HTS discovery of PARP1-HPF1 complex inhibitors in cancer.

PARP1/2 inhibitors (PARPi) are effective clinically used drugs for the treatment of cancers with BRCA deficiencies. PARPi have had limited success and applicability beyond BRCA deficient cancers, and their effect is diminished by resistance mechanisms. The recent discovery of Histone PARylation Factor (HPF1) and the role it plays in the PARylation reaction by forming a shared active site with PARP1 raises the possibility that novel inhibitors that target the PARP1-HPF1 complex can be identified. Herein we describe a simple and cost-effective high-throughput screening (HTS) method aimed at discovering inhibitors of the PARP1-HPF1 complex. Upon HTS validation, we first applied this method to screen a small PARP-focused library of compounds and then scale up our approach using robotic automation to conduct a pilot screen of 10,000 compounds and validating >100 hits. This work demonstrates for the first time the capacity to discover potent inhibitors of the PARP1-HPF1 complex, which may have utility as probes to better understand the DNA damage response and as therapeutics for cancer.

Slow Dissociation from the PARP1-HPF1 Complex Drives Inhibitor Potency.

PARP1, upon binding to damaged DNA, is activated to perform poly ADP-ribosylation (PARylation) on itself and other proteins, which leads to relaxation of chromatin and recruitment of DNA repair factors. HPF1 was recently discovered as a protein cofactor of PARP1 that directs preferential PARylation of histones over other targets by contributing to and altering the PARP1 active site. Inhibitors of PARP1 (PARPi) are used in the treatment of BRCA-/- cancers, but the basis for their potency in cells, especially in the context of HPF1, is not fully understood. Here, we demonstrate the simple one-step association for eight different PARPi to PARP1 with measured rates of association (kon) of 0.8-6 µM-1 s-1. We find only minor differences in these on rates when comparing PARP1 with the PARP1-HPF1 complex. By characterizing the rates of dissociation (koff) and the binding constants (KD) for two more recently discovered PARPi, we find, for example, that saruparib has a half-life for dissociation of 22.5 h and fluzoparib has higher affinity for PARP1 in the presence of HPF1, just like the structurally related compound olaparib. By using the measured KD and kon to calculate koff, we find that the potency of PARPi in cells correlates best with the koff from the PARP1-HPF1 complex. Our data suggest that dissociation of a drug compound from the PARP1-HPF1 complex should be the parameter of choice for guiding the development of next-generation PARPi.

Dynamics of endogenous PARP1 and PARP2 during DNA damage revealed by live-cell single-molecule imaging.

PARP1 contributes to genome architecture and DNA damage repair through its dynamic association with chromatin. PARP1 and PARP2 (PARP1/2) recognize damaged DNA and recruit the DNA repair machinery. Using single-molecule microscopy in live cells, we monitored the movement of PARP1/2 on undamaged and damaged chromatin. We identify two classes of freely diffusing PARP1/2 and two classes of bound PARP1/2. The majority (>60%) of PARP1/2 diffuse freely in both undamaged and damaged nuclei and in the presence of inhibitors of PARP1/2 used for cancer therapy (PARPi). Laser-induced DNA damage results in a small fraction of slowly diffusing PARP1 and PARP2 to become transiently bound. Treatment of cells with PARPi in the presence of DNA damage causes subtle changes in the dynamics of bound PARP1/2, but not the high levels of PARP1/2 trapping seen previously. Our results imply that next-generation PARPi could specifically target the small fraction of DNA-bound PARP1/2.

Analyzing PARP1 Activity: Small Molecule Reactants and Attached Chains of Poly (ADP-Ribose).

We describe a method for analyzing multiple products of PARylation by PARP1 and/or PARP2 using high-pressure liquid chromatography. The method quantitates the small molecules NAD+ (the substrate), nicotinamide (the byproduct of PARylation or hydrolysis of NAD+), and ADPR, the product of NAD+ hydrolysis. The method also quantitates the products of PARylation following digestion of the PAR chains into "ends," "middles," and "branches." This method is useful for dissecting both the activity and the partitioning of PARylation products between different outcomes (i.e., long chains vs. short chains, PARylation vs. hydrolysis).

The structure of the human LACTB filament reveals the mechanisms of assembly and membrane binding.

Mitochondria are complex organelles that play a central role in metabolism. Dynamic membrane-associated processes regulate mitochondrial morphology and bioenergetics in response to cellular demand. In tumor cells, metabolic reprogramming requires active mitochondrial metabolism for providing key metabolites and building blocks for tumor growth and rapid proliferation. To counter this, the mitochondrial serine beta-lactamase-like protein (LACTB) alters mitochondrial lipid metabolism and potently inhibits the proliferation of a variety of tumor cells. Mammalian LACTB is localized in the mitochondrial intermembrane space (IMS), where it assembles into filaments to regulate the efficiency of essential metabolic processes. However, the structural basis of LACTB polymerization and regulation remains incompletely understood. Here, we describe how human LACTB self-assembles into micron-scale filaments that increase their catalytic activity. The electron cryo-microscopy (cryoEM) structure defines the mechanism of assembly and reveals how highly ordered filament bundles stabilize the active state of the enzyme. We identify and characterize residues that are located at the filament-forming interface and further show that mutations that disrupt filamentation reduce enzyme activity. Furthermore, our results provide evidence that LACTB filaments can bind lipid membranes. These data reveal the detailed molecular organization and polymerization-based regulation of human LACTB and provide new insights into the mechanism of mitochondrial membrane organization that modulates lipid metabolism.

Automated modeling of protein accumulation at DNA damage sites using qFADD.py.

Eukaryotic cells are constantly subject to DNA damage, often with detrimental consequences for the health of the organism. Cells mitigate this DNA damage through a variety of repair pathways involving a diverse and large number of different proteins. To better understand the cellular response to DNA damage, one needs accurate measurements of the accumulation, retention, and dissipation timescales of these repair proteins. Here, we describe an automated implementation of the "quantitation of fluorescence accumulation after DNA damage" method that greatly enhances the analysis and quantitation of the widely used technique known as laser microirradiation, which is used to study the recruitment of DNA repair proteins to sites of DNA damage. This open-source implementation ("qFADD.py") is available as a stand-alone software package that can be run on laptops or computer clusters. Our implementation includes corrections for nuclear drift, an automated grid search for the model of a best fit, and the ability to model both horizontal striping and speckle experiments. To improve statistical rigor, the grid-search algorithm also includes automated simulation of replicates. As a practical example, we present and discuss the recruitment dynamics of the early responder PARP1 to DNA damage sites.

PARP inhibitors trap PARP2 and alter the mode of recruitment of PARP2 at DNA damage sites.

Dual-inhibitors of PARP1 and PARP2 are promising anti-cancer drugs. In addition to blocking PARP1&2 enzymatic activity, PARP inhibitors also extend the lifetime of DNA damage-induced PARP1&2 foci, termed trapping. Trapping is important for the therapeutic effects of PARP inhibitors. Using live-cell imaging, we found that PARP inhibitors cause persistent PARP2 foci by switching the mode of PARP2 recruitment from a predominantly PARP1- and PAR-dependent rapid exchange to a WGR domain-mediated stalling of PARP2 on DNA. Specifically, PARP1-deletion markedly reduces but does not abolish PARP2 foci. The residual PARP2 foci in PARP1-deficient cells are DNA-dependent and abrogated by the R140A mutation in the WGR domain. Yet, PARP2-R140A forms normal foci in PARP1-proficient cells. In PARP1-deficient cells, PARP inhibitors - niraparib, talazoparib, and, to a lesser extent, olaparib - enhance PARP2 foci by preventing PARP2 exchange. This trapping of PARP2 is independent of auto-PARylation and is abolished by the R140A mutation in the WGR domain and the H415A mutation in the catalytic domain. Taken together, we found that PARP inhibitors trap PARP2 by physically stalling PARP2 on DNA via the WGR-DNA interaction while suppressing the PARP1- and PAR-dependent rapid exchange of PARP2.

Inhibitors of PARP: Number crunching and structure gazing.

SignificancePARP is an important target in the treatment of cancers, particularly in patients with breast, ovarian, or prostate cancer that have compromised homologous recombination repair (i.e., BRCA-/-). This review about inhibitors of PARP (PARPi) is for readers interested in the development of next-generation drugs for the treatment of cancer, providing insights into structure-activity relationships, in vitro vs. in vivo potency, PARP trapping, and synthetic lethality.

The BRCT domain of PARP1 binds intact DNA and mediates intrastrand transfer.

PARP1 is a key player in the response to DNA damage and is the target of clinical inhibitors for the treatment of cancers. Binding of PARP1 to damaged DNA leads to activation wherein PARP1 uses NAD+ to add chains of poly(ADP-ribose) onto itself and other nuclear proteins. PARP1 also binds abundantly to intact DNA and chromatin, where it remains enzymatically inactive. We show that intact DNA makes contacts with the PARP1 BRCT domain, which was not previously recognized as a DNA-binding domain. This binding mode does not result in the concomitant reorganization and activation of the catalytic domain. We visualize the BRCT domain bound to nucleosomal DNA by cryogenic electron microscopy and identify a key motif conserved from ancestral BRCT domains for binding phosphates on DNA and phospho-peptides. Finally, we demonstrate that the DNA-binding properties of the BRCT domain contribute to the "monkey-bar mechanism" that mediates DNA transfer of PARP1.

HPF1 and nucleosomes mediate a dramatic switch in activity of PARP1 from polymerase to hydrolase.

Poly(ADP-ribose) polymerase 1 (PARP1) is an important player in the response to DNA damage. Recently, Histone PARylation Factor (HPF1) was shown to be a critical modulator of the activity of PARP1 by facilitating PARylation of histones and redirecting the target amino acid specificity from acidic to serine residues. Here, we investigate the mechanism and specific consequences of HPF1-mediated PARylation using nucleosomes as both activators and substrates for PARP1. HPF1 provides that catalytic base Glu284 to substantially redirect PARylation by PARP1 such that the histones in nucleosomes become the primary recipients of PAR chains. Surprisingly, HPF1 partitions most of the reaction product to free ADP-ribose (ADPR), resulting in much shorter PAR chains compared to reactions in the absence of HPF1. This HPF1-mediated switch from polymerase to hydrolase has important implications for the PARP1-mediated response to DNA damage and raises interesting new questions about the role of intracellular ADPR and depletion of NAD+.

Histone Parylation factor 1 contributes to the inhibition of PARP1 by cancer drugs.

Poly-(ADP-ribose) polymerase 1 and 2 (PARP1 and PARP2) are key enzymes in the DNA damage response. Four different inhibitors (PARPi) are currently in the clinic for treatment of ovarian and breast cancer. Recently, histone PARylation Factor 1 (HPF1) has been shown to play an essential role in the PARP1- and PARP2-dependent poly-(ADP-ribosylation) (PARylation) of histones, by forming a complex with both enzymes and altering their catalytic properties. Given the proximity of HPF1 to the inhibitor binding site both PARPs, we hypothesized that HPF1 may modulate the affinity of inhibitors toward PARP1 and/or PARP2. Here we demonstrate that HPF1 significantly increases the affinity for a PARP1 - DNA complex of some PARPi (i.e., olaparib), but not others (i.e., veliparib). This effect of HPF1 on the binding affinity of Olaparib also holds true for the more physiologically relevant PARP1 - nucleosome complex but does not extend to PARP2. Our results have important implications for the interpretation of PARP inhibition by current PARPi as well as for the design and analysis of the next generation of clinically relevant PARP inhibitors.

Bridging of nucleosome-proximal DNA double-strand breaks by PARP2 enhances its interaction with HPF1.

Poly(ADP-ribose) Polymerase 2 (PARP2) is one of three DNA-dependent PARPs involved in the detection of DNA damage. Upon binding to DNA double-strand breaks, PARP2 uses nicotinamide adenine dinucleotide to synthesize poly(ADP-ribose) (PAR) onto itself and other proteins, including histones. PAR chains in turn promote the DNA damage response by recruiting downstream repair factors. These early steps of DNA damage signaling are relevant for understanding how genome integrity is maintained and how their failure leads to genome instability or cancer. There is no structural information on DNA double-strand break detection in the context of chromatin. Here we present a cryo-EM structure of two nucleosomes bridged by human PARP2 and confirm that PARP2 bridges DNA ends in the context of nucleosomes bearing short linker DNA. We demonstrate that the conformation of PARP2 bound to damaged chromatin provides a binding platform for the regulatory protein Histone PARylation Factor 1 (HPF1), and that the resulting HPF1•PARP2•nucleosome complex is enzymatically active. Our results contribute to a structural view of the early steps of the DNA damage response in chromatin.

Probing the Conformational Changes Associated with DNA Binding to PARP1.

Poly(ADP-ribose) polymerase 1 (PARP1) is an important first responder in the mechanism of DNA repair in eukaryotic cells. It is also a validated drug target, with four different PARP inhibitors (PARPi) approved for the treatment of BRCA-negative cancers. Despite past efforts, many aspects of PARPi are poorly understood, in particular their ability to trap PARP1 on chromatin and the relationships between their potencies, cellular toxicities, and trapping efficiencies. Because PARP trapping is widely believed to originate in allosteric coupling between DNA binding and the catalytic site, we further investigated the binding properties of PARP1 to a model for DNA with a double-strand break in the presence and absence of PARPi. Specifically, we have used sequential mixing stopped-flow spectroscopy to identify a slow conformational change that follows rapid DNA binding. Using a range of DNA concentrations and different mutants of PARP1 we demonstrate that this conformational change is one of the steps of the "monkey bar mechanism" that promotes DNA-dependent dissociation of DNA. This conformational change also corresponds to the previously identified conformational change associated with DNA-dependent activation of PARP1. Despite linking the conformational change associated with DNA binding and release to DNA activation, we find no evidence for PARPi perturbing this allosteric coupling.

Nonspecific Binding of RNA to PARP1 and PARP2 Does Not Lead to Catalytic Activation.

Poly(ADP-ribose) polymerases 1 and 2 (PARP1 and PARP2, respectively), upon binding damaged DNA, become activated to add long chains of poly(ADP-ribose) (PAR) to themselves and other nuclear proteins. This activation is an essential part of the DNA damage response. The PAR modifications recruit the DNA repair machinery to sites of DNA damage and result in base excision and single-strand break repair, homologous recombination, nucleotide excision repair, and alternative nonhomologous end joining. More recently, both PARP1 and PARP2 have been shown to bind to or be activated by RNA, a property that could interfere with the function of PARP1 and PARP2 in the response to DNA damage or lead to necrosis by depletion of cellular NAD+. We have quantitatively evaluated the in vitro binding of a variety of RNAs to PARP1 and PARP2 and queried the ability of these RNAs to switch on enzymatic activity. We find that while both proteins bind RNAs without specificity toward sequence or structure, their interaction with RNA does not lead to auto-PARylation. Thus, although PARP1 and PARP2 are promiscuous with respect to activation by DNA, they both demonstrate exquisite selectivity against activation by RNA.

Kinetics of DNA-protein association and dissociation by stopped-flow spectroscopy.

The ability of proteins to recognize a specific nucleotide sequence or structural feature in the genome is central to DNA replication and repair as well as gene transcription. To understand how DNA-binding proteins accomplish this feat it is important to characterize how they efficiently recognize specific DNA sequences or various structural features in the milieu of a crowded nucleus containing 50mM basepairs of DNA. This minimally entails the determination of thermodynamic (equilibrium) parameters that characterize the relative affinities of DNA-binding proteins with their cognate and non-cognate DNA sequences of interest. Additionally, elucidating the kinetic parameters that govern DNA-protein association and dissociation can provide essential mechanistic insights into the underlying mechanisms, and can inform on how these processes can be controlled under varying cellular conditions. Unfortunately, the acquisition of kinetic data can appear intimidating and are usually not included in the characterization of most DNA binding proteins. Here we provide detailed theoretical considerations and practical, easy-to-follow protocols for using stopped-flow spectroscopy to study the kinetics of DNA-protein association and dissociation, using poly-(ADP ribose) polymerase 1 as a practical example.

Activation loop dynamics are controlled by conformation-selective inhibitors of ERK2.

Conformational selection by small molecules expands inhibitory possibilities for protein kinases. Nuclear magnetic resonance (NMR) measurements of the mitogen-activated protein (MAP) kinase ERK2 have shown that activation by dual phosphorylation induces global motions involving exchange between two states, L and R. We show that ERK inhibitors Vertex-11e and SCH772984 exploit the small energetic difference between L and R to shift the equilibrium in opposing directions. An X-ray structure of active 2P-ERK2 complexed with AMP-PNP reveals a shift in the Gly-rich loop along with domain closure to position the nucleotide in a more catalytically productive conformation relative to inactive 0P-ERK2:ATP. X-ray structures of 2P-ERK2 complexed with Vertex-11e or GDC-0994 recapitulate this closure, which is blocked in a complex with a SCH772984 analog. Thus, the L→R shift in 2P-ERK2 is associated with movements needed to form a competent active site. Solution measurements by hydrogen-exchange mass spectrometry (HX-MS) reveal distinct binding interactions for Vertex-11e, GDC-0994, and AMP-PNP with active vs. inactive ERK2, where the extent of HX protection correlates with R state formation. Furthermore, Vertex-11e and SCH772984 show opposite effects on HX near the activation loop. Consequently, these inhibitors differentially affect MAP kinase phosphatase activity toward 2P-ERK2. We conclude that global motions in ERK2 reflect conformational changes at the active site that promote productive nucleotide binding and couple with changes at the activation loop to allow control of dephosphorylation by conformationally selective inhibitors.

Q-FADD: A Mechanistic Approach for Modeling the Accumulation of Proteins at Sites of DNA Damage.

The repair of DNA damage requires the ordered recruitment of many different proteins that are responsible for signaling and subsequent repair. A powerful and widely used tool for studying the orchestrated accumulation of these proteins at damage sites is laser microirradiation in live cells, followed by monitoring the accumulation of the fluorescently labeled protein in question. Despite the widespread use of this approach, there exists no rigorous method for characterizing the recruitment process quantitatively. Here, we introduce a diffusion model that explicitly accounts for the unique sizes and shapes of individual nuclei and uses two variables: Deff, the effective coefficient of diffusion, and F, the fraction of mobile protein that accumulates at sites of DNA damage. Our model quantitatively describes the accumulation of three test proteins, poly-ADP-ribose polymerases 1 and 2 (PARP1/2) and histone PARylation factor 1. Deff for PARP1, as derived by our approach, is 6× greater than for PARP2 and in agreement with previous literature reports using fluorescence correlation spectroscopy and fluorescence recovery after photobleaching. Our data indicate that histone PARylation factor 1 arrives at sites of DNA damage independently of either PARP. Importantly, our model, which can be applied to existing data, allows for the direct comparison of the coefficient of diffusion for any DNA repair protein between different cell types, obtained in different laboratories and by different methods, and also allows for the interrogation of cell-to-cell variability.

Poly(ADP-ribose) polymerase 1 searches DNA via a 'monkey bar' mechanism.

Poly(ADP-ribose) polymerase 1 (PARP1) is both a first responder to DNA damage and a chromatin architectural protein. How PARP1 rapidly finds DNA damage sites in the context of a nucleus filled with undamaged DNA, to which it also binds, is an unresolved question. Here, we show that PARP1 association with DNA is diffusion-limited, and release of PARP1 from DNA is promoted by binding of an additional DNA molecule that facilitates a 'monkey bar' mechanism, also known as intersegment transfer. The WGR-domain of PARP1 is essential to this mechanism, and a point mutation (W589A) recapitulates the altered kinetics of the domain deletion. Demonstrating the physiological importance of the monkey bar mechanism for PARP1 function, the W589A mutant accumulates at sites of DNA damage more slowly following laser micro-irradiation than wild-type PARP1. Clinically relevant inhibitors of PARP1 did not alter the rate or mechanism of the release of PARP1 from DNA.