SUNY at Albany
June 19-23, 2001
Unifying Themes in the Enzymatic Polymerization of DNA
DNA polymerases faithfully copy a DNA template by the stepwise addition of nucleotides onto the terminal 3'-OH of a DNA primer. These enzymes have vastly different sizes, cellular roles, replication fidelities, and sensitivities to nucleotide analogs. Until recently, it appeared that the sole unifying theme among all DNA polymerases was a common chemistry for nucleotidyl transfer that utilizes bound divalent metals and several clusters of conserved residues. However, the structural portraits of DNA polymerases from four different families have revealed similar silhouettes, featuring a U-shaped DNA binding cleft that resembles a partially opened right hand with fingers, thumb, and palm subdomains, as was originally described by Steitz and coworkers for the Klenow fragment of Escherichia coli DNA polymerase I. Despite differences in their topologies and secondary structures, the emerging data suggest that all DNA polymerases employ analogous catalytic schemes for template-directed DNA synthesis. This common strategy for DNA synthesis is evidenced by a similar geometry of the bound DNA, nucleotide, and metals in ternary complexes of DNA polymerases from different families. The structures feature a conserved active site geometry that includes a kink in the 5'-end of the template strand and, most strikingly, a large conformational change in the fingers in response to substrate binding. Recent single molecule experiments show that stretching the DNA template markedly alters the rate of DNA synthesis catalyzed by various DNA polymerases. These data have been interpreted as evidence that T7 DNA polymerase and two related enzymes convert anywhere from two to four (n = 2-4) single-stranded template bases to double helix geometry in the active site during each catalytic cycle. We show why structural data are more consistent with only one base shortening from ssDNA to dsDNA geometry during the conformational change from an open to closed complex. We present model calculations for the observed variation of replication rate by mechanical tension as an explicit function of a few structural parameters. Unlike previous phenomenological models which relied upon experimental force-extension curves for DNA, our model does not presume equilibrium within the DNA during replication and considers the rate to be modulated solely by the nanoscale neighborhood of the enzyme. Our theoretical analysis, with n=1, reconciles the results of recent single molecule experiments with structural data on DNA polymerases.
Anita Goel (1a,b), Sylvie Doublié (2) Maxim D. Frank-Kamenetskii (3), Dudley Herschbach (1b), and Tom Ellenberger (4)
Department of Physics (1a) and Harvard-MIT Joint Division of Health Sciences and Technology and Department of Chemistry (1b) and Chemical Biology, Harvard University, Cambridge, MA 02138, USA,