对小胶质细胞响应肿瘤学化合物的全蛋白质组高通量扰动筛选揭示了不同的免疫调节和毒性特征

Microglia are critical for maintaining neural homeostasis. However, under pathological conditions, they can adopt a pro-inflammatory M1 phenotype characterized by the release of IL12, IL1beta, TNFalpha, and IL6. Chronically, the M1 state contributes to neurodegenerative diseases including Alzheimer's, Parkinson's, and multiple sclerosis (MS). M1-polarized microglia are of particular interest in glioblastoma (GBM) because their pro-inflammatory cytokine programs can both suppress antitumor immunity and promote tumor growth; influencing disease progression and shaping effectiveness of emerging therapeutic strategies. Identifying potential therapies requires capturing the breath of inflammatory proteins produced by microglia, and signals of potential toxicities to minimize adverse events (AEs). The high cost of capturing these diverse signals at scale limits typical drug discovery efforts to simple low-plex readouts. To address this, we previously described the Nomic platform, a proteomics tool capable of quantifying >1000 proteins simultaneously. Here, we leverage Nomic's Omni 1000 to screen 510 bioactive small molecules in microglia stimulated with LPS to mimic a pro-inflammatory state., generating >6,500 samples and >6.5 million data points. LPS led to expected increased expression of TNFɑ, IL1beta, and IL6, as well as ISG15 and IL12 p40, indicators of chronic inflammation associated with neurodegenerative diseases. Approximately 1/3 of the compounds screened blocked inflammation, of which ¼ presented clear signs of toxicity, characterized by leaking of intracellular content into the supernatant. To further assess potential toxicities, we treated cardiomyocytes and hepatocytes with these compounds, identifying expected and novel drugs with promising properties. For example, Berberine (BBR) is being developed to revert M1 microglia to M2 by blocking NF-κB signaling. We observed significantly reduced IL12 p40, IL10, and CXCL6 upon addition of BBR, with minimal signs of toxicities. Methylprednisolone (MP), a synthetic glucocorticoid, is used for inflammatory flares in MS. Here, MP significantly reduced TNFɑ, IL6, IL1beta, and IL12 p40 in LPS-treated microglia. However, in hepatocytes, MP led to increased SAA, and IL6, consistent with clinical reports of hepatic AEs with high MP exposure. In addition, several compounds under clinical investigation for GBM, including proteasome inhibitors (Bortezomib), and topoisomerase inhibitors (Doxorubicin, Camptothecin derivatives), showed distinct immunomodulatory and toxicity profiles, revealing microglial cytokine programs that may inform potential therapeutic outcomes or combinatorial strategies Our results demonstrate the value of high-throughput proteomics to simultaneously identify immunomodulatory compounds and characterize their efficacy and safety profile.