Matt Jones jonesmat at
Wed Oct 16 14:31:50 EST 1996

In article <32651078.13BE at> Nigel Foden, n.foden at writes:
>	You should be a little careful of people giving 'answers' to simply 
>put questions.  Propofol may bind to many receptor subtypes, and produce 
>actions at most of them (opioid, GABAa, etc.).  It is a great leap to
>therefore that an anaesthetic agent acts *ultimately* through a
>single receptor type.  If you are entering anaesthetic research I cannot 
>recommend too strongly the following references, and the groups that 
>produced them:

Hi Nigel,

Here're an unsolicited digression on traditional and modern methods for
demonstrating anesthetic mechanisms. 

I agree with your advice quoted above entirely. However, I also think
it's useful to
look at general trends with regard to possible anesthetic mechanisms. For
example, you probably know the Meyer-Overton theory, paraphrased as
follows: "Because there is an excellent correlation between anesthetic
potency and lipid solubility coefficient, anesthetics must exert their
actions by partitioning into membranes." After almost 100 years of
further research, this theory may seem incredibly naive (to me anyway).
However, the result is still correct, anesthetic potency correlates with
lipid solubility.

Now that we know about enzymes and receptors and ion channels, it may be
useful to
look for similar correlations between anesthetic potency and effects on
some relevant biomolecules. Franks and Lieb's studies of the correlation
between potency and effects on firefly luciferase were very useful, even
though luciferase is not a "relevant" biomolecule, because they showed
that anesthetics can interact with *proteins* instead of just lipids. If
I recall, they also showed a similar correlation at a moluscan potassium

So what kinds of correlation can we find with "relevant" biomolecules?
Well, I would suggest that major neurotransmitter receptors such as GABA
and glutamate receptors are potentially relevant. Also, that the major
voltage-gated channels may be relevant, and probably a lot of
transporters and pumps. It turns out that correlating anesthetic potency
with effects on these molecules works for some of them, but not for
others. For example, general anesthetics don't work very well on sodium
channels *at concentrations relevant to anesthesia*. On the other hand,
there is a very good correlation between anesthetic potency and actions
at GABA-A receptor channels, and these actions are consistent with
enhanced synaptic inhibition and sedation:

Zimmerman, S.A., Jones, M.V. and Harrison, N.L. (1994) Potentiation of
GABA-A receptor Cl- current correlates with in vivo anesthetic potency.
J. Pharm. Exp. Ther. 270, 987-991.

I agree totally that the GABA-A receptor may not be the sole relevant
site of action of general anesthetics, but the majority of data suggest
that at anesthetic concentrations these drugs *are* dramatically
affecting GABAergic signalling, in a manner appropriate to produce the
anesthetic syndrome. It would be helpful if we had a large body of such
correlation studies, in order to "weed out" the molecules that are *not*
relevant targets of anesthetics, and focus closer attention on those that

Ketamine is a major fly in the ointment for proving a unified mechanism
of anesthetic action through GABA receptors. It does not really affect
GABA channels much at all (acts at NMDA channels, and some others too),
but still produces anesthesia. On the other hand, the "quality" of
ketamine anesthesia is different than, say, barbiturate anesthesia.
Patients may be anesthetized, but still have their eyes open, and tend to
hallucinate during onset and recovery.  It's also quite soluble, which
means it falls way off any Meyer-Overton type curve. So maybe there's a
lot of different types of anesthesia. But for the common agents,
barbiturates, volatile anesthetics, steroid anesthetics, propofol and
urethane (in animal experimentation), my money is on the GABA-A receptor
as *a major* relevant site of action. 

Thanks for your patience.

More information about the Neur-sci mailing list