PARG inhibition eliminates base excision repair-deficient and non-proliferating cancer cells through toxic PAR accumulation

Technical Summary: PARG Inhibition Eliminates Base Excision Repair-Deficient and Non-Proliferating Cancer Cells Through Toxic PAR Accumulation

Problem Statement
Poly(ADP-ribose) glycohydrolase (PARG) is the primary enzyme responsible for degrading poly(ADP-ribose) (PAR) chains synthesized by PARP enzymes during DNA damage responses. While PARG inhibition has emerged as a potential anticancer strategy, the specific mechanisms driving its cytotoxicity and the tumor contexts most susceptible to it remain incompletely understood. Previous models suggested mechanisms involving NAD⁺ depletion or replication stress, but robust biomarkers for sensitivity and a clear understanding of how PARG inhibition affects non-proliferating or therapy-resistant (senescent) cancer cells were lacking.

Methodology
The authors developed two novel small-molecule PARG inhibitors, MOD000568 (MOD568) and MOD000582 (MOD582), and characterized them through:

• Structural Biology: X-ray crystallography of the human PARG catalytic domain in complex with MOD582 to determine binding modes.
• Biochemical Assays: PAR ELISA-based assays to measure PARG inhibitory activity (IC₅₀).
• Genetic Screening: A focused siRNA loss-of-function screen across three cancer cell lines (MCF7, HCC1806, HeLa) targeting DNA repair and nucleotide metabolism pathways.
• Genetically Modified Models: Utilization of TK6, MCF7, and TOV-112D cell lines with specific knockouts (KO) or knockdowns (KD) of BER genes (XRCC1, POLβ), HR genes (BRCA1, BRCA2, PALB2), and PARP1. Auxin-inducible degron (AID) systems were used for conditional protein depletion.
• In Vivo Models: Subcutaneous xenograft models in nude mice using LoVo and A549 cells to assess antitumor efficacy and tolerability.
• Mechanistic Analyses:
  • Measurement of NAD⁺/NADH levels and PAR accumulation in the presence of NAD⁺ precursors (NRH).
  • Cell cycle analysis via EdU incorporation and flow cytometry.
  • Assessment of replication stress markers (γH2AX, p-CHK1, p-RPA2).
  • Subcellular fractionation and immunofluorescence to track PAR localization and Apoptosis-Inducing Factor (AIF) translocation.
  • Seahorse metabolic flux analysis to measure mitochondrial function (OCR, ATP production).
  • Induction of G1 arrest (using palbociclib) and therapy-induced senescence (TIS) to test efficacy in non-proliferating states.

Key Contributions and Results

1. Development of Potent PARG Inhibitors:
   The study identified MOD568 and MOD582, with MOD582 showing single-digit nanomolar potency (IC₅₀ = 0.0017 µM). Crystallographic analysis revealed that MOD582 occupies the PAR-binding pocket, forming a unique hydrogen-bond network that likely contributes to its enhanced potency compared to previous inhibitors. In vivo, MOD568 demonstrated significant antitumor activity in xenograft models with minimal body weight loss, confirming target engagement via PAR accumulation and PARP1 cleavage.

2. BER Deficiency as a Primary Sensitizer:
   An siRNA screen identified that depletion of Base Excision Repair (BER) components, particularly XRCC1, LIG1, and DUT, consistently sensitized cells to PARG inhibition. While Homologous Recombination (HR) deficiency (e.g., BRCA1/2 loss) increased sensitivity in some contexts, it was not a consistent predictor across cell lines. Genetic validation in TK6 and MCF7 models confirmed that BER-deficient cells (XRCC1-KO, POLβ-KO, or double KO) exhibited markedly higher sensitivity to MOD582 than wild-type cells. This hypersensitivity correlated with excessive accumulation of PARylated proteins.

3. Mechanism: Toxic PAR Accumulation vs. NAD⁺ Depletion:
   The authors challenged the hypothesis that NAD⁺ depletion is the primary driver of cell death. Supplementation with NRH (a precursor to NAD⁺) increased intracellular NAD⁺ levels and further enhanced PAR accumulation, yet it did not rescue cells from death; instead, it exacerbated cytotoxicity. This indicates that the toxic accumulation of PARylated proteins, rather than NAD⁺ depletion, is the primary determinant of cell death following PARG inhibition.

4. Replication-Independent Cell Death:
   Contrary to models suggesting PARG inhibitors act via replication stress, the study found that MOD582-induced cytotoxicity occurred largely independently of canonical replication stress pathways.

   • MOD582 treatment did not significantly alter EdU incorporation or cell cycle distribution in BER-deficient cells, unlike the topoisomerase I inhibitor camptothecin (CPT).
   • Markers of replication stress (γH2AX, p-CHK1, p-RPA2) were minimally activated by MOD582.
   • Sensitivity to MOD582 was unaffected by SLFN11 status, whereas CPT sensitivity was SLFN11-dependent.

5. Parthanatos-Like Signaling and Mitochondrial Dysfunction:
   The mechanism of death involves a parthanatos-like pathway. PARG inhibition led to early mitochondrial dysfunction (reduced ATP production and oxygen consumption rate) followed by the nuclear translocation of Apoptosis-Inducing Factor (AIF). This process was accompanied by caspase-dependent apoptosis (cleaved PARP1 detection), suggesting a convergence of parthanatos and apoptotic pathways driven by PAR overload.

6. Efficacy in Non-Proliferating and Senescent Cells:
   A critical finding is that PARG inhibition retains cytotoxicity in non-proliferating cells.

   • In G1-arrested cells (induced by palbociclib), MOD582 remained highly cytotoxic to BER-deficient cells, whereas replication-dependent drugs (CPT, talazoparib) lost efficacy.
   • In therapy-induced senescence (TIS) models, the combination of MOD582 with low-dose MMS efficiently eliminated senescence-like cells, preventing colony formation. This contrasts with CPT, which failed to eradicate the senescent population.

Significance
The paper establishes that PARG inhibition represents a distinct therapeutic strategy that targets cancer cells through the toxic accumulation of PARylated proteins, a mechanism driven by BER deficiency and elevated PAR metabolism rather than NAD⁺ depletion or replication stress.

The authors claim this approach offers a unique advantage over conventional DNA-damaging therapies: it can effectively eliminate non-proliferating cancer cells, including those in G1 arrest and therapy-induced senescence. This addresses a major limitation in current oncology, where residual, non-dividing tumor cells often persist and lead to relapse. The study suggests that tumors with functional DNA damage loads (potentially exacerbated by the tumor microenvironment or oxidative stress) may be vulnerable to PARG inhibition, even in the absence of specific genetic defects in core BER genes. Consequently, PARG inhibitors could serve as a strategy to target therapy-resistant and non-proliferating tumor populations that are refractory to standard treatments.