Before communication went electronic, telephone operators sat at huge switchboards, plugging wires into the proper slots and connecting callers. If they occasionally slipped, the resulting conversations were short and unfulfilling.

In the cell, a group of G protein-coupled receptors fulfill a similar function, acting as sensors for signals external to the cell and the modulators of the response within the cell. In a study appearing online in Proceedings of the National Academy of Sciences, Drs. Olivier Lichtarge, professor of molecular and human genetics, and Theodore Wensel, professor of biochemistry, both at Baylor College of Medicine, married the strengths of their labs – computational biology and bench experiments to recode a receptor for the neurotransmitter dopamine (specifically the D2 dopamine receptor) so that it now responds to another neurotransmitter called serotonin.

Both neurotransmitters affect the brain and are involved in memory, cognition, mood and behavior. They often act antagonistically, said Lichtarge. The receptor the two laboratories studied are the D2 dopamine receptor and the 5-HT2A receptor. Both are members of a class of G protein-coupled receptors that are the targets of 30 to 50 percent of all therapeutic drugs, said Lichtarge.

Wensel and Lichtarge asked, how does the dopamine receptor distinguish between the two neurotransmitters when the genetic sequences are so similar, particularly where they cross the membrane?

"Typically, one receptor 'sees' and responds to one ligand (in this case, a neurotransmitter," said Lichtarge.

To answer that, they began with an "evolutionary trace approach that Dr. Lichtarge's group developed," said Wensel. "It uses information inherent in evolution that can be read out from the sequence database. They mine the billions of 'experiments' that have occurred in evolution to find the amino acid residue (the part of the molecule that binds to another amino acid, leaving out water) conserved in dopamine and in serotonin receptors throughout evolution but that are different between them."

Wensel and Lichtarge expected to find these sequences in the sites that bind to dopamine and serotonin. And they did. But they also found other key residues at the site in the protein that controls whether it responds to the binding or not.

"Once the chemical is bound, how does it happen to activate the cellular signals?" said Lichtarge. "The signal is extracellular, but this triggers activation of intracellular proteins. This is the sensory protein that sees what is outside and triggers a change inside. It binds, and then if it matches the receptor internal code there is a shape transformation."

As he and his colleagues followed their evolutionary trace, they ranked every amino acid in the protein for importance in terms of evolution. They mapped the important proteins to identify "hot spots," sets of amino acids most important to the code of these receptors.

One by one, they altered the amino acids. Some mutations increased the ability of the dopamine receptor to bind to serotonin, but the serotonin's signal was weak. Another key set of mutations did not change the binding ability of the receptor. Instead it changed its response so that it was more likely to respond to the serotonin.

Lichtarge compared it to a lock and a set of keys. All the keys come from the same blank. They can all fit into the lock but only one can turn the lock because the pins (like carved hills and valleys) on key have a corresponding set of spring-loaded driver pins in the lock. When a properly cut key is inserted into the lock, the driver pins rise causing them align exactly with those on the key. This allows you to unlock the door.

Within the cell, a ligand may be able to bind to the receptor, but unless there is a match within, the proper intercellular signal to alter function never occurs. However, altering those amino acids one at a time, painstakingly changes the alignment and eventually, the "lock" turns and the message gets through.

"We have been able to use evolution to peer inside the protein and identify those key determinants of specificity, and we switched them," said Lichtarge.

The finding has implications for drug development. Most drugs currently target the binding site, but the site within the cell where the neurotransmitter is activated provides another target. A water-filled cavity at that site in this particular system is particularly tantalizing, said Lichtarge.

"It would also be wonderful to be able to peer inside the protein and pick out specific amino acids and figure out which ones confer key functions," said Lichtarge. "Patients who have mutations at the activation site would have abnormal responses to dopamine and serotonin."

"Inside the protein is a computer that analyzes what is going on and what kind of signal it should put out," said Lichtarge. "It is not a simple conduit. The pathway is a logic circuit that analyzes what binds and whether it fits the criteria for transferring information. This is like a little computer that sits there and says you have bound to this receptor but now I have to decide if you are the right one."

What Lichtarge and Wensel have done is reprogram that circuit to change the meaning of the binding event.

Wensel said future research will look at whether the same approach will work with other kinds of proteins that affect the same receptor.

Others who took part in this work include Dr. Gustavo J. Rodriguez, a postdoctoral associate in the Department of Biochemistry and Molecular Biology and Rong Yao, a bioinformatics programmer in the Department of Molecular and Human Genetics. Both are at BCM.

Funding for this work came from the National Institutes of Health and The Welch Foundation.

Lichtarge holds the Cullen Foundation Professorship in Molecular and Human Genetics. Wensel holds the Roberts A. Welch Chair in Biochemistry.