Stroke is an acute cerebrovascular disorder caused by the sudden occlusion or rupture of a cerebral blood vessel. Approximately 80% of strokes are ischemic, whereas the remaining 20% are hemorrhagic. Stroke is a leading cause of death and long-term disability worldwide and the leading cause of death and disability in China. Clinically, stroke typically presents with the abrupt onset of focal neurological deficits, such as hemiplegia, aphasia, and facial weakness. For acute ischemic stroke, intravenous thrombolysis and endovascular thrombectomy are the two main reperfusion therapies. Although these treatments have substantially improved clinical outcomes, numerous neuroprotective strategies targeting ischemic neuronal injury have failed to demonstrate efficacy in clinical trials.

Stroke also induces sustained changes in the immune, cardiovascular, respiratory, and gastrointestinal systems, while complications including infection, early recurrent vascular events, and cardiac dysfunction further contribute to mortality and long-term disability. Stroke is therefore increasingly recognized as a systemic disease that provokes widespread responses throughout the body.

The immune response to stroke has a dual nature. Following cerebral ischemia, injured brain cells release damage-associated molecular patterns (DAMPs), including high-mobility group box 1 (HMGB1) and cell-free DNA (cfDNA), which activate innate immune signaling and initiate inflammatory responses. On the one hand, immune cells contribute to neuroinflammation and secondary tissue injury within the brain. On the other, cerebral injury disrupts peripheral immune homeostasis through neurohumoral pathways and the release of damage-associated molecular patterns.

How signals be relayed from the brain to periphery?

Inflammasomes are key molecular platforms through which the innate immune system detects danger signals. Several well-characterized inflammasomes have been identified, including NLRP1, NLRP3, NLRC4, AIM2, and pyrin, which respond to distinct inflammatory stimuli. Among these, AIM2 is particularly relevant to stroke because it directly recognizes cytosolic double-stranded DNA (dsDNA), providing a potential link between circulating cell-free DNA (cfDNA) and systemic inflammation. Upon binding to DNA, AIM2 recruits the adaptor protein ASC, leading to caspase-1 activation and the subsequent maturation and release of IL-1β.

The Liesz group found that circulating cfDNA levels increased after ischemic stroke and identified neutrophil extracellular traps (NETs) as the major source of this increase. In experimental stroke models, serum from stroke-affected mice induced caspase-1 activation in bone marrow-derived macrophages, while NET-derived DNA promoted ASC speck formation, indicating inflammasome activation.

By comparing cells deficient in different DNA-sensing pathways and inflammasome-related genes, the investigators further demonstrated that stroke serum–induced caspase-1 activation was primarily dependent on AIM2, rather than cGAS, NLRP1, or NLRP3. Notably, AIM2 deficiency abolished this response, highlighting its critical role in sensing stroke-associated cfDNA. These findings established a mechanistic axis linking brain injury, circulating cfDNA, AIM2 inflammasome activation in myeloid cells, and IL-1β release, providing a mechanistic explanation for how a focal cerebral injury can rapidly trigger a systemic immune response.

Chasing the culprit of cfNDA post stroke: neutrophil matters

Once elevated cfDNA levels were observed after stroke, an important question was its cellular origin. Analysis based on human cell type–specific DNA methylation signatures showed that the increase in plasma cfDNA in patients with stroke was predominantly attributable to neutrophils, whereas the contributions from astrocytes, neurons, and oligodendrocytes were relatively limited. Animal and ex vivo experiments further demonstrated that stroke rapidly induced neutrophil extracellular trap formation, or NETosis, and that serum from patients with stroke was itself sufficient to trigger NETosis in otherwise healthy neutrophils.Neutrophil extracellular traps (NETs) consist of a DNA scaffold decorated with histones and granular proteins and normally help capture pathogens during infection. Excessive NETosis, however, releases large amounts of DNA with pro-inflammatory and procoagulant properties into the circulation. Pharmacological inhibition of NET formation with a PAD4 inhibitor reduced circulating cfDNA levels and attenuated caspase-1 activation in splenic monocytes. These findings support NETosis as an important upstream event driving both the rise in circulating cfDNA and systemic AIM2 inflammasome activation after stroke.

Post-stroke immune responses amplify plaque inflammation, rendering the brain to recurrent stroke

Patients remain at considerable risk of recurrent ischemic events during the early period after stroke, particularly those with large-artery atherosclerotic stroke. Liesz group established an animal model with vulnerable carotid plaques and induced primary cerebral ischemia on the contralateral side to mimic the clinical course of recurrent ischemia shortly after an initial stroke. The results showed that circulating NET-derived DNA released after the primary stroke entered distant vulnerable plaques, where it activated the AIM2 inflammasome in plaque-associated myeloid cells, increased caspase-1 activity, and promoted IL-1β release.

Inflammasome activation subsequently enhanced monocyte infiltration, local immune-cell proliferation, and matrix metalloproteinase (MMP) activity. These changes drove collagen remodeling within the fibrous cap, reduced plaque stability, and promoted thrombus formation. Plaque-associated thrombosis ultimately triggered artery-to-artery embolism, resulting in new ischemic lesions in the cerebral hemisphere contralateral to the primary stroke. Consistent findings were observed in human carotid endarterectomy specimens: compared with plaques from patients with asymptomatic carotid stenosis, plaques obtained from patients with acute stroke exhibited greater monocyte infiltration, caspase-1 activation, MMP9 activity, and deposition of activated coagulation factor XII (FXIIa).

The study also provided direct evidence for therapeutic intervention. Clearance of circulating cfDNA with recombinant DNase or inhibition of inflammasome signaling with a caspase-1 inhibitor attenuated plaque inflammation and instability and markedly reduced experimental stroke recurrence. The relative risk reductions achieved with these two approaches were approximately 75% and 82%, respectively. These findings suggest that, in addition to antiplatelet and lipid-lowering therapies, secondary stroke prevention may benefit from short-term, mechanism-based immunomodulatory strategies targeting residual inflammatory risk.

Stroke induces long-term cardiac injury through trained immunity

Beyond infection and early recurrence, stroke can also lead to dysfunction in distant organs that persists for months or even longer. Using multi-organ single-cell transcriptomic profiling, Liesz group found that peripheral monocytes and tissue macrophages maintained a prolonged pro-inflammatory transcriptional state after brain injury, with these changes detectable for up to three months after stroke. By one month after experimental stroke, the heart exhibited myeloid-cell infiltration, interstitial fibrosis, and diastolic dysfunction. These alterations were observed in mouse models, with supporting evidence from analyses of cardiac tissue and clinical data from patients with stroke.

Mechanistically, the acute post-stroke increase in IL-1β acted on bone marrow hematopoietic stem and progenitor cells, inducing changes in chromatin accessibility and histone modifications. These epigenetic changes endowed their monocyte progeny with a persistent pro-inflammatory “trained” phenotype. When bone marrow from mice with stroke was transplanted into naïve recipients, the recipients still developed cardiac inflammation and fibrosis, demonstrating that this form of myeloid immune memory was transferable. Trained monocytes subsequently continued to infiltrate the heart, where they released mediators such as MMP9 and activated cardiac fibroblasts, ultimately driving interstitial fibrosis and chronic cardiac dysfunction.

At the therapeutic level, early neutralization of IL-1β after stroke prevented the long-term epigenetic reprogramming of myeloid cells. Limiting the recruitment of pro-inflammatory monocytes with a CCR2/CCR5 inhibitor likewise attenuated cardiac fibrosis and functional impairment. By linking acute post-stroke inflammation to chronic cardiac complications, this study showed that cardiac injury after stroke cannot be attributed solely to sympathetic activation or pre-existing cardiovascular risk factors. Persistent innate immune memory also represents an important pathological mechanism underlying long-term cardiac dysfunction.

Taken together, these studies delineate the temporal evolution of the systemic immune response after stroke. During the acute phase, brain injury induces NETosis and the release of cell-free DNA (cfDNA), which activate the AIM2 inflammasome in myeloid cells. Within vulnerable atherosclerotic plaques, this pathway amplifies inflammation, promotes fibrous cap degradation and thrombosis, and thereby increases the risk of early recurrent stroke. During the subacute and chronic phases, IL-1β induces persistent epigenetic changes in the bone marrow, maintaining myeloid cells in a prolonged pro-inflammatory state and ultimately driving cardiac fibrosis and dysfunction.

These findings indicate that the post-stroke immune response cannot be characterized simply as either “inflammation” or “immunosuppression.” Rather, it varies according to cell type, organ, and disease stage. Future immunotherapies will therefore need to account for patient subtype, therapeutic window, and the specific level of intervention. Broad inhibition of IL-1β may increase the risk of infection, whereas clearance of NET-derived DNA, selective blockade of AIM2–caspase-1 signaling, or restriction of specific monocyte subsets may provide more precise therapeutic approaches. As the current evidence is derived primarily from experimental models and supportive analyses of human tissue, prospective clinical studies are still needed to determine whether these targeted interventions can safely reduce stroke recurrence and complications affecting distant organs.

1. Roth S, Cao J, Singh V, et al. Post-injury immunosuppression and secondary infections are caused by an AIM2 inflammasome-driven signaling cascade. Immunity. 2021;54(4):648–659.e8. doi:10.1016/j.immuni.2021.02.004.

2. Cao J, Roth S, Zhang S, et al. DNA-sensing inflammasomes cause recurrent atherosclerotic stroke. Nature. 2024;633:433–441. doi:10.1038/s41586-024-07803-4.

3. Simats A, Zhang S, Messerer D, et al. Innate immune memory after brain injury drives inflammatory cardiac dysfunction. Cell. 2024;187(17):4637–4655.e26. doi:10.1016/j.cell.2024.06.028.