Structural basis for allosteric PARP-1 retention on DNA breaks

Abstract

The success of poly(ADP-ribose) polymerase–1 (PARP-1) inhibitors (PARPi) to treat cancer relates to their ability to trap PARP-1 at the site of a DNA break. Although different forms of PARPi all target the catalytic center of the enzyme, they have variable abilities to trap PARP-1. We found that several structurally distinct PARPi drive PARP-1 allostery to promote release from a DNA break. Other inhibitors drive allostery to retain PARP-1 on a DNA break. Further, we generated a new PARPi compound, converting an allosteric pro-release compound to a pro-retention compound and increasing its ability to kill cancer cells. These developments are pertinent to clinical applications where PARP-1 trapping is either desirable or undesirable.

Dynamics of the HD regulatory subdomain of PARP-1; substrate access and allostery in PARP activation and inhibition

“…We propose, therefore, that activation of PARP-1 is in all probability a two-step process, in which first DNA binding and then substrate binding contribute consecutively and additively to successive destabilisation of the folded structure of the HD subdomain, as summarised in Figure 6H. Previous work has shown that DNA binding also causes changes in the HD subdomain, with the strongest effects on NH solvent exchange seen for helices B, D and F ( Figure 1D) (16), while similar effects for EB-47 binding as reported here were found in a parallel HXMS study of allostery involving PARP inhibitors and their influence on the HD subdomain (58). Comparison of the crystal structures of the inhibitor complexes with that of the DNA-bound complex (4DQY) also shows differences, most notably that DNA binding causes helix F to become straight ( Figure 1J,K), whereas inhibitor binding, particularly of EB-47, has the opposite effect of increasing the bend in helix F, with movements in other helices, particularly E, that also differ between DNA and EB-47 binding ( Figures 6A-D).…”

“…Our data show that it is principally the slower conformational exchange processes, those detected by the NH solvent exchange experiments, that show differences amongst the WT protein, mutants and inhibitor complexes, whereas for the fastest processes, those detected mainly by the 15 shows very poor electron density and consistently high B factors throughout (58), while in solution,…”

“…The very different displacements of the HD in these structures demonstrates that crystal packing forces must differ substantially in the two cases. This in turn must presumably underlie another big difference between them, which is that for 6VKQ, but not 7AAB, the quality of the observed electron density for the HD was particularly weak, which led to considerable difficulties in modelling this portion of the structure (58). This is also reflected in the much higher B factors for the HD subdomain than those for the ART subdomain for 6VKQ ( Figure S21); similar effects were seen in the closely related structure of a complex of PARP-1 CAT domain with the inhibitor UKTT15 (6VKO) that was also reported recently (58), whereas for 7AAB (and for the other structures discussed here) the quality of the electron density was essentially normal for both the HD and ART subdomains.…”

“…The SPR data is also relevant to another outstanding question, namely how it is that binding of veliparib causes mild stabilisation of the HD subdomain, and thus leads to allosteric effects in the opposite sense to those caused by EB-47 (prior inhibitor binding causes weaker binding of PARP-1 to DNA damage sites in the case of veliparib, but stronger binding in the case of EB-47) (58). The most noticeable characteristic of veliparib relative to the other inhibitors studied here is its small size; when bound, no part of the veliparib molecule comes close to any part of the HD.…”

Single-molecule measurements reveal that PARP1 condenses DNA by loop formation

“…Much recent work on analysing PARP1 has been stimulated by the discovery that inhibition of PARP activity results in synthetic lethality towards BRCA mutated cancer cells 8,9 . PARP inhibitors trap the protein at sites of DNA damage by inhibiting enzymatic activity and thereby block poly(ADP-ribose) induced unbinding [10][11][12][13] . This leads to increased double-strand breaks via collapse of replication forks 14,15 and confers sensitivity of BRCA mutated cells to PARP inhibitor drugs.…”