Biochemistry/ Structural Biology

K. Braig, Z. Otwinowski, R. Hegde, D.C. Boisvert, A.Joachimiak, A.L. Horwich, P.B. Sigler, "The crystal structure of the bacterial chaperonin GroEL at 2.8 angstroms," Nature, 371:578-86, 1994. (Cited in more than 100 publications as of February 1996) W.A. Fenton, Y. Kashi, K. Furtak, A. Horwich, "Residues in chaperonin GroEL required for polypeptide binding and release," Nature, 371:614-9, 1994. (Cited in more than 60 publications as of February 1996) Comments by Arthur Horwich and Paul Sigler, Yale University School of Medicine and Howard Hughes Medical Institute A 'Mystical Experience': Teams led by Yale researchers Arthur Horwich, left, and Paul Sigler solved the structure of the bacterial chaperonin GroEL. This pair of papers describes the three-dimensional structure of the bacterial chaperonin GroEL, and relates the structural features to the function of GroEL in facilitating protein folding. GroEL is a large, cylindrical protein complex with central channels. These channels are required for folding newly synthesized and newly translocated proteins into their final active forms. How such large structures adjust the conformation of proteins-through the use of energy in the form of adenosine triphosphate -- has been of major interest to the biological community over the last few years. The bacterial chaperonin GroEL has become the most popular target for understanding how these components work. Early electron microscopy (em) studies detected the large central channel in GroEL and showed that proteins that were not folded could bind inside this channel. Now, through the collaboration of two Yale University research labs, the atomic-level details of GroEL structure and functional implications have been elucidated. "The production of decent, well-diffracting crystals was a major problem," comments Arthur Horwich, a professor of genetics at Yale University School of Medicine and an associate investigator of the Howard Hughes Medical Institute. "We, like others, tried the normal version of GroEL from E. coli, but also versions of GroEL from special high-temperature bacteria, thinking that these might make better crystals. When these approaches failed, we got desperate and tried setting up all the different mutant forms of the E. coli protein that were sitting in the lab. One of them, containing two benign mutations, worked." This was the contribution of one of the paper's first authors, Kerstin Braig, a postdoc in Horwich's lab. Determining the structure of GroEL was also a major task. As Paul Sigler, a professor of molecular biophysics and biochemistry at Yale and a Howard Hughes Medical Institute investigator, comments: "We needed a crackerjack postdoc to solve the structure, and Zbyszek Otwinowski rose to the occasion." Sigler notes that GoeEL's subunits are large and parts of them are not well tied-down, and are thus difficult to visualize crystallographically. However, he says, "we reckoned this flexibility is part and parcel of how these molecules work, so our difficulties were telling us much of the take-home lesson." Horwich comments: "When we first saw the solved structure even at low resolution, it was like a mystical experience. Here it was, with the giant central channel, 45 angstroms in diameter." Next the two teams conducted a high-resolution analysis of both the structure and function. Sigler's group could quickly see that GroEL's domains would have individual roles and that the whole assembly could also work as one unit. Horwich's group, along with Yale geneticist Wayne Fenton, attacked GroEL's structurally interesting features. They found amino acid residues critical to protein binding; ATP hydrolysis; and binding of the cooperating single-ring "lid" protein called GroES. "Finding the peptide-binding region was the most exciting problem," comments Horwich. "From EM work we and others surmised that peptides would be bound inside the channel, but 'How deep?' was the question. We started to look at the underside of the so-called apical domain that forms the end of the cylinder, but that didn't identify any mutations affecting binding. Then, we looked harder at electron micrographs of Helen Saibil and colleagues [from the department of crystallography at Birkbeck College in London; H. Saibil et al., Current Biology, 3:265-73, 1993], and there was density in their reconstructions that appeared to be bound protein just inside the terminal ends of the channel. "We mutated those residues. There they were -- all hydrophobic residues that were required for binding non-native proteins. So it seemed that non-folded proteins must be recognized by binding to this greasy surface of GroEL through their own exposed greasy surfaces, which would become buried in the native forms." Sigler notes, "This first structure [of GroEL] will facilitate the structural analysis of its more complicated functional states. With the work on structures, biochemistry, and genetics at Yale and other institutions, we will probably be able to figure out how this interesting [molecular] machine works." Neeraja Sankaran is a science writer based in the Los Angeles area. The articles listed here, all less than two years old, have received a substantially greater number of citations than others of the same type and vintage, according to data from the Science Citation Index of the Institute for Scientific Information, Philadelphia. Why have these research reports become such standouts? In conversations with The Scientist, the authors attempt to provide an answer.

Biochemistry/ Structural Biology

Interested in reading more?

Become a Member of