UCSD Discovery May Provide Novel Method To Generate Medically Useful Proteins

In the September 18 advance on-line publication of thejournal Nature Structural and Molecular Biology, the researchersdescribe the mechanism by which a virus that infects bacteria—called abacteriophage, or phage—can generate a kaleidoscope of variants of aparticular protein. The paper will appear in print in Nature Structuraland Molecular Biology in October.

Since this degree of proteindiversity is extremely rare, recreating the process in a test tubecould give researchers a new way to generate therapeutic enzymes,vaccines and other medically important proteins.

“This is onlythe second type of massively variable protein ever discovered,”explained Partho Ghosh, a professor of chemistry and biochemistry atUCSD who headed the research team. “Only antibodies have more variationthan this protein in phage. However, the genetic mechanism used by thephage to generate this diversity is completely different from that usedby animals to produce antibodies, and has the advantage of giving theprotein greater stability.”

“If we can learn from these organismshow to set up a system that churns out proteins with enormousvariability, it may be possible to target these new proteins tospecific cells to treat disease,” said Stephen McMahon, a formerpostdoctoral fellow in Ghosh’s lab who conducted much of the research.“This idea has already been picked up by the biotech industry.”

Thefunction of the massively variable phage protein is to tether the phageto the bacteria they infect. The phage “predator” protein fits into a“prey” protein on the bacteria like a three-dimensional puzzle piece.However, the bacteria are constantly changing the proteins on theirsurface. To keep up with the unpredictable changes in the prey protein,the phage must generate many different predator proteins for at leastone to have an acceptable fit.

In their paper, the researchersdescribe how by altering the amino acids at one or more of just 12sites on the predator protein, the phage are able to generate 10trillion proteins, each with the potential to bind to a different preyprotein. This variability arises as DNA is being copied into the RNAblueprint for the protein. The sequence of DNA bases at the 12 siteshas unique characteristics that cause frequent mistakes to be made inthe copying process. As a result, the RNA ends up specifying adifferent amino acid, and a protein with different structural andchemical properties is created.

Antibodies are another typepredator protein that must respond to rapidly evolving prey proteins,because microorganisms are constantly altering proteins on theirsurfaces to evade the immune system. Unlike the phage protein,antibodies have a complicated loop structure. The size of the loopsvaries in addition to the amino acid building blocks that constitutethe antibody protein. Although this mechanism can generate more than100 trillion different antibodies, the researchers say replicating itin a test tube would be very challenging because the loops would havethe tendency to fold incorrectly.

“Because of its stability, thephage protein makes a better model to create protein diversity in atest tube,” explained Jason Miller, a graduate student in Ghosh’s labwho conducted much of the research. “Our discovery shows that naturehas provided at least two completely different methods to generate ahuge amount of protein variability, and it opens up a whole newplatform for protein development.”

Other contributors to thepaper were Jeffrey Lawton, Department of Chemistry, Eastern University;Donald Kerkow, The Scripps Research Institute; Marc Marti-Renom, EswarNarayanan, and Andrej Sali, Departments of Biopharmaceutical Sciencesand Pharmaceutical Chemistry, University of California, San Francisco;Asher Hodes, and Jeffrey Miller, Department of Microbiology,Immunology, and Molecular Genetics, David Geffen School of Medicine andthe Molecular Biology Institute, University of California, Los Angeles;and Sergei Doulatov, Department of Microbiology and Medical Genetics,University of Toronto.

Stephen McMahon is now at the Centre for Biomolecular Sciences at The University of St. Andrews in Scotland.

This research was supported by a W.M. Keck Distinguished Young Scholars in Medicine Award and a UC Discovery Grant.