Laurence Hurley

PMID: 18355457;PMCID: PMC2585383;Abstract:

In its simplest form, a DNA G-quadruplex is a four-stranded DNA structure that is composed of stacked guanine tetrads. G-quadruplex-forming sequences have been identified in eukaryotic telomeres, as well as in non-telomeric genomic regions, such as gene promoters, recombination sites, and DNA tandem repeats. Of particular interest are the G-quadruplex structures that form in gene promoter regions, which have emerged as potential targets for anticancer drug development. Evidence for the formation of G-quadruplex structures in living cells continues to grow. In this review, we examine recent studies on intramolecular G-quadruplex structures that form in the promoter regions of some human genes in living cells and discuss the biological implications of these structures. The identification of G-quadruplex structures in promoter regions provides us with new insights into the fundamental aspects of G-quadruplex topology and DNA sequence-structure relationships. Progress in G-quadruplex structural studies and the validation of the biological role of these structures in cells will further encourage the development of small molecules that target these structures to specifically modulate gene transcription. © 2008 Elsevier Masson SAS. All rights reserved.

Abstract:

As expected, Bizelesin, which is a biscyclopropa[c]pyrrolo[3,2-e]indol-4(5H)-one [(+)-CPI]-derived DNA-DNA cross-linker, has a high interstrand cross-linking reactivity with the palindromic sequence 5'-d(CGTAATTȦCG)2. Contrary to expectations, the target duplex is rearranged to yield two products: one (major product) contains an AT step wherein both adenines are syn-oriented and hydrogen bonded to thymines forming a stable Hoogsteen base-paired region flanked by Watson-Crick base-paired regions (5HG); the other (minor product) contains anti-oriented AT-step adenines that show no evidence of hydrogen bonding with pairing thymines (5OP) in an otherwise normally base-paired duplex. In another unexpected outcome, the reaction of two 'uncoupled' monoalkylating (+)-CPI 'halves' of Bizelesin with the same duplex alkylates same-strand adenines three base pairs apart [5'-d(CGTAȦTTȦCG)2] rather than the anticipated opposite-strand adenines six base pairs apart (which would mimic Bizelesin). To probe the molecular mechanism that leads to Bizelesin's unusual DNA rearrangement, which appears to be a requirement for DNA-DNA interstrand cross-linking, we have carried out conformational exchange analyses (NOESY and ROESY) and restrained molecular dynamics simulations of these adducts. These studies suggest that Bizelesin controls the rearrangement of the six-base-pair target prior to cross-linkage and restricts the cross-linked DNA adduct's range of motion, freezing-out adduct conformers defined by alternative drug-DNA hydrogen-bond regimes. The two competing cross-linkage pathways share a common first step, the opening of the central AT-step base pairs, an event that is facilitated by the energetics of monoadduct-induced DNA bending distortion. One pathway (to 5HG) stabilizes these open bases by reorganizing the AT-step region into two Hoogsteen base pairs, the thymine bases of which also hydrogen bond with Bizelesin's ureadiyl subunit. A second pathway (to 5OP) directly stabilizes the open bases by forming a hydrogen-bonding complex between the AT-step thymines and Bizelesin's ureadiyl subunit. Cross-linked DNA motion drives both of the 5HG and 5OP adducts from one ephemeral hydrogen-bonding regime to another, a process documented in the NOESY conformational exchange data and simulated in restrained molecular dynamics trajectories. These results, together with the analysis of other six-base-pair Bizelesin cross-linked species, suggest a novel mechanism for sequence recognition by this cross-linker where monoalkylation distortive stress associated with a bent DNA conformation must be dissipated by a cooperative interaction between drug and duplex to produce a straight B-form-like structure before cross-linking can proceed. This example provides a new mechanism for DNA sequence recognition involving a 'drug-induced rearrangement' of DNA that critically depends upon the interplay of drug and sequence recognition elements.

PMID: 8672503;Abstract:

The T-antigen-induced structural changes of the SV40 replication origin were probed with three DNA-reactive antitumor agents: (+)-CC-1065, bizelesin, and pluramycin, (+)-CC-1065 is an N3 adenine minor groove alkylating agent that selectively reacts with AT-rich DNA sequences with a bent conformation; bizelesin also reacts with the minor groove of AT-rich sequences but is selective for a straight DNA conformation. Pluramycin is an intercalative guanine alkylator whose reactivity is increased by unwinding and decreased by compression of the minor and/or major grooves of DNA. We show that while binding of T-antigen reduced the ability of (+)-CC-1065 to alkylate the AT tract in the SV40 replication origin, it did not interfere with bizelesin modification of the same sequence. These unexpected results suggest that when T-antigen binds to the SV40 origin the AT tract is in a straight DNA conformation. High-resolution DNase I footprinting experiments indicate that at least three helically in-phase T-antigen binding sites exist in the GC box region located immediately downstream of the AT tract. The binding of T- antigen enhances the reactivity of (+)-CC-1065 to the two 5'-AGTTA* (the asterisk indicates the covalent bonding site) drug modification sites in the GC box region, demonstrating that these sites are in a bent conformation. In contrast, T-antigen inhibited the reactivity of pluramycin at sequences within the GC box region that are known not to bind T-antigen. These data, in combination with the DNase I footprinting results, suggest that T-antigen binding induces a conformational change in the DNA that no longer favors pluramycin intercalation. Based on our results, we propose that T-antigen binds tightly to the upstream region of the AT tract of SV40 replication origin forming double hexamers. In the downstream region, binding of T- antigen to the helically in-phase sites in the GC box region induces DNA bending in the opposite direction of the natural AT tract bending, while simultaneously transforming the naturally bent AT tract DNA into a straight conformation.

PMID: 10956013;Abstract:

The quinobenzoxazines, a group of structural analogues of the antibacterial fluoroquinolones, are topoisomerase II inhibitors that have demonstrated promising anticancer activity in mice. It has been proposed that the quinobenzoxazines form a 2:2 drug-Mg²⁺ self-assembly complex on DNA. The quinobenzoxazine (S)-A-62176 is photochemically unstable and undergoes a DNA-accelerated photochemical reaction to afford a highly fluorescent photoproduct. Here we report that the irradiation of both supercoiled DNA and DNA oligonucleotides in the presence of (S)-A-62176 results in photochemical cleavage of the DNA. The (S)-A-62176-mediated DNA photocleavage reaction requires Mg²⁺. Photochemical cleavage of supercoiled DNA by (S)-A-62176 is much more efficient that the DNA photocleavage reactions of the fluoroquinolones norfloxacin, ciprofloxacin, and enoxacin. The photocleavage of supercoiled DNA by (S)-A-62176 is unaffected by the presence of SOD, catalase, or other reactive oxygen scavengers, but is inhibited by deoxygenation. The photochemical cleavage of supercoiled DNA is also inhibited by 1 mM KI. Photochemical cleavage of DNA oligonucleotides by (S)-A-62176 occurs most extensively at DNA sites bound by drug, as determined by DNase I footprinting, and especially at certain G and T residues. The nature of the DNA photoproducts, and inhibition studies, indicate that the photocleavage reaction occurs by a free radical mechanism initiated by abstraction of the 4'- and 1'-hydrogens from the DNA minor groove. These results lend further support for the proposed DNA binding model for the quinobenzoxazine 2:2 drug-Mg²⁺ complex and serve to define the position of this complex on the minor groove of DNA.