Our research focuses on DNA/protein interactions with an emphasis on the DNA integration reaction of HIV and other retroviruses as a model system. Retroviruses integrate a DNA copy of their genome into chromosomal DNA of the host cell as an obligatory step in their replication cycle. The viral integrase protein carries out the key steps of the DNA integration reaction and a major focus of our research is understanding this reaction at the molecular level. In vivo, the viral DNA copy to be integrated into cellular DNA is made by reverse transcription of viral RNA within the cytoplasm of the infected cell. Reverse transcription occurs within a large nucleoprotein complex derived from the core of the infecting virion. This complex includes reverse transcriptase and integrase together with other viral and host proteins. The newly synthesized viral DNA remains associated with this complex, which is called the preintegration complex. We are studying the components and organization of MLV preintegration complexes to understand their role in the retroviral replication cycle. In particular, we are studying the mechanism by which a cellular protein (BAF ) blocks self-destructive autointegration of retroviral DNA and the function of this protein for the host cell.
HIV integrase carries out the key DNA cutting and joining steps
involved in DNA integration. In the first step, called 3' end processing, two
nucleotides are removed from each 3' end of the initially blunt-ended viral
DNA. In the next step, called DNA strand transfer, the 3' hydroxyls at each
end the processed viral DNA attack a pair of phosphodiester bonds in the target
DNA. In the case of HIV, the sites of attack on each target DNA strand are separated
by five nucleotides. The resulting integration intermediate can be repaired
by cellular enzymes to complete the integration process.
HIV
integrase is comprised of three domains that are structurally and functionally
distinct. The structure of the central core domain (shown on the left) has been
solved by X-ray crystallography. It contains a triad of acidic residues (shown
in red) that form part of the active site. This central catalytic domain is
flanked by the N-terminal and C-terminal domains. The structures of these two
domains have been solved by NMR spectroscopy. The N-terminal domain has a three-helix
bundle structure that is stabilized by the binding of zinc. The C-terminal domain
has a similar fold to that of the SH3 domain of spectrin. Although the structures
of all three domains of HIV integrase have now been determined as isolated domains,
we still no little of how they interact with each other and with DNA substrate
to carry out catalysis of 3' processing and DNA strand transfer, the key biochemical
steps of retroviral DNA integration. Understanding these interactions is a major
goal of our current and future research
We
are studying preintegration complexes isolated from cells infected with Moloney
murine leukemia (MLV) virus to investigate aspects of retroviral DNA integration
that have not yet been reproduced in simplified in vitro systems with purified
integrase protein and synthetic DNA substrates. Preintegration complexes contain
a double strand DNA copy of the viral genome together with a number of viral
and cellular proteins. These complexes efficiently integrate their DNA into
an exogenously added target DNA in vitro. We have developed a novel footprinting
technique to probe the nucleoprotein organization of MLV preintegration complexes.
Several hundred base pairs at each end of the viral DNA were found to be organized
in a higher order nucleoprotein complex, which we call the intasome This complex
is not formed when preintegration complexes are made by infection with integrase-minus
virus, demonstrating the involvement of integrase in the complex. Functional
interference experiments demonstrate that the integrity of the complex is required
for normal intermolecular integration into a target DNA.
The viral DNA within preintegration complexes efficiently integrates into an exogenously added target DNA in vitro, but the viral DNA itself is refractory as a target. Recent work has focused on understanding the mechanism of this protection against autointegration. We have previously shown that treating the preintegration complexes with high salt can abolish the protection against autointegration. After separating the stripped complexes from free proteins by gel filtration, the barrier to autointegration can be restored by incubation with a cytoplasmic extract of uninfected NIH3T3 fibroblasts. We have used this reconstitution assay to purify this barrier to autointegration factor (BAF) from NIH 3T3 cells.
BAF
is a 10 kDa protein with an amino amino acid sequence that does not match any
previously identified protein, although a search of EST databases reveals that
many species including humans, zebra fish and C. elegans express a transcript
that encodes a highly conserved homologue. The human homologue of BAF has been
cloned and comparison of its amino acid sequence with its murine counterpart
reveals an 86/89 amino acid identity. BAF is a non-specific DNA binding protein
that bridges together segments of double stranded DNA. Our current model for
the mechanism of blocking autointegration proposes that the DNA-bridging property
of BAF compacts the viral DNA, making inaccessible as a target for DNA integration,
hence promoting proper intermolecular integration into cellular DNA. BAF is
not present in virions and must therefore be recruited by the preintegration
complex from the cytoplasm after viral entry. We are studying the interactions
between BAF and DNA and the role of BAF for the host cell.
Craigie Home Page / Recent Publications / Research Facilities / Lab Personnel / Postdoctoral Positions / Back to top of page
Last updated 12th August, 2002
