Explore the vast range of titles available.

Astrocytes and microglia are brain‐resident glia that can establish harmful inflammatory environments in disease contexts and thereby contribute to the progression of neuronal loss in neurodegenerative disorders. Correcting the diseased properties of glia is therefore an appealing strategy for treating brain diseases. Previous studies have shown that serum/ glucocorticoid related kinase 1 (SGK1) is upregulated in the brains of patients with various neurodegenerative disorders, suggesting its involvement in the pathogenesis of those diseases. In this study, we show that inhibiting glial SGK1 corrects the pro‐inflammatory properties of glia by suppressing the intracellular NFκB‐, NLRP3‐inflammasome‐, and CGAS‐STING‐mediated inflammatory pathways. Furthermore, SGK1 inhibition potentiated glial activity to scavenge glutamate toxicity and prevented glial cell senescence and mitochondrial damage, which have recently been reported as critical pathologic features of and therapeutic targets in Parkinson disease (PD) and Alzheimer disease (AD). Along with those anti‐inflammatory/neurotrophic functions, silencing and pharmacological inhibition of SGK1 protected midbrain dopamine neurons from degeneration and cured pathologic synuclein alpha (SNCA) aggregation and PD‐associated behavioral deficits in multiple in vitro and in vivo PD models. Collectively, these findings suggest that SGK1 inhibition could be a useful strategy for treating PD and other neurodegenerative disorders that share the common pathology of glia‐mediated neuroinflammation. Synopsis Pathogenic involvement of SGK1 has been implicated in various neurodegenerative disorders. In this study, we show that inhibition of SGK1 in glia treats Parkinson disease (PD) via suppressing glial inflammation and potentiating glial neurotrophic functions. SGK inhibition reduces neuroinflammation and protects dopamine neurons from degeneration. PD – associated behavioral deficits are improved in PD mice after SGK inhibition. Inhibition of SGK1 could be a useful strategy for treating PD and other neurodegenerative disorders. Graphical Abstract Pathogenic involvement of SGK1 has been implicated in various neurodegenerative disorders. In this study, we show that inhibition of SGK1 in glia treats Parkinson disease (PD) via suppressing glial inflammation and potentiating glial neurotrophic functions.

Intraneuronal inclusions of misfolded α-synuclein (α-syn) and prion-like spread of the pathologic α-syn contribute to progressive neuronal death in Parkinson’s disease (PD). Despite the pathologic significance, no efficient therapeutic intervention targeting α-synucleinopathy has been developed. In this study, we provide evidence that astrocytes, especially those cultured from the ventral midbrain (VM), show therapeutic potential to alleviate α-syn pathology in multiple in vitro and in vivo α-synucleinopathic models. Regulation of neuronal α-syn proteostasis underlies the therapeutic function of astrocytes. Specifically, VM-derived astrocytes inhibited neuronal α-syn aggregation and transmission in a paracrine manner by correcting not only intraneuronal oxidative and mitochondrial stresses but also extracellular inflammatory environments, in which α-syn proteins are prone to pathologic misfolding. The astrocyte-derived paracrine factors also promoted disassembly of extracellular α-syn aggregates. In addition to the aggregated form of α-syn, VM astrocytes reduced total α-syn protein loads both by actively scavenging extracellular α-syn fibrils and by a paracrine stimulation of neuronal autophagic clearance of α-syn. Transplantation of VM astrocytes into the midbrain of PD model mice alleviated α-syn pathology and protected the midbrain dopamine neurons from neurodegeneration. We further showed that cografting of VM astrocytes could be exploited in stem cell–based therapy for PD, in which host-to-graft transmission of α-syn pathology remains a critical concern for long-term cell therapeutic effects.

There is a compelling need to develop disease-modifying therapies for Alzheimer’s disease (AD), the most common neuro-degenerative disorder. Together with recent progress in vector development for efficiently targeting the central nervous system, gene therapy has been suggested as a potential therapeutic modality to overcome the limited delivery of conventional types of drugs to and within the damaged brain. In addition, given increasing evidence of the strong link between glia and AD pathophysiology, therapeutic targets have been moving toward those addressing glial cell pathology. Nurr1 and Foxa2 are transcription/epigenetic regulators that have been reported to cooperatively regulate inflammatory and neurotrophic response in glial cells. In this study, we tested the therapeutic potential of Nurr1 and Foxa2 gene delivery to treat AD symptoms and pathologies. A series of functional, histologic, and transcriptome analyses revealed that the combined expression of Nurr1 and Foxa2 substantially ameliorated AD-associated amyloid β and Tau proteinopathy, cell senescence, synaptic loss, and neuro-inflammation in multiple in vitro and in vivo AD models. Intra-cranial delivery of Nurr1 and Foxa2 genes using adeno-associated virus (AAV) serotype 9 improved the memory and cognitive function of AD model mice. The therapeutic benefits of gene delivery were attained mainly by correcting pathologic glial function. These findings collectively indicate that AAV9-mediated Nurr1 and Foxa2 gene transfer could be an effective disease-modifying therapy for AD.

As DNNs are widely adopted in various application domains while demanding increasingly higher compute and memory requirements, designing efficient and performant NPUs (Neural Processing Units) is becoming more important. However, existing architectural NPU simulators lack support for high-speed simulation, multi-core modeling, multi-tenant scenarios, detailed DRAM/NoC modeling, and/or different deep learning frameworks. To address these limitations, this work proposes ONNXim, a fast cycle-level simulator for multi-core NPUs in DNN serving systems. It takes DNN models represented in the ONNX graph format generated from various deep learning frameworks for ease of simulation. In addition, based on the observation that typical NPU cores process tensor tiles from on-chip scratchpad memory with deterministic compute latency, we forgo a detailed modeling for the computation while still preserving simulation accuracy. ONNXim also preserves dependencies between compute and tile DMAs. Meanwhile, the DRAM and NoC are modeled in cycle-level to properly model contention among multiple cores that can execute different DNN models for multi-tenancy. Consequently, ONNXim is significantly faster than existing simulators (e.g., by up to 384x over Accel-sim) and enables various case studies, such as multi-tenant NPUs, that were previously impractical due to slow speed and/or lack of functionalities. ONNXim is publicly available at https://github.com/PSAL-POSTECH/ONNXim.

Emerging Compute Express Link (CXL) enables cost-efficient memory expansion beyond the local DRAM of processors. While its CXL\\(.\\)mem protocol provides minimal latency overhead through an optimized protocol stack, frequent CXL memory accesses can result in significant slowdowns for memory-bound applications whether they are latency-sensitive or bandwidth-intensive. The near-data processing (NDP) in the CXL controller promises to overcome such limitations of passive CXL memory. However, prior work on NDP in CXL memory proposes application-specific units that are not suitable for practical CXL memory-based systems that should support various applications. On the other hand, existing CPU or GPU cores are not cost-effective for NDP because they are not optimized for memory-bound applications. In addition, the communication between the host processor and CXL controller for NDP offloading should achieve low latency, but existing CXL\\(.\\)io/PCIe-based mechanisms incur \\(\\)s-scale latency and are not suitable for fine-grained NDP. To achieve high-performance NDP end-to-end, we propose a low-overhead general-purpose NDP architecture for CXL memory referred to as Memory-Mapped NDP (M\\(^2\\)NDP), which comprises memory-mapped functions (M\\(^2\\)func) and memory-mapped \\(\\)threading (M\\(^2\\)thread). M\\(^2\\)func is a CXL\\(.\\)mem-compatible low-overhead communication mechanism between the host processor and NDP controller in CXL memory. M\\(^2\\)thread enables low-cost, general-purpose NDP unit design by introducing lightweight \\(\\)threads that support highly concurrent execution of kernels with minimal resource wastage. Combining them, M\\(^2\\)NDP achieves significant speedups for various workloads by up to 128x (14.5x overall) and reduces energy by up to 87.9% (80.3% overall) compared to baseline CPU/GPU hosts with passive CXL memory.