Duplication of DNA, the information repository for life, is a vital process for all cellular organisms. Therefore the replication process must be faithfully performed at the passing of each generation to maintain the species. Duplication of DNA requires that the two strands of the double helix be separated; each strand is then used as a template to produce two new daughter strands from the original parental duplex DNA. Presently, these daughter chromosomes will segregate into two newly formed cells.
Numerous proteins cooperate to perform this delicate and vital task. Initially, several proteins conspire to pry apart the two strands at a position in the chromosome called an origin. A helicase then enters the opening and unzips, or unwinds, the DNA strands, acting with topoisomerases that remove the spiral turns. A multiprotein replicase machinery then coordinates the synthesis of two new strands at the same time.
One of our goals is to understand the functions of each of the numerous proteins involved in the chromosome replication process. Our studies are focused on two different organisms. As a representative of bacteria we study Escherichia coli. As a model for eukaryotes we study Baker's yeast, Saccharomyces cerevisiae. The E. coli replicase, DNA polymerase III (Pol III) holoenzyme, consists of 10 different protein "gears"; some are present in multiple copies, for a total of 17 proteins. Within this machine are two copies of the DNA polymerase, the protein catalyst that synthesizes DNA, using one DNA strand as a template. The presence of two polymerases in one particle enables it to copy both strands of the parental duplex DNA at the same time.
Another subunit, called ε, is an exonuclease that can degrade DNA, thereby proofreading the DNA made by the polymerase. When the polymerase makes a mistake, the exonuclease removes the incorrect nucleotide, allowing the polymerase to try again. The other proteins manipulate the DNA polymerases to ensure efficient and coordinated duplication of the chromosome. For example, the β subunit encircles DNA like a ring and acts as a mobile sliding tether to hold the polymerase to DNA, which enables the replicase to synthesize DNA at a rate of 800 nucleotide units every second. The structure of the β dimer, solved in collaboration with John Kuriyan's laboratory (HHMI, University of California, Berkeley), shows it to be in the shape of a ring, consistent with biochemical studies of β sliding on DNA. The use of a ring-shaped sliding clamp for the polymerase is conserved in eukaryotes. Our collaboration with the Kuriyan lab has led to structures of the human clamp as well as that of T4 phage (a bacterial virus). The eukaryotic sliding clamp is called proliferating cell nuclear antigen (PCNA), as its presence correlates with the proliferative state. Overall, it is remarkable how the sliding clamp has retained its shape throughout evolution. (These studies are supported by a grant from the National Institutes of Health.)
Sliding clamps do not get onto DNA by themselves. For this, a group of five subunits, called γ complex, act as a "clamp loader" to open the β ring, position it onto DNA, and then close it. The eukaryotic replicase also utilizes a five-subunit clamp loader, called RFC (replication factor C). Clamp loaders harness the energy of ATP hydrolysis to drive the clamp-loading reaction. In the E. coli clamp loader, the clamp is cracked open by the δ subunit. The crystal structure of δ•β (solved in collaboration with the Kuriyan lab) illustrates how δ distorts an interface of the β dimer and opens it. The crystal structure of the clamp-loading complex shows the subunits are organized as a circular pentamer that sits on top of the clamp. We have recently solved the crystal structure of the β clamp with a primed DNA; it holds DNA in a highly tilted fashion. The steeply angled DNA will rapidly flip between the two protomers of the dimer, which explains how the single primed site that passes through the ring can switch between two polymerases attached to the same clamp. The β-DNA structure also reveals a binding site on the surface of the clamp for template single-strand DNA (ssDNA), explaining how the clamp may remain at a primed site after clamp loading for use by a DNA polymerase. (This research was supported by a grant from the National Institutes of Health.)
The overall architecture of the proteins at the replication fork, called a replisome, has emerged from these studies (see the movie). At the center is the clamp loader; it has protein arms that extend out the "top" that bind two Pol III cores for simultaneous synthesis of both strands of duplex DNA. The clamp loader arms also bind the helicase, DnaB, a ring-shaped hexamer that encircles one DNA strand. The leading-strand Pol III continuously extends DNA forward as it travels with the helicase during DNA unwinding. Due to the antiparallel structure of duplex DNA, the lagging strand is synthesized in the opposite direction of the leading strand. Thus the lagging strand is made as a series of small fragments, each being started by an RNA primer (e.g., by primase). RNA primers are extended by Pol III, resulting in a DNA loop. When the polymerase finishes, it bumps into a fragment made previously and disengages from its β ring, allowing it to recycle to the next upstream RNA primer onto which the clamp loader has loaded a new clamp. Overall these actions result in collapse of the loop and the start of another. (This research was supported by a grant from the National Institutes of Health.)
We have also studied how the replisome bypasses blocks to the replication fork. Interestingly, DNA lesions can be skipped over, leaving them behind in ssDNA gaps. The process involves loading of a β clamp at a primed site ahead of the DNA lesion. Then the stalled Pol III hops over the lesion to the new clamp. In our collaboration with Myron Goodman (University of Southern California), we find that E. coli contains a specialized DNA polymerase (Pol V) that is induced upon DNA damage and functions with β to extend DNA across a lesion, resulting in a mutation. Two additional DNA polymerases are also induced by DNA damage, Pol II and Pol IV, the functions of which are obscure. We find that Pol II and Pol IV can replace Pol III within a moving replisome and slow the fork, thereby giving time for DNA lesions to be repaired by normal, high-fidelity DNA repair processes, a preferred outcome over mutagenic lesion bypass by Pol V. (These studies were supported by a grant from the National Institutes of Health.)
Studies of the eukaryotic replication proteins are equally fascinating. Exciting studies from other laboratories have identified the proteins that function at eukaryotic replication forks. Interestingly, there exist eukaryotic equivalents for all the E. coli replication proteins, suggesting that the fundamental principles of replication have been conserved during evolution. We are studying the six-subunit ring-shaped eukaryotic MCM helicase, which we have expressed and reconstituted, to determine the order of the six subunits around the ring and to examine its actions during translocation on duplex and ssDNA. Our biochemical studies, and structural studies in collaboration with the Kuriyan lab, reveal striking similarities between the bacterial and eukaryotic clamps and clamp loaders. Unlike E. coli, eukaryotes utilize two different DNA polymerases for leading and lagging strands, Pol ε and Pol δ, respectively, and lagging-strand priming is performed by a four-subunit Pol α/primase that makes a hybrid RNA/DNA primer. Our studies of eukaryotic Pol δ demonstrate exceedingly high processivity with PCNA, and a specific mechanism that enables Pol δ to eject from PCNA when a DNA segment is complete, thereby enabling Pol δ to recycle rapidly during lagging-strand synthesis. PCNA clamps that are left on DNA must be recycled before the intracellular pool of PCNA is depleted. Eukaryotes contain alternative RFC clamp loaders in which the RFC1 subunit is replaced by another protein. We find that the Rad24-RFC alternative clamp loader, which functions with a different clamp, unloads PCNA from DNA. We are also studying the numerous replication proteins that are unique to eukaryotes, with the expectation that they may provide novel functions or add levels of control to the replication apparatus of eukaryotic cells.