Introduction
Ischemic stroke remains a leading cause of global mortality and disability (Collaborators, 2021). During cerebral ischemia, excessive oxidative stress and the overproduction of reactive oxygen species (ROS) act as early drivers of irreversible neuronal injury (Campbell et al., 2019; Liu et al., 2025). Accumulating evidence indicates that ROS-induced DNA damage triggers PARthanatos, a PARP1-dependent, caspase-independent form of programmed cell death that serves as the principal executioner of ischemic neuronal demise (Andrabi et al., 2006). Pathological ROS directly cause DNA strand breaks, which provoke PARP1 hyperactivation. This hyperactivation depletes cellular nicotinamide adenine dinucleotide (NAD+) and ATP reserves, precipitating a lethal bioenergetic crisis (Fatokun et al., 2014; Wu et al., 2025). Concurrently, the accumulation of free poly (ADP-ribose) (PAR) polymers induces mitochondrial depolarization and the nuclear translocation of apoptosis-inducing factor (AIF), which then associates with the PAAN (PAR-activated AIF-associated nuclease)/MIF nuclease complex to execute large-scale terminal DNA fragmentation (Huang et al., 2022; Mace et al., 2026).
Given the central role of oxidative stress in driving PARthanatos, pharmacological interception of this lethal cascade has been actively pursued. Direct catalytic inhibition of PARP1, using clinically approved agents such as olaparib or our recently identified natural quassinoid bruceine E (Wei et al., 2026), effectively blocks PARthanatos. Yet this approach presents a fundamental therapeutic paradox. Basal PARP1 activity is indispensable for recruiting DNA repair proteins, including XRCC1 and DNA ligase III, to resolve single-strand breaks through base excision repair (Gibson and Kraus, 2012). Consequently, indiscriminate PARP1 inhibition compromises genomic stability and may paradoxically exacerbate ROS-induced DNA lesions (Ray Chaudhuri and Nussenzweig, 2017). This fundamental limitation has motivated the search for downstream intervention nodes that would permit selective blockade of PARthanatos without impairing physiological DNA repair.
Within the ischemic penumbra, the c-Jun N-terminal kinase (JNK) pathway acts as a critical pro-death node exquisitely sensitive to oxidative stress (Hu et al., 2026). Upstream of JNK, death-associated protein kinase 1 (DAPK1), a calcium/calmodulin-regulated serine/threonine kinase extensively characterized in excitotoxicity (Tu et al., 2010), integrates multiple death stimuli, including ROS and DNA damage, to propagate lethal signaling (Zhang et al., 2025a). Notably, DAPK1 activates JNK through the GADD45a/MDM2 axis, linking oxidative stress sensing to mitochondrial dysfunction, ROS overproduction, and caspase-independent cell death. Whether DAPK1 functions as a kinase relay coupling PARP1 hyperactivation to the JNK–mitochondrial depolarization–AIF axis in PARthanatos, however, has remained unknown. Elucidating this position is of considerable importance, as it may reveal a druggable node entirely separable from PARP1-dependent DNA maintenance.
Through high-throughput phenotypic screening of a customized NMPA-approved library against an N-methyl-N′-nitro-N-nitrosoguanidine (MNNG)-induced PARthanatos model (Tang et al., 2019), we identified wogonoside (WS), a natural flavonoid glycoside derived from Scutellaria baicalensis (Pan et al., 2021; Zhang et al., 2025b), as a potent anti-PARthanatos lead compound with favorable preliminary tolerability. In this study, we employed a phenotype-to-target deconvolution strategy and identified DAPK1 as the direct molecular target of WS (Wang et al., 2017). Using integrated biophysical assays (Molina et al., 2013), 100-ns all-atom molecular dynamics simulations, and CRISPR/Cas9 genetic ablation (Le Cong et al., 2013), we demonstrate that WS acts as a ligand-induced degrader (KD = 6.07 μM) that triggers the proteasome-dependent clearance of DAPK1. This targeted degradation dismantles the JNK–mitochondrial depolarization–AIF execution axis while leaving PARP1 catalytic activity and genomic DNA repair fully intact. In a mouse permanent distal middle cerebral artery occlusion (pdMCAO) model, a single dose of WS reduced infarct volume by approximately 50%, rescued neurons, suppressed microglial activation, and preserved microvascular integrity (Iadecola, 2017; Iadecola and Anrather, 2011). Single-nucleus RNA sequencing revealed Dapk1 enrichment in activated microglia and co-variation with neuronal MAPK activity, suggesting a potential dual-compartment mechanism of protection. Collectively, these results identify proteasome-dependent DAPK1 degradation as a mechanistically distinct antioxidant strategy for halting ROS-driven PARthanatos in acute ischemic stroke.
Chun-Ze Zhai|Yin Cao|Yu-Xin Shen|Yi-Zhe Wang|Zhi-Peng Liu|Fang Liu|Jin-Liu Yu|Xiu-Xiu Wang|Zhao-Huan Zhang|Xiao-Hui Xu