Taking structural biology to a new plateau
The coiled-coil tails of Rad50 proteins link to each other with zinc hooks.
November 4, 2002—Less than a year ago the National Cancer Institute (NCI) launched a multidisciplinary "program project" named Structural Cell Biology of DNA Repair Machines (SBDR), aimed at elevating the understanding of DNA repair from the level of individual protein structures to the level of complex interactions within the cell.
On August 1, 2002, Nature published the project's first major result, the unexpected discovery of the Rad50 "zinc hook," a metal-mediated interface that the essential Mre11/Rad50 protein complex uses to link broken DNA strands so that they can be rejoined.
SBDR is a $18 million program involving 18 investigators at 11 institutions, pursuing five research programs in three major areas of DNA repair. Based at Berkeley Lab, SBDR's principal investigator is John Tainer, a professor of molecular biology at the Scripps Research Institute, with a visiting faculty appointment in Berkeley Lab's Life Sciences Division. Co-principal investigator is Priscilla Cooper, acting director of the Life Sciences Division.
"We believe that SBDR will be powerful for providing insights on the function of dynamically assembled DNA repair machines," says Tainer. "These insights absolutely require advanced technologies that bridge the gaps from molecular structure to assemblies to repair networks to cell biology. Such a combination of technologies cannot be found in individual laboratories but instead requires a highly collaborative effort."
Understanding complex molecular machines requires an arsenal of techniques, among them traditional molecular biology and biochemistry laboratory methods, electron microscopy, mass spectrometry, and the use of synchrotron radiation.
An essential component of SBDR will be the versatile SIBYLS beamline funded by the Department of Energy's Office of Science. SIBYLS, an acronym for "structurally integrated biology for life sciences," will have the unique capability of performing both x-ray crystallography, to determine the structure of crystallized proteins, and small-angle x-ray scattering (SAXS), which studies proteins in solution, conditions more nearly resembling their natural environment. When the crystal structures of individual proteins are known, SAXS can sometimes be used to reveal how they work together in complexes; it can also be an important tool for gaining structural information on impossible-to-crystallize proteins.
Two of the many ways the Mre11/Rad50 protein complex could help repair damaged DNA: (A) recombination with a sister chromatid; (B) bringing together broken double strands.
Life under siege
DNA is incessantly attacked outside and inside the organism by radiation, toxic chemicals, and other factors, and can also be damaged by accidents of cell division. Some of the results are chemically corrupt single bases, groups of missing or altered bases, the wrong base pairs in a gene sequence, and broken single or double strands of DNA. Because damaged DNA contributes to birth defects, developmental abnormalities, cancers, and other diseases, constant repair of DNA is one of the most basic functions of life.
Different proteins act together along different pathways to repair different kinds of damage; some groups have been doing DNA repair for millions or billions of years and are found in organisms from bacteria to humans. While the specific proteins involved in a particular repair process are not always identical among different organisms, their group geometries and functional "strategies" are often similar.
"Previously DNA repair was thought of as a collection of clean, neat, separate pathways, but now we know that these pathways are interwoven," says Susan Tsutakawa of the Life Sciences Division, a crystallographer in SBDR's molecular biology lab. "Proteins that were thought to have a role in only one pathway are now understood to be involved in many."
The proteins Mre11 and Rad50, for example, form repair complexes found in all kingdoms of life; the Mre11 complex's remarkably diverse functions include roles in cell division, maintenance of chromosome-capping telomeres, and repairing double-stranded breaks in DNA in two different ways. The August 1 Nature paper by Tainer and his colleagues compares the Mre11 complex in eight organisms: a virus, an archaeon, a bacterium, two yeasts, a nematode, a plant, and humans. They found a new and surprising mechanism by which the Mre11 complex accomplishes DNA repair.
Rad50 tails link together when four cysteine amino-acid residues in the CXXC motifs (two cysteines each, in a pair of Rad50 tails) bind to a single, doubly ionized zinc atom.
The ties that bind
To repair a double-strand break, the ends of the broken strands must be stripped of damaged or redundant bases, then brought together with a new length of DNA. The new length may be assembled intact from a nearby "sister" chromatid—the adjacent arm of the chromosome—or it may involve an imperfectly matched length of DNA in a process called nonhomologous end-joining. In both processes, the preparatory work on a broken strand is done by a complex of two Mre11s and two Rad50s that form a "binding head."
The binding head is compact, but each of its Rad50s has a long tail, extending up to 600 angstroms in mammals, which consists of a coil of amino acids coiled back around itself—like a twisted fiber pinched in the middle, its two halves wound around each other to form a loose piece of twine.
At the pinch in the middle of the primary coil, where the tail bends back on itself, there is a sequence of four amino acid residues similar in all the organisms studied. The first and fourth residues in this sequence are always cysteines, with various other residues in the second and third places—thus the label "CXXC motif."
Through x-ray crystallography the researchers learned that the CXXC motifs were shaped like hooks, with the potential to grapple the tails of other Rad50 proteins. Using laboratory techniques of gel filtration and ultracentrifugation, the group confirmed that successful linkage of two Rad50 proteins depends on the presence of zinc.
Chemical and structural studies showed that four cysteines (two each, in a pair of Rad50 tails) bind to a single, doubly ionized zinc atom. Discovery of this zinc-hook mechanism immediately suggested ways the Mre11 complex could connect DNA strands and bring them together.
Electron micrographs show two possible structures formed by Rad50 zinc hooks: (A) two Mre11/Rad50 complexes could link their tails and attach to separate double strands of DNA with their binding heads; (B) a single Mre11/Rad50 complex, its twin tails linked in a circle, joins a broken double strand.
A biological multitool kit
In a series of compelling electron micrographs of human and archaean Mre11 complexes, the researchers found several different conformations of linked Rad50. Some pairs were joined by their tail hooks to form a bola shape, with the binding heads of the Mre11 complex at both ends.
In one micrograph, the twin tails of a single binding head link to each other, forming what looks like a finger ring, with the binding head as its stone; an even more unusual micrograph shows a single-headed complex of this kind actually bound to a length of DNA.
These conformations suggest distinct mechanisms by which linked Mre11 complexes can bridge sister chromatids or perform nonhomologous end-joining, and how even a circular, single-headed Mre11 could bring together two broken DNA ends.
"Identification of the Rad50 zinc hook will allow us to understand the mechanism for dynamic assembly and disassembly of the Mre11 complex," says Tainer, "which is critical for the repair of double-strand breaks in DNA and thus for the avoidance of cancer-causing mutations."
The discovery is exciting not only for the questions it answers but for the new questions it raises. Do binding heads begin their work separately, thrashing their tails about until they grapple another bound complex that's likely to help complete the repair? Do they link tails first and search together for broken DNA to repair?
Other models are possible; which are correct and how the process works in detail are yet to be determined, but other coordinating proteins and interactions are likely to be involved. Thus finding the correct model requires the complete NCI-funded team.
SBDR's leaders emphasize the cyclical nature of their work, first drawing on genome sequencing and biochemical and genetic studies to suggest the proteins most worth studying, then determining the structures of these proteins through x-ray crystallography and other methods, in search of understanding of how they work together to accomplish complex biological tasks.
This knowledge, in turn, suggests biological experiments that can give clues to the make-up and mechanisms underlying the complex machines that mediate DNA repair, filling in more details of their intricate interactions—a new kind of research design cycle that provides a novel prototype for productively linking structure and chemistry with biology.—by Paul Preuss
Media contact: Paul Preuss, LBNL Public Information Dept, (510) 486-6249, firstname.lastname@example.org
Technical contact: Priscilla Cooper, LBNL Life Sciences Division, (510) 486-7346, PKCooper@lbl.gov
Related Web Links
"The Rad50 Zinc-Hook is a Structure Joining Mre11 Complexes in DNA Recombination and Repair," Karl-Peter Hopfner, Lisa Craig, Gabriel Moncalian, Robert A. Zinkel, Takehiko Usui, Barbara A. L. Owen, Annette Karcher, Brendan Henderson, Jean-Luc Bodmer, Cynthia T. Mcmurray, James P. Carney, John H. J. Petrini & John A. Tainer, Nature 418, 562 - 566, August 1, 2002 [subscription required]
Program Project in Structural Cell Biology of DNA Repair Machines: Overview
Small Angle X-Ray Scattering at the SIBYLS Beamline
John A. Tainer, Ph.D.: Structural Molecular Biology and Protein Design, Scripps Research Institute
LBNL Life Sciences Division
Advanced Light Source
Funding: Key funding was provided by the U.S. Department of Energy's (DOE's) Office of Science and the National Cancer Institute (NCI), the American Cancer Society, the Canadian Institutes of Health Research, National Institutes of Health, Human Frontiers Science Program, Skaggs Institute for Chemical Biology, the National Institutes of Mental Health, and The Swiss National Science Foundation.
Lawrence Berkeley National Laboratory: The advantages of basing the SBDR Program at LBNL come directly from the broad base of biological expertise in DNA damage and repair in the Life Sciences Division, coupled to the structural expertise and advanced biophysical technologies at the Advanced Light Source. The established strength of Berkeley Lab in mounting and sustaining large multidisciplinary programs provides an added major advantage.
LBNL is a multiprogram national laboratory where research in advanced materials, life sciences, energy efficiency, detectors and accelerators serves America's needs in technology and the environment. LBNL is managed by the University of California for the DOE.
Author: Paul Preuss has been a science writer at Lawrence Berkeley National Laboratory for the last six years, covering the broad range of research conducted by the Lab, from astrophysics to earth sciences to chemistry and materials sciences to genetics to cell biology, etc. Before joining LBNL, Preuss spent 20 years as a novelist, writing mostly science fiction, even collaborating with Arthur C. Clarke on his "Venus Prime" series, which is still prominent on Amazon.com. Preuss said, "I've stopped writing science fiction. The truth is that real science is far more fascinating. The fictional stuff is a pale reflection." For more science news, see LBNL's Science Beat.