Proteins are vital to life. Enzymes that extract energy from food or allow DNA to copy itself, are proteins. Muscle fibers that make the heart beat, are proteins. Antibodies, which protect against infection, also are proteins.
The physical shape of a protein determines how it works. Proteins consist of chains of amino acids that fold to form their three-dimensional structures. Scientists often seek to understand the way a protein's architecture impacts a cell's physiology or how altered proteins cause disease.
In 1972, Christian Anfinsen shared a Nobel Prize in Chemistry for research from the 1950s showing that the linear sequence of the amino acids of a protein contains the information the molecule needs to form its functional shape. He unfolded an enzyme in a test tube, and the biomolecule re-formed correctly.
But by the late 1980s, some scientists were questioning whether the amino acid sequence alone was sufficient to direct proper protein folding in the cell. Evidence was accumulating that under stressful conditions, such as excess heat, cells use inducible "heat-shock proteins" as protective agents. These proteins appear to prevent or reverse the damaging multiprotein aggregation that can occur when excessive heat melts a protein's structure. But do cells use similar assistance under normal conditions?
In late 1987, Arthur Horwich and his group stumbled onto a yeast mutant that ultimately demonstrated such assistance was occurring. It became clear over the subsequent year that a cellular machine they had discovered, a large cylindrical assembly called a chaperonin, itself a heat-shock protein, is required for the proper folding of many proteins under normal conditions in the living cell. The amino acid sequence determined the proper fold of these proteins, as inferred by Anfinsen, but a chaperonin also was needed.
Horwich and his coworkers have studied the mechanism of chaperonin assistance ever since. The machine helps at two levels. First, the cylinder's open ring selectively binds an unfolded protein's "greasy" surface, which eventually becomes buried in the folded state. Protein binding to the chaperonin prevents misfolding and aggregation. Second, a lid structure, a second protein called a cochaperonin, is added to the open ring, thus encapsulating the unfolded protein and allowing it to fold in solitary confinement, without the possibility of aggregation, following Anfinsen's principles. After 10 seconds, the lid pops off and out comes the properly folded protein, like a jack-in-the-box.
Horwich came to research with a clinical background. He was valedictorian of his medical school class at Brown, but became intrigued during his pediatric residency at Yale by a genetic experiment: a single gene from a polyoma virus could turn a normal cell cancerous. How did it do this?
Pursuing his curiosity, Horwich went to the Salk Institute as a postdoctoral fellow to work with Walter Eckhart and Tony Hunter, who studied polyoma virus cancer transformation. Horwich watched Hunter make a spectacular discovery. Hunter, who had decided to study virus-encoded proteins that have attached phosphate groups, ran an electrophoresis separation late one night with old buffer. He observed a new phosphoamino acid spot never before seen. Recognizing it as a phosphate added to tyrosine, Hunter went on to show that tyrosine kinases and phosphorylation are key regulators of cell growth.
Having experienced the excitement of experimental inquiry, and wanting to marry his interests in molecular biology and medicine, Horwich returned to Yale in 1981 for further postdoctoral training in medical genetics with Leon Rosenberg. While there, Horwich, Rosenberg, and coworkers cloned the coding sequence for human ornithine transcarbamylase (OTC), an enzyme that detoxifies waste nitrogen by converting ammonia to urea. X-linked deficiency of OTC often leads to ammonia intoxication in affected newborn males, who appear normal a day or two after birth but then lapse into irreversible coma. The sequence offered the possibility of DNA diagnosis for affected families.
But how the cell made OTC also fascinated Horwich. The enzyme subunit, encoded by the X chromosome, was produced in the cytoplasm as a precursor, and then imported across the two mitochondrial membranes to the innermost matrix. Horwich decided to use yeast genetics to identify the steps in the process. Initially, he showed that his cloned human OTC expressed in yeast went to its mitochondria and became enzymatically active. The first mutants he found in the pathway blocked maturation of the OTC precursor to its mature size.
Then he and his student Ming Cheng had a prescient thought late one night after a day of mutant screening. It was known that proteins unfold to cross mitochondrial membranes. Everyone assumed they then spontaneously refolded to the active form. But what if a "machine" helped the refolding? After all, protective heat-shock proteins maintained protein conformations. What if there were machines that assisted folding under normal conditions? A mutant affecting such a machine would allow import into the mitochondrion of a mature-sized OTC enzyme, but it would have no activity.
They soon found such a mutant. Skeptical of their result, they performed multiple experiments to assess whether normal yeast mitochondrial proteins also were affected in the mutant. Every experiment, and later ones with collaborators from Munich, confirmed the result. They then identified the gene affected in the mutant. This gene turned out to encode a mildly heat-inducible mitochondrial protein, dubbed heat-shock protein 60, or Hsp60, which also functions at normal temperatures. Hsp60 is the subunit component that makes up the chaperonin complex.
Over the years, Horwich continued to study the machine, but working on a more easily manipulable bacterial version, called GroEL. The genetic studies proceeded to biochemical studies, in collaboration with Ulrich Hartl; then to crystallographic studies, in collaboration with the late Paul Sigler; to electron micrograph studies with Helen Saibil; to structure-function studies with Wayne Fenton; and recently to NMR studies with Kurt Wüthrich. The result is a dynamic view of the working cycle of this machine.
But major questions remain about the fate of the folding proteins and exactly how they are assisted. Horwich keeps dissecting the mechanism, doing hands-on experiments side-by-side with his staff, unlike most of his colleagues. His work in the laboratory, he says, keeps him humble before Mother Nature.
As Horwich's work on chaperonins progressed, other researchers showed that many neurodegenerative conditions result from protein misfolding. Recently, he has begun investigating how a misfolded protein contributes to Lou Gehrig's disease, where motor neuron loss and paralysis occur.
The complexity of Lou Gehrig's disease has been particularly daunting, involving multiple spinal cord cell types and numerous proteins in those cells, far from the three-protein GroEL system he has been analyzing in a test tube.
But he remains motivated. Throughout his career, he says, he has taken experimental risks that somehow paid off. He feels privileged to have a chance now to tackle a human condition related to his ongoing work about how proteins "get into shape."