Cyanobacteria and humans/ethanol: This is something you never would have found with any sort of logical approach
The breakthrough came when Marc Delarue and his colleagues at the Pasteur Institute sequenced the genome of cyanobacteria Gloeobacter violaceus. They noted a protein sequence on the bacteria that is remarkably similar to the sequence of a group of ligand-gated ion channels in the human brain. They were able to crystallize this protein. Harris saw the results and immediately got in touch.
"This is something you never would have found with any sort of logical approach," he said. "You never would have guessed that this obscure bacterium would have something that looks like a brain protein in it. But the institute, because of Pasteur's fascination with bacteria, has this huge collection of obscure bacteria, and over the last few years they've been sequencing the genomes, keeping an eye out for interesting properties."
Harris and Howard asked their French colleagues to collaborate, got the cyanobacteria, changed one amino acid to make it sensitive to alcohol, and then crystallized both the original bacteria and the mutated one. They compared the two to see whether they could identify where the alcohol bound to the mutant. With further tests they confirmed that it was a meaningful site.
"Everything validated that the cavity in which the alcohol bound is important," said Harris. "It doesn't account for all the things that alcohol does, but it appears to be important for a lot of them, including some of the 'rewarding' effects and some of the negative, aversive effects."
Going forward, Harris and his lab plan to use mice to observe how changes to the key protein affect behavior when the mice consume alcohol.