From Models to Medications:Identification of Medication Leads for Treating Methamphetamine Addiction
Methamphetamine (METH) is an addictive psychostimulant1 that interferes with dopamine signaling in the brain. In vitro METH activates Trace Amine-Associated Receptor 1 (TAAR1), a recently cloned G protein-coupled receptor (GPCR).2 METH-activated TAAR1-mediated signaling contributes to the drug’s stimulatory effects in mice and are predictive of METH’s addiction liability in humans, making this receptor an exciting and unconventional new target for medications designed to prolong abstinence from METH abuse.
An integral part of my experimentation process involved virtually screening compounds to 3D homology models of the human TAAR1 (hTAAR1) based on the crystal structures of related GPCRs: bovine Rhodopsin, human β2 Adrenergic, and human α2a Adenosine receptor.3 I found that the hTAAR1 model based on the Rhodopsin template better supports the empirical data obtained from cell culture assays. This confirms that the bovine Rhodopsin template built to the hTAAR1 is the better homology model for in silico experimentation.
My approach to exploring the thesis that hTAAR1 plays a role in human METH abuse led to my formulation of three null hypotheses: (a) three binding sites will not be identified in more than one hTAAR1 model; (b) mutating key residues will have no effect on ligand-receptor interactions; and (c) top compounds will not be identified by more than one hTAAR1 model.
While experimenting with the molecular models I developed, I discovered two new activation sites (which I call Sites II and III) on the receptor protein. I predicted that certain chemicals that prefer to bind to these new sites can change the shape of the receptor, making it impossible for METH to stick. If the lock changes, the old key cannot fit! So, guided by the computer-generated 3D structures of the two new TAAR1 binding sites, I designed new compounds and verified by in silico modeling that they would match the shape of the new activation sites. These new compounds may be preferred over others because their chemical structures and shape give them a stronger potential to bind to the receptor. The Oregon Health and Science University has filed a patent application on my discovery of the two binding sites and my invention of the novel compounds.
The top compounds had free energy scores that ranged between -5 to -3 Kcal/mol in the extra precision mode. In order for compounds to have potential as medications in vitro, they usually need to have a free energy score lower than -5 Kcal/mol. The novel YTN compounds I designed have the most favorable free energy scores (scores lower than -5 Kcal/mol) at Sites II and III, which indicates the promising potential of these compounds as allosteric modulators. My findings should be helpful in the rational design of a novel anti-METH medication lead. These findings can also be applied to other neurological disorders,4 and I also believe that the TAAR1 can be applied in the environment as a biosensor.
1. Miller et al., (2011) Journal of Neurochemistry 116:164-176
2. Borowsky et al., (2001) Proceedings of the National Academy of Sciences 98:8966-8971; Zucchi et al., (2006) British Journal of Pharmacology 149:967-978
3. Cherezov et al., (2007) Science 318:1258-1265