For the first time, scientists have determined the structure of the initiator of bacterial DNA replication. It is already known that such replication is controlled by a protein known as DnaA, a member of the AAA+ superfamily of ATPases. What has now been discovered is that the core of the initiator is not the closed-ring structure expected for this system. Instead, DnaA forms an open right-handed helix. In addition, the architecture indicates that this AAA+ superhelix will wrap coils of the DNA around its exterior, causing the DNA double helix to deform as a first step in the separation and unwinding of its strands. Eukaryotic and archaeal initiators also have the structural elements that promote open-helix formation, indicating that a spiral, open-ring AAA+ assembly is a conserved element from a common evolutionary ancestor of Archaea, Bacteria, and Eukarya.

When We Were One… Billions of years ago, life was simple. There were no plants, no animals, no bacteria. Life wasn't even a single cell. It was a "soup" of small free-floating molecules. The famous 1953 experiment by Stanley Miller and Harold Urey, in which they made amino acids by applying sparks to a test tube of hydrogen, methane, ammonia, and water, gave rise to this "primodial soup" theory of life. Small molecules eventually came together and formed complex molecules. Eventually the first cell was formed. It is uncertain exactly how that first cellular life form came into being. However, at the beginning was replication. And recently, scientists identified one of the first tools of replication, a helical substructure within a superfamily of proteins called AAA+, which spans all three domains of life—Archaea, Bacteria, and Eukarya. Studies of the bacteria Aquifex aeolicus and Escherichia coli, Drosophila melanogaster (the fruit fly), and earlier research identifying AAA+ proteins in archaeal organisms indicate that DNA replication is highly conserved, that is, it comes from the last common ancestor of all extant life. Understanding replication in simpler life forms will help us understand such mechanisms in higher-order life forms. It may also help us understand how these mechanisms have been conserved and may bring us closer to pinpointing how the ability of cells to duplicate their genomes came into being.

Structure of ATP-DnaA, which forms a right-handed helical filament with 8₁ symmetry. (a) Overlay of the four DnaA monomers in the asymmetric unit. The AAA+ module is colored green and red (domains IIIA and IIIB, respectively); domain IV, the DNA-binding element, is yellow. The ATP analog AMP-PCP is shown as black sticks. The helix-turn-helix (HTH) motif in domain IV and the N and C termini are labeled. (b,c) Side and axial views of four symmetry-related DnaA tetramers are shown. Monomers are colored by domain as in (a).

At the beginning of replication, ATP binds with DnaA, causing it to change from a monomer to a large oligomeric complex consisting of DnaA monomers bound to a series of DnaA "boxes" (9-base-pair sequences). Although this DnaA/DnaA-box interaction is highly conserved in all bacteria, the mechanism by which ATP activates DnaA oligomerization has been poorly understood. However, University of California, Berkeley, researchers using ALS Beamline 8.3.1 have determined the structure of ATP-bound DnaA from the bacterium Aquifex aeolicus. Using data collected from a single crystal, they assembled a high-resolution model of an ATP-bound DnaA molecule using an ATP analog (AMP-PCP). Each asymmetric unit contains four structurally similar DnaA molecules arranged in a head-to-tail manner.

Several important pieces of information came out of this study. First, by comparing the ATP-DnaA binding pocket to those of other AAA+ assemblies, it was determined that the ATP-DnaA active site is in a closed configuration, which allows it to bind nucleotides using conserved residues from neighboring AAA+ protomers. The precise geometry of these contacts depends on the subunit arrangement arising from the ATP-DnaA spiral, strongly supporting the physiological relevance of the helical filament architecture.

The researchers also discovered that a V-shaped α-helical steric wedge protrudes away from the core AAA+ fold and reorients the AAA+ interface, preventing a flat-ring assembly. This establishes the unique DnaA superhelix. While the steric wedge directs this helical architecture, an ATP-specific conformational change within the AAA+ domain accommodates and stabilizes subunit-subunit interactions necessary to support oligomer formation at replication origins. This rearrangement adjusts for internal incompatibility and is crucial for filament creation.

The initiator helical insert drives filament formation. Comparison of ATP-DnaA helix with a classic AAA+ protein such as NSF, which is a closed-ring assembly, reveals a novel AAA+ assembly mode. The central symmetry axes of each assembly are depicted as rods. Helices α3 and α4 of DnaA (a) form a V-shaped steric wedge that blocks assembly of DnaA domains into a closed, planar array like NSF (b).

The mechanism that allows DnaA to render an origin competent for replisome activity is the linker helix in domain IV of ATP-DnaA. This helix is bent at a ≥40° angle, and the ATPase and DNA binding domains of DnaA are tethered but conformationally uncoupled, allowing the origin DNA to wrap around the outside of a DnaA core (consistent with previous E. coli modeling studies).

Understanding replisome initiation allows us to start answering some long-held questions and start asking some new ones. Although the DnaA/DnaA-box interaction is highly conserved in all bacteria, the origins of different bacteria vary greatly. How does a conserved initiator protein accommodate this origin heterogeneity? A filamentous DnaA assembly provides a ready mechanism, as it could grow or shrink at either end depending on the size and organization of the origin.

The DnaA filament in the context of the nucleoprotein complex. (a) DNA engagement by oligomerized DnaA requires a rotation of the DNA-binding domain (yellow) about the hinge at the base of the connector helix from its position in the filament (gray). DNA is modeled onto domain IV. (b) The outward rotation orients DNA on the outside of the helical assembly, as predicted from electron microscopy (EM) and DNA-protection studies. (c,d) Modulation of filament size enables the engagement of origins of different lengths with highly diverse numbers of DnaA boxes (shown in red; orientations indicated by arrows), exemplified by the E. coli and Aquifex origins.

The structural similarities of AAA+ in archaeal, eukaryotic, and bacterial initiators further suggest these proteins share substantial mechanistic properties. A recent classification of AAA+ proteins reveals that all initiators share a specific helical insert in their ATPase cores. We must then ask, do the AAA+ domains of archaeal and eukaryotic initiators assemble in a manner similar to that of bacterial ATP-DnaA? Recent EM reconstructions of the Drosophila melanogaster origin recognition complex reveal a helical feature within the initiator that accommodates a five-subunit, DnaA-like AAA+ assembly. This finding indicates that the DnaA oligomer is a useful model to further our understanding of higher-order initiator architecture and function.

Research conducted by J.P. Erzberger and M.L. Mott (University of California, Berkeley) and J.M. Berger (University of California, Berkeley, and Berkeley Lab).

Research funding: G. Harold and Leila Y. Mathers Charitable Foundation and the U.S. National Institutes of Health. Operation of the ALS is supported by U.S. Department of Energy, Office of Basic Energy Sciences (BES).

Publication about this research: J.P. Erzberger, M.L. Mott and J.M. Berger, "Structural basis for ATP-dependent DnaA assembly and replication-origin remodeling," Nat. Struct. Mol. Biol. 13, 665 (2006).