Potential Protein Signatures for Recurrence Prediction of Ischemic Stroke.

BACKGROUND: Acute ischemic stroke is a major cause of mortality and disability worldwide, with approximately 7.4% to 7.7% recurrence within the first 3 months. This study aimed to identify potential biomarkers for predicting stroke recurrence. METHODS AND RESULTS: We conducted a nested case-control study using a hospital-based cohort from the Third China National Stroke Registry selecting 214 age- and sex-matched patients with ischemic stroke with hypertension and no history of diabetes or heart disease. Using data-independent acquisition for discovery and multiple reaction monitoring for quantitative validation, we identified 26 differentially expressed proteins in large-artery atherosclerosis (Causative Classification of Ischemic Stroke [CCS]1), 16 in small-artery occlusion (CCS3), and 25 in undetermined causes (CCS5) among patients with recurrent stroke. In the CCS1 and CCS3 subgroups, differentially expressed proteins were associated with platelet aggregation, neuronal death/cerebroprotection, and immune response, whereas differentially expressed proteins in the CCS5 subgroup were linked to altered metabolic functions. Validated recurrence predictors included proteins associated with neutrophil activity and vascular inflammation (TAGLN2 [transgelin 2], ITGAM [integrin subunit α M]/TAGLN2 ratio, ITGAM/MYL9 [myosin light chain 9] ratio, TAGLN2/RSU1 [Ras suppressor protein 1] ratio) in the CCS3 subgroup and proteins associated with endothelial plasticity and blood-brain barrier integrity (ITGAM/MYL9 ratio and COL1A2 [collagen type I α 2 chain]/MYL9 ratio) in the CCS3 and CCS5 subgroups, respectively. CONCLUSIONS: These findings provide a foundation for developing a blood-based biomarker panel, using causative classifications, which may be used in routine clinical practice to predict stroke recurrence.

Nur77 increases mitophagy and decreases aggregation of α-synuclein by modulating the p-c-Abl/p-PHB2 Y121 in α-synuclein PFF SH-SY5Y cells and mice.

Parkinson's disease (PD) is characterized by the progressive death of dopamine (DA) neurons and the pathological accumulation of α-synuclein (α-syn) fibrils. In our previous study, simulated PHB2 phosphorylation was utilized to clarify the regulatory role of c-Abl in PHB2-mediated mitophagy in PD models. In this investigation, we employed an independently patented PHB2Y121 phosphorylated antibody in the PD model to further verify that the c-Abl inhibitor STI571 can impede PHB2Y121 phosphorylation, decrease the formation of α-Syn polymers, and improve autophagic levels. The specific involvement of Nur77 in PD pathology has remained elusive. We also investigate the contribution of Nur77, a nuclear transcription factor, to α-syn and mitophagy in PD. Our findings demonstrate that under α-syn, Nur77 translocates from the cytoplasm to the mitochondria, improving PHB-mediated mitophagy by regulating c-Abl phosphorylation. Moreover, Nur77 overexpression alleviates the expression level of pS129-α-syn and the loss of DA neurons in α-syn PFF mice, potentially associated with the p-c-Abl/p-PHB2 Y121 axis. This study provides initial in vivo and in vitro evidence that Nur77 protects PD DA neurons by modulating the p-c-Abl/p-PHB2 Y121 axis, and STI571 holds promise as a treatment for PD.

APOE from patient-derived astrocytic extracellular vesicles alleviates neuromyelitis optica spectrum disorder in a mouse model.

Neuromyelitis optica spectrum disorder (NMOSD) is an autoimmune astrocytopathy of the central nervous system, mediated by antibodies against aquaporin-4 water channel protein (AQP4-Abs), resulting in damage of astrocytes with subsequent demyelination and axonal damage. Extracellular communication through astrocyte-derived extracellular vesicles (ADEVs) has received growing interest in association with astrocytopathies. However, to what extent ADEVs contribute to NMOSD pathogenesis remains unclear. Here, through proteomic screening of patient-derived ADEVs, we observed an increase in apolipoprotein E (APOE)-rich ADEVs in patients with AQP4-Abs-positive NMOSD. Intracerebral injection of the APOE-mimetic peptide APOE130-149 attenuated microglial reactivity, neuroinflammation, and brain lesions in a mouse model of NMOSD. The protective effect of APOE in NMOSD pathogenesis was further established by the exacerbated lesion volume in APOE-deficient mice, which could be rescued by exogenous APOE administration. Genetic knockdown of the APOE receptor lipoprotein receptor-related protein 1 (LRP1) could block the restorative effects of APOE130-149 administration. The transfusion ADEVs derived from patients with NMOSD and healthy controls also alleviated astrocyte loss, reactive microgliosis, and demyelination in NMOSD mice. The slightly larger beneficial effect of patient-derived ADEVs as compared to ADEVs from healthy controls was further augmented in APOE-/- mice. These results indicate that APOE from astrocyte-derived extracellular vesicles could mediate disease-modifying astrocyte-microglia cross-talk in NMOSD.

Evaluating the clinical utility of artificial intelligence assistance and its explanation on the glioma grading task.

Clinical evaluation evidence and model explainability are key gatekeepers to ensure the safe, accountable, and effective use of artificial intelligence (AI) in clinical settings. We conducted a clinical user-centered evaluation with 35 neurosurgeons to assess the utility of AI assistance and its explanation on the glioma grading task. Each participant read 25 brain MRI scans of patients with gliomas, and gave their judgment on the glioma grading without and with the assistance of AI prediction and explanation. The AI model was trained on the BraTS dataset with 88.0% accuracy. The AI explanation was generated using the explainable AI algorithm of SmoothGrad, which was selected from 16 algorithms based on the criterion of being truthful to the AI decision process. Results showed that compared to the average accuracy of 82.5±8.7% when physicians performed the task alone, physicians' task performance increased to 87.7±7.3% with statistical significance (p-value = 0.002) when assisted by AI prediction, and remained at almost the same level of 88.5±7.0% (p-value = 0.35) with the additional assistance of AI explanation. Based on quantitative and qualitative results, the observed improvement in physicians' task performance assisted by AI prediction was mainly because physicians' decision patterns converged to be similar to AI, as physicians only switched their decisions when disagreeing with AI. The insignificant change in physicians' performance with the additional assistance of AI explanation was because the AI explanations did not provide explicit reasons, contexts, or descriptions of clinical features to help doctors discern potentially incorrect AI predictions. The evaluation showed the clinical utility of AI to assist physicians on the glioma grading task, and identified the limitations and clinical usage gaps of existing explainable AI techniques for future improvement.

Plasma Reference Ranges for Brain Injury Biomarkers (NfL, NfH, MCP-1, and MMP-9) in Healthy Chinese.

BACKGROUND: Brain injury triggers neuroaxonal injury and neural death, that leads to the development of secondary sequelae. Throughout this process, brain injury factors released into circulation via the injured neurovascular unit are important prognostic parameters. Plasma NfL, NfH, MCP-1, and MMP-9 have been identified as potential indicators in this regard. METHODS: Using a microfluidic ELISA platform, we measured plasma from 273 healthy subjects that underwent quantifications of NfL, NfH, MCP-1, and MMP-9 levels. We investigated the possible associations between biomarkers and basic demographics. RESULTS: The median concentration of plasma NfL was 10.40 (IQR = 6.73 - 16.60) pg/mL, NfH was 70.70 (IQR = 39.75 - 125.50) pg/mL, MCP-1 was 191.0 (IQR = 162.0 - 237.5) pg/mL, and MMP-9 was 169,255 (IQR = 107,657 - 231,276) pg/mL. Among all four biomarkers, plasma NfL and NfH levels were positively correlated with age (r = 0.557, p < 0.001, r = 0.364, p = 0.003). NfL was also correlated with NfH (r = 0.391, p = 0.002). CONCLUSIONS: These data provide a basis for the potential application of a brain-injury biomarker panel in routine clinical practice. It lays a significant foundation in supporting circulating CNS-biomarkers as noninvasive biomarkers for neurological disorders.

CD4+ T cells aggravate hemorrhagic brain injury.

Leukocyte infiltration accelerates brain injury following intracerebral hemorrhage (ICH). Yet, the involvement of T lymphocytes in this process has not been fully elucidated. Here, we report that CD4+ T cells accumulate in the perihematomal regions in the brains of patients with ICH and ICH mouse models. T cells activation in the ICH brain is concurrent with the course of perihematomal edema (PHE) development, and depletion of CD4+ T cells reduced PHE volumes and improved neurological deficits in ICH mice. Single-cell transcriptomic analysis revealed that brain-infiltrating T cells exhibited enhanced proinflammatory and proapoptotic signatures. Consequently, CD4+ T cells disrupt the blood-brain barrier integrity and promote PHE progression through interleukin-17 release; furthermore, the TRAIL-expressing CD4+ T cells engage DR5 to trigger endothelial death. Recognition of T cell contribution to ICH-induced neural injury is instrumental for designing immunomodulatory therapies for this dreadful disease.

Bioinformatic identification of hub genes Myd88 and Ccl3 and TWS-119 as a potential agent for the treatment of massive cerebral infarction.

Background: Massive cerebral infarction (MCI) causes severe neurological deficits, coma and can even result in death. Here, we identified hub genes and pathways after MCI by analyzing microarray data from a murine model of ischemic stroke and identified potential therapeutic agents for the treatment of MCI. Methods: Microarray expression profiling was performed using the GSE28731 and GSE32529 datasets from the Gene Expression Omnibus (GEO) database. Data from a sham group (n = 6 mice) and a middle cerebral artery occlusion (MCAO) group (n = 7 mice) were extracted to identify common differentially expressed genes (DEGs). After identifying gene interactions, we generated a protein-protein interaction (PPI) network with Cytoscape software. Then, the MCODE plug-in in Cytoscape was used to determine key sub-modules according to MCODE scores. Enrichment analyses were then conducted on DEGs in the key sub-modules to evaluate their biological functions. Furthermore, hub genes were identified by generating the intersections of several algorithms in the cytohubba plug-in; these genes were then verified in other datasets. Finally, we used Connectivity MAP (CMap) to identify potential agents for MCI therapy. Results: A total of 215 common DEGs were identified and a PPI network was generated with 154 nodes and 947 edges. The most significant key sub-module had 24 nodes and 221 edges. Gene ontology (GO) analysis showed that the DEGs in this sub-module showed enrichment in inflammatory response, extracellular space and cytokine activity in terms of biological process, cellular component and molecular function, respectively. Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis revealed that TNF signaling was the most enriched pathway. Myd88 and Ccl3 were identified as hub genes and TWS-119 was identified as the most potential therapeutic agent by CMap. Conclusions: Bioinformatic analysis identified two hub genes (Myd88 and Ccl3) for ischemic injury. Further analysis identified TWS-119 as the best potential candidate for MCI therapy and that this target may be associated with TLR/MyD88 signaling.

Group 2 innate lymphoid cells resolve neuroinflammation following cerebral ischaemia.

BACKGROUND: Acute brain ischaemia elicits pronounced inflammation, which aggravates neural injury. However, the mechanisms governing the resolution of acute neuroinflammation remain poorly understood. In contrast to regulatory T and B cells, group 2 innate lymphoid cells (ILC2s) are immunoregulatory cells that can be swiftly mobilised without antigen presentation; whether and how these ILC2s participate in central nervous system inflammation following brain ischaemia is still unknown. METHODS: Leveraging brain tissues from patients who had an ischaemic stroke and a mouse model of focal ischaemia, we characterised the presence and cytokine release of brain-infiltrating ILC2s. The impact of ILC2s on neural injury was evaluated through antibody depletion and ILC2 adoptive transfer experiments. Using Rag2-/-γc-/- mice receiving passive transfer of IL-4-/- ILC2s, we further assessed the contribution of interleukin (IL)-4, produced by ILC2s, in ischaemic brain injury. RESULTS: We demonstrate that ILC2s accumulate in the areas surrounding the infarct in brain tissues of patients with cerebral ischaemia, as well as in mice subjected to focal cerebral ischaemia. Oligodendrocytes were a major source of IL-33, which contributed to ILC2s mobilisation. Adoptive transfer and expansion of ILC2s reduced brain infarction. Importantly, brain-infiltrating ILC2s reduced the magnitude of stroke injury severity through the production of IL-4. CONCLUSIONS: Our findings revealed that brain ischaemia mobilises ILC2s to curb neuroinflammation and brain injury, expanding the current understanding of inflammatory networks following stroke.

Generating post-hoc explanation from deep neural networks for multi-modal medical image analysis tasks.

Explaining model decisions from medical image inputs is necessary for deploying deep neural network (DNN) based models as clinical decision assistants. The acquisition of multi-modal medical images is pervasive in practice for supporting the clinical decision-making process. Multi-modal images capture different aspects of the same underlying regions of interest. Explaining DNN decisions on multi-modal medical images is thus a clinically important problem. Our methods adopt commonly-used post-hoc artificial intelligence feature attribution methods to explain DNN decisions on multi-modal medical images, including two categories of gradient- and perturbation-based methods. • Gradient-based explanation methods - such as Guided BackProp, DeepLift - utilize the gradient signal to estimate the feature importance for model prediction. • Perturbation-based methods - such as occlusion, LIME, kernel SHAP - utilize the input-output sampling pairs to estimate the feature importance. • We describe the implementation details on how to make the methods work for multi-modal image input, and make the implementation code available.

Targeting CCL5 signaling attenuates neuroinflammation after seizure.

BACKGROUND: Epilepsy is a neurological condition that causes unprovoked, recurrent seizures. Accumulating evidence from clinical and experimental studies indicates that neuroinflammation exacerbates seizure activity. METHODS: We investigated the transcriptional changes occurring in specific brain domains of a seizure mouse model, using 10× Genomics spatial transcriptomics. Differential gene expression and pathway analysis were applied to investigate potential signaling targets for seizure, including CCL5/CCR5 pathway. Maraviroc, an FDA-approved C-C chemokine receptor 5 (CCR5) antagonist, was used to verify the impact of CCL5/CCR5 signaling in seizure mice. RESULTS: We found distinguished regional transcriptome features in the hippocampus of seizure mice. The hippocampus exhibited unique inflammatory gene signatures, including glia activation, apoptosis, and immune response in seizure mice. Especially, we observed notable expression of C-C chemokine ligand 5 (CCL5) throughout the entire seizure hippocampus. Blockade of CCL5/CCR5 signaling via maraviroc prevented microglia activation and neuron degeneration in seizure mice. CONCLUSIONS: This study supports the potential of CCL5/CCR5 signaling for targeting neuroinflammation after seizure.

Guidelines and evaluation of clinical explainable AI in medical image analysis.

Explainable artificial intelligence (XAI) is essential for enabling clinical users to get informed decision support from AI and comply with evidence-based medical practice. Applying XAI in clinical settings requires proper evaluation criteria to ensure the explanation technique is both technically sound and clinically useful, but specific support is lacking to achieve this goal. To bridge the research gap, we propose the Clinical XAI Guidelines that consist of five criteria a clinical XAI needs to be optimized for. The guidelines recommend choosing an explanation form based on Guideline 1 (G1) Understandability and G2 Clinical relevance. For the chosen explanation form, its specific XAI technique should be optimized for G3 Truthfulness, G4 Informative plausibility, and G5 Computational efficiency. Following the guidelines, we conducted a systematic evaluation on a novel problem of multi-modal medical image explanation with two clinical tasks, and proposed new evaluation metrics accordingly. Sixteen commonly-used heatmap XAI techniques were evaluated and found to be insufficient for clinical use due to their failure in G3 and G4. Our evaluation demonstrated the use of Clinical XAI Guidelines to support the design and evaluation of clinically viable XAI.

Nonreceptor Tyrosine Kinase c-Abl-Mediated PHB2 Phosphorylation Aggravates Mitophagy Disorder in Parkinson's Disease Model.

Mitophagy and oxidative stress play important roles in Parkinson's disease (PD). Dysregulated mitophagy exacerbates mitochondrial oxidative damage; however, the regulatory mechanism of mitophagy is unclear. Here, we provide a potential mechanistic link between c-Abl, a nonreceptor tyrosine kinase, and mitophagy in PD progression. We found that c-Abl activation reduces the interaction of prohibitin 2 (PHB2) and microtubule-associated protein 1 light chain 3 (LC3) and decreases the expressive level of antioxidative stress proteins, including nuclear factor erythroid 2-related factor 2 (Nrf2), NADPH quinone oxidoreductase-1 (NQO-1), and the antioxidant enzyme heme oxygenase-1 (HO-1) in 1-methyl-4-phenylpyridinium- (MPP+-) lesioned SH-SY5Y cells. Importantly, we found that MPP+ can increase the expression of phosphorylated proteins at the tyrosine site of PHB2 and the interaction of c-Abl with PHB2. We showed for the first time that PHB2 by changing tyrosine (Y) to aspartate (D) at site 121 resulted in impaired binding of PHB2 and LC3 in vitro. Moreover, silencing of PHB2 can decrease the interaction of PHB2 and LC3 and exacerbate the loss of dopaminergic neurons. We also found that STI 571, a c-Abl family kinase inhibitor, can decrease dopaminergic neuron damage and ameliorate MPTP-induced behavioral deficits in PD mice. Taken together, our findings highlight a novel molecular mechanism for aberrant PHB2 phosphorylation as an inhibitor of c-Abl activity and suggest that c-Abl and PHB2 are potential therapeutic targets for the treatment of individuals with PD. However, these results need to be further validated in PHB2 Y121D mice.

Brain-derived programmed death-ligand 1 mediates immunosuppression post intracerebral hemorrhage.

Immunosuppression commonly occurs after a stroke, which is believed to be associated with the increased risk of infectious comorbidities of stroke patients, while the mechanisms underlying post-stroke immunosuppression is yet to be elucidated. In the brains of intracerebral hemorrhage (ICH) patients and murine ICH models, we identified that neuron-derived programmed death-ligand 1 (PD-L1) is reduced in the perihematomal area, associating increased soluble PD-L1 level in the peripheral blood. ICH induced a significant decrease of T and natural killer (NK) cell numbers in the periphery with an upregulation of programed death-1 (PD-1) in these cells. Blocking PD-1 pathway with an anti-PD1 monoclonal antibody prevented the T and NK cell compartment contraction and spleen atrophy post-ICH, with reduced pulmonary bacterial burden and improved neurological outcome. Thus, we here identified that brain-derived PD-L1 as a new mechanism driving post-stroke immunosuppression, and anti-PD1 treatment could be potentially developed to reducing the risk of post-stroke infections.

Role of Immune and Inflammatory Mechanisms in Stroke: A Review of Current Advances.

Stroke accounts for a large proportion of morbidity and mortality burden in China. Moreover, there is a high prevalence of the leading risk factors for stroke, including hypertension and smoking. Understanding the underlying mechanisms and developing effective therapeutic interventions for patients with stroke is a key imperative. The pathophysiology of stroke involves a complex interplay between the immune and inflammatory mechanisms. Focal brain inflammation triggered by neuronal cell death and the release of factors such as damage-associated molecular patterns can further exacerbate neuronal injury; in addition, impairment of the blood-brain barrier, oxidative stress, microvascular dysfunction, and brain edema cause secondary brain injury. Immune cells, including microglia and other infiltrating inflammatory cells, play a key role in triggering focal and global brain inflammation. Anti-inflammatory therapies targeting the aforementioned mechanisms can alleviate primary and secondary brain injury in the aftermath of a stroke. Further experimental and clinical studies are required to explore the beneficial effects of anti-inflammatory drugs in stroke.

Pertussis toxin-induced inflammatory response exacerbates intracerebral haemorrhage and ischaemic stroke in mice.

BACKGROUND: Stroke is a devastating disease, including intracerebral haemorrhage (ICH) and ischaemic stroke. Emerging evidences indicate that systemic inflammatory cascades after stroke contribute to brain damage. However, the direct effects and features of systemic inflammation on brain injury, especially comparing between ischaemic and haemorrhagic stroke, are still obscure. METHODS: Pertussis toxin (PT) was used to build a pro-inflammatory milieu after ICH and ischaemic stroke in mouse model. The neurodeficits, stroke lesion, immune response and blood-brain barrier (BBB) destruction were assessed. RESULTS: In ICH mouse model, PT-induced systemic inflammation exacerbated neurological deficits, and enlarged haemorrhage lesion and perihaematomal oedema. We also found promoted leucocyte infiltration and inflammatory cytokine release into the brain after PT treatment. Moreover, the integrity of the BBB was further disrupted after receiving PT. Furthermore, we demonstrated that PT enhanced brain inflammation and aggravated stroke severity in middle cerebral artery occlusion mouse model. CONCLUSIONS: Our results suggest that PT increases inflammatory response that exacerbates brain injury after ICH or ischaemic stroke in mouse model.

miR-1224 contributes to ischemic stroke-mediated natural killer cell dysfunction by targeting Sp1 signaling.

BACKGROUND: Brain ischemia compromises natural killer (NK) cell-mediated immune defenses by acting on neurogenic and intracellular pathways. Less is known about the posttranscriptional mechanisms that regulate NK cell activation and cytotoxicity after ischemic stroke. METHODS: Using a NanoString nCounter® miRNA array panel, we explored the microRNA (miRNA) profile of splenic NK cells in mice subjected to middle cerebral artery occlusion. Differential gene expression and function/pathway analysis were applied to investigate the main functions of predicted miRNA target genes. miR-1224 inhibitor/mimics transfection and passive transfer of NK cells were performed to confirm the impact of miR-1224 in NK cells after brain ischemia. RESULTS: We observed striking dysregulation of several miRNAs in response to ischemia. Among those miRNAs, miR-1224 markedly increased 3 days after ischemic stroke. Transfection of miR-1224 mimics into NK cells resulted in suppression of NK cell activity, while an miR-1224 inhibitor enhanced NK cell activity and cytotoxicity, especially in the periphery. Passive transfer of NK cells treated with an miR-1224 inhibitor prevented the accumulation of a bacterial burden in the lungs after ischemic stroke, suggesting an enhanced immune defense of NK cells. The transcription factor Sp1, which controls cytokine/chemokine release by NK cells at the transcriptional level, is a predicted target of miR-1224. The inhibitory effect of miR-1224 on NK cell activity was blocked in Sp1 knockout mice. CONCLUSIONS: These findings indicate that miR-1224 may serve as a negative regulator of NK cell activation in an Sp1-dependent manner; this mechanism may be a novel target to prevent poststroke infection specifically in the periphery and preserve immune defense in the brain.

Neuroblast senescence in the aged brain augments natural killer cell cytotoxicity leading to impaired neurogenesis and cognition.

Normal aging is accompanied by escalating systemic inflammation. Yet the potential impact of immune homeostasis on neurogenesis and cognitive decline during brain aging have not been previously addressed. Here we report that natural killer (NK) cells of the innate immune system reside in the dentate gyrus neurogenic niche of aged brains in humans and mice. In situ expansion of these cells contributes to their abundance, which dramatically exceeds that of other immune subsets. Neuroblasts within the aged dentate gyrus display a senescence-associated secretory phenotype and reinforce NK cell activities and surveillance functions, which result in NK cell elimination of aged neuroblasts. Genetic or antibody-mediated depletion of NK cells leads to sustained improvements in neurogenesis and cognitive function during normal aging. These results demonstrate that NK cell accumulation in the aging brain impairs neurogenesis, which may serve as a therapeutic target to improve cognition in the aged population.