Our general long-term goal is to determine the detailed molecular mechanisms by which the proteins and nucleic acids involved in the central dogma of molecular biology (DNA replication, transcription, and translation) achieve their biological functions. Virtually all aspects of the maintenance and expression of information stored in the genome involve interactions between proteins and nucleic acids. Over the past three decades we have obtained detailed structural insights into the mechanisms by which specific proteins and nucleic acids catalyze and control the fundamental processes of DNA replication, mRNA synthesis, and protein synthesis, as well as DNA recombination.
To establish the structural basis of DNA replication, we have been studying DNA polymerases and associated proteins involved in replication. Following on our earlier structures of Escherichia coli DNA polymerase I Klenow fragment and Thermus aquaticus DNA polymerase and their DNA substrate complexes, we established the crystal structure of a replicative DNA polymerase (from phage RB69) that is homologous to the eukaryotic B-family polymerases. The structures of the RB69 polymerase complexed with duplex DNA substrate, bound both at the editing site and the polymerase site, and of the sliding clamp complexed with a polymerase carboxyl-terminal peptide allowed construction of a replisome core structure. These structures showed that this macromolecular machine, charged with the responsibility of faithfully copying the DNA genome, undergoes large conformational changes throughout its catalytic cycle. These changes are associated with the enzyme's fidelity-enhancing mechanisms and translocation.
We have also determined the structures of another B-family DNA polymerase, φ29 DNA polymerase, which initiates replication by attaching the first nucleotide of the phage genome to a serine side chain of protein primer called terminal protein, as well as its binary and ternary substrate complexes. The basis of DNA strand displacement activity exhibited by this enzyme is explained by the template strand passing through a tunnel that is too small to accommodate the nontemplate strand that is displaced. The extreme processivity of this polymerase is explained by its topological encirclement of the substrate and product DNA at the active site. The structure of the φ29 DNA polymerase bound to terminal protein provides the first structural insights into the mechanism of protein-primed DNA replication, suggesting that a four-helix domain containing the priming serine must back out of the duplex DNA product-binding site as DNA synthesis proceeds.
Toward our goal of understanding eubacterial replication, we determined the structure of the T. aquaticus DNA polymerase III, which we discovered exhibits no similarity to that of the archaeal or eukaryotic replicating polymerases, but rather possesses a catalytic domain that is homologous to that of repair DNA polymerase β. Furthermore, our structure of a ternary complex of Pol III with substrates shows that the DNA and nucleoside triphosphate (NTP) substrates bind identically in these two polymerases. The possibility that the last common ancestor had a ribozyme-replicating polymerase is thus raised. Our structures of the hexomeric DnaB helicase and its complex with the helicase-binding domain of primase are beginning to illuminate the structural bases of primosome function.
Transposable elements encode recombination proteins that catalyze recombination of DNA at specific sequences. Our structure of a γδ resolvase synaptic tetramer bound to two DNA duplexes captured in an intermediate state of the recombination process shows a cleaved DNA substrate covalently linked to the protein, with the ends to be recombined separated by 50 Å. The very flat interface between the protein dimers linked to the DNAs to be recombined suggests that recombination is achieved by an unprecedented 180° rotation of one dimer relative to the other. Our recent structure of a synaptic tetramer of the homologous Gin recombinase exhibits a state in which one dimer is in a rotated position relative to the other dimer when compared with their orientation in γδ resolvase, consistent with the rotation hypothesis.
Genes encoded in DNA are transcribed into mRNA by DNA-dependent RNA polymerases that can initiate RNA synthesis at a specific promoter sequence. To understand this process and its regulation and to explain how RNA polymerases differ from DNA polymerases, we have determined the crystal structures of several T7 RNA polymerase complexes with promoter DNAs, mRNAs, and incoming NTP. These structures show how portions of the RNA polymerase recognize the bases in the duplex DNA promoter and denature part of the promoter to form a transcription initiation bubble. In an initiation complex, three nucleotides of transcript are seen base-paired to the template strand. We have also captured this polymerase in a transcription elongation phase as a complex with 30 base pairs of DNA and a 17-nucleotide RNA transcript. The transition from the initiation to the elongation phases of transcription is accompanied by a massive structural rearrangement of the amino-terminal domain, which eliminates the promoter DNA-binding site on the enzyme and creates a tunnel through which the transcript exits the enzyme, thus explaining the high processivity of the elongation phase. Our recent structures of initiation complexes with either a 7- or 8-nucleotide transcript show intermediates in this structural transition in which the promoter binding domain rotates by 45° to accommodate the growing transcript.
The structures of T7 RNA polymerase elongation complexes captured at each step of nucleotide incorporation show a 22° rotation of a five-helix subdomain upon NTP binding and upon pyrophosphate release. The conformational change that accompanies pyrophosphate release produces both the translocation of the product heteroduplex and the strand separation of downstream duplex DNA.
Our structural studies of the proteins and nucleic acids involved in translating the gene sequence carried in the messenger RNA into the protein products are providing insights into the translation of the genetic code. This includes our earlier structural studies of aminoacyl-tRNA synthetases, as well as more recent structural studies explaining how the CCA-adding enzyme is able to mature or repair the 3' CCA end of tRNA without using a nucleic acid template. We have established the structures of the CCA-adding enzyme captured in the steps of adding penultimate C and final A as well as the product tRNA.
We have been pursuing high-resolution structural studies of the machine that synthesizes proteins, the ribosome, and have determined the atomic structure of the 1.6-mDa ribosomal subunit that catalyzes the formation of the peptide bond. The structures of the large subunit with either substrate or product analogs bound to the active site of peptide synthesis show a peptidyltransferase center that is made entirely of RNA. Ribosomal RNA positions the substrate α-amino group appropriately for attack of the peptidyl-tRNA, and it also interacts with the latter's A76 2'-OH group, which may function as a proton shuttle between the α-amino group and the A76 3'-OH. More recently, we have obtained the structure of the 70S ribosome complexed with fMet-tRNA in the P site and an essential protein factor EF-P that is seen to be interacting with the tRNA and a rearranged L1 stalk, suggesting that it may be stimulating the first step of protein synthesis by correctly positioning the fMet-tRNA in the P site.
We have also established the structures of nearly two dozen different antibiotics that target the large ribosomal subunit in complex with the large subunit as well as complexes with a T. thermophilus 70S ribosome, including two members of the tuberactinomycin family of antibiotics that are used to treat tuberculosis. These structures not only establish how these antibiotics stop peptide synthesis but also are providing the basis for structure-based design of new antibiotics (by Rib-X Pharmaceuticals, Inc.) that are effective against ribosomes containing antibiotic-resistance mutations.
Portions of this work were supported in part by grants from the National Institutes of Health and the Agouron Institute.
As of February 09, 2011