Anesthetic cutoff in cycloalkanemethanols. A test of current theories.

N-alkanols containing up to 12 carbons are anesthetic; however, those with more than 12 carbons are not. This phenomenon has been termed cutoff. Lipid disordering theories of anesthesia suggest that cutoff occurs because the alkyl chains of long-chain alcohols approach the length and shape of the lipids of neuronal membranes and, therefore, intercalate into membranes without perturbing them. Protein theories suggest that cutoff occurs because the size of long-chain alcohols exceeds that of a protein binding site having finite dimensions. These theories were tested with a new series of alcohols, the cycloalkanemethanols, c(CnH2n-1).CH2.OH. Anesthetic potency was measured in Rana pipiens tadpoles using the reversible loss of righting reflexes as the endpoint. The change in order parameter induced by cycloalkanemethanols and n-alkanols in lipid bilayers made of egg phosphatidylcholine and cholesterol was measured with electron paramagnetic resonance spectroscopy. On ascending the series from cyclopropanemethanol (EC50 = 54 +/- 3.2 mM) to cycloundecanemethanol (EC50 = 7.0 +/- 0.12 microM) anesthetic potencies first increased exponentially but then decreased sharply at cyclododecanemethanol (EC50 = 13 +/- 0.2 microM). Cyclotetradecanemethanol was found not to cause anesthesia in tadpoles, even after 48 h of exposure, although saturated solutions shifted the dose-response curve of octanol from 66 +/- 2.6 to 47 +/- 2.8 microM. A linear loss in the ability to disorder lipid bilayers was observed on ascending both alcohol series such that cyclotetradecanemethanol and n-tridecanol actually increased bilayer order. Molecular length does not correlate with anesthetic cutoff in these two alcohol series. Cutoff is predicted by the ability of both series of alcohols to disorder lipid bilayers and correlates with their molecular volume.