely that a solution to these problems will require an understanding of the effects of alcohol on specific ion channels and the kinases that modulate them. Specific binding sites for alcohols have recently been described in the transmembrane domain of the superfamily of glycine, GABA, nicotinic acetylcholine, and 5-HT3 receptors. Our hypothesis is that alcohols bind within cavities that are bounded by transmembrane segments of these receptors. Our goal is to define the properties of those sites that regulate binding and efficacy of alcohols. These binding sites may provide a common motif for binding of alcohols within other classes of ion channels. We will build computational models of binding sites and design specific site-directed mutations to test this hypothesis. These mutations will be expressed and tested by our collaborators, Drs. R. Adron Harris and S. John Mihic, in separately funded experiments. Specifically:
Aim 1. We will define specific amino acid residues that determine the """"""""cutoff"""""""" length of long-chain alcohols. We have previously shown that mutation of S267 in transmembrane segment 2 (TM2) and A288 in TM3 of the glycine alphal receptor can change the alcohol """"""""cutoff' from heptanol to dodecanol. We will develop computational molecular models that allow us to suggest mutations that will determine additional residues in TM1 and TM4 that may also form """"""""walls"""""""" of the putative binding cavities. We will refine our models by iterations in which we optimize the structure of an initial model, use it to predict mutations, test if the model is consistent with the resulting experimental data, and then modify the model in a way that would better fit the data.
Aim 2. We will determine the structural requirements of alcohols for potentiation of agonist potency by providing models in which a series of alcohol analogs are covalently linked to site-directed cysteine mutations in the putative binding cavities. Since we will know that a single alcohol analog is bound to the putative site, we can distinguish binding from efficacy.
Aim 3. We will define the proximity of amino acid residues important for alcohol potentiation of agonists by building models that predict double site-directed cysteine mutations that are appropriate for cross-linking. We will predict pairs of residues that could be linked by direct disulfide formation or with bi-functional methanethiosulfonate reagents with 1-5 carbon spacers. In summary, these computational studies will provide new knowledge about determinants of alcohol binding and efficacy.
Showing the most recent 10 out of 41 publications
