Molecules: You'd better learn to live with them.
Last night I mentioned that diphenhydramine is a somewhat promiscuous molecule, binding to a number of disparate receptors. One of its effects is some inhibition of serotonin reuptake. As you've probably read in countless pop science articles by now, this is one mechanism of action for antidepressants. Take a look at diphenhydramine again:
And Fluoxetine/Prozac:
Note the similarity in structure to diphenhydramine. Also note the trifluoromethyl group (-CF3). You don't really see organofluorine compounds in Nature (it's so reactive it's been caught up in rocks and ores for ages), but they are ubiquitous in medchem. This C&E News article estimates that one in five drugs contains at least one fluorine atom. Fluorine is such a singularly unique element it has its own journals. Fluorine chemistry is a field in and of itself.
For an overview of some of the history of psychopharmacological agents (including the story of diphenhydramine's kinship with modern anti-depressants), take a look at this review article from Psychosomatic Medicine.
Comments
I think I am going to fall in love with this blog... lol I love chem. Pls Pls don't close this blog :)
Posted by: Celeritas | November 8, 2006 10:08 AM
In NMR, F-19 (100% natural abundance) resonates at 94.08% the frequency or magnetic field of protons and with about the same sensitivity. It would be interesting to slightly downgrade an old MRI machine, slip in a Prozactivated head, and see whether its brain is overflowing with trapped fluorinated metabolites. Betcha it lights up like a cheaply decorated Christmas tree.
Posted by: Uncle Al | November 8, 2006 11:15 AM
Sheesh, one in five eh? I can't think of too many off the top of my head, as I'm most familiar with adrenergics. Two of my favorite drugs, moxonidine and clonidine have chlorines though.
Have you though of using JMOL to embed 3D models in your blog?
Posted by: Ian Musgrave | November 8, 2006 11:34 PM
Good reason why not, JMOL is a nightmnare to set up (at least for me). I have a range of teaching presentations set up using the MDL chime plugin, See here for example (you must have the chime plugin enabled to see the manipulable 3D structures), but Chime is no longer supported, and the plugin must be loaded onto each and every computer. Getting that sort of thing supported by our IT folks is a pain.
JMOL is a JAVA implemented system, and will work on anything that allows JAVA to run. The JAVA should reside on the local server (which may be a pain with our IT folks, but less of a pain than having to push an unsupported plugin to several thousand computers. But I am having severe problems trying to work out how to convert Chime scripts of JMOL scripts.
If I ever work this out (and Eric Martz has, see Fisrtglance in JMOL) I'll let you know how I go.
Posted by: Ian Musgrave | November 9, 2006 9:10 PM