Scientists searching for protein-protein interactions generally must look for them in vitro. But available techniques, such as chemical crosslinking and coimmunoprecipitation, are prone to false-positive and false-negative results. Cell lysis procedures, for instance, may bring into contact proteins that normally are compartmentalized in the cell, while wash procedures can dissociate fragile intermolecular interactions. And uncontrolled chemical methods can make even the most solitary polypeptide seem promiscuous.
Now, a team of researchers at the Max Planck Institute of Molecular Cell Biology and Genetics in Dresden, Germany, has created a new approach that may make detecting protein interactions easier, and more reliable in living mammalian cells.1
The new research builds on pioneering work at the Scripps Research Institute in San Diego, where Peter Schultz developed a method for modifying tRNAs to selectively incorporate non-natural amino acids into a single position within a protein. But Schultz's method is time consuming and requires extensive genetic manipulations of both the protein to be crosslinked and a tRNA gene. Coauthor Christoph Thiele and colleagues found a way to incorporate newly designed, photoactivatable amino acids – photomethionine and photoleucine – into multiple positions without such genetic modification. The new technique is better suited to detect new interactions, Thiele says.
"This method is very reliable and easy to use," says Thiele, a biochemist at the Max Planck Institute. "We feed cells the photoamino acids and ultraviolet light and [through] Western blotting ... you can see resolvable bands and distinct interactions where you would see nothing anymore with the chemical crosslinkers."
Thiele and colleagues created the new photoamines (which contain a photoactivatable diazirine ring but are otherwise similar to their naturally occurring counterparts) by alphabromination of azicarboxylic acids, followed by aminolysis. Despite initial success with yeast, the group found incorporation into mammalian cell lines difficult. "The big step forward for us was to find growth conditions [under which] the cells would efficiently incorporate these amino acids," says Thiele. They succeeded by culturing mammalian COS7 cells in a medium lacking methionine and leucine, and sometimes also isoleucine and valine, supplemented with the photoamines. As a result, the crosslinking agents are widely incorporated into new protein.
One advantage of this approach is that the modified molecules do not affect cell viability. "It's a nice piece of work ... they've found a balance," says Tamara Hendrickson, assistant chemistry professor at Johns Hopkins University (who was not involved in the study). "They get enough [incorporation] to see crosslinking, but not enough to see deleterious effects on cell growth or morphology."
Thiele's team used the new reagents to uncover previously unknown interactions between the progesterone-binding protein PGRMC1, and the cholesterol regulating proteins Insig-1 and SCAP. The researchers plan to continue their search for new pathways in lipid regulation, and to better understand interactions with PGRMC1 and Insig-1 in cholesterol homeostasis.