Abstract: Approximately 15%-20% of PEOPLE INFECTED with COVID-19 develop symptoms of acute infection, which develops into excessive inflammation, which is caused by a storm of cytokines in the body. Researchers at Fred Hutchinson, a renowned cancer center, have found that compounds identified using AI technology can effectively clear the inflammatory response in mouse cells, and they have application prospects in patients with new crown infection or prostate cancer.
According to the Fred Hutchinson Cancer Research Center, researchers at the center have used machine learning, deep neural networks and other artificial intelligence tools to screen, identify and validate compounds in a small number of FDA-approved drugs, and were pleasantly surprised to find that these compounds can cope with the cytokine storm that occurs in patients with acute COVID-19 infection and prostate cancer.
Cytokine storms, also known as hypercytosis or runaway immune system responses, are physiological responses that occur in humans and other animals, in which uncontrolled and excessive release of pro-inflammatory cytokines caused by the innate immune system is the primary cause. Usually, cytokines are part of the body's immune response to infection, but their sudden release can lead to multisystem organ failure and even death[1]. Cytokine storms are the main enemy of many COVID-19 patients struggling to survive in the ICU and an important factor in the death of patients triggered by the COVID-19 pandemic.
The study was published in Molecular Systems Biology as a paper titled "Machine learning identifies molecular regulators and therapeutics for targeting SARS-CoV2-induced cytokine release" (Figure 1).
Figure 1 Research results (Source: [2])
Corresponding author Gujral said that identifying and validating FDA-approved or clinical-grade compounds in the context of the coronavirus outbreak can effectively inhibit the production of inflammatory cytokines (Figure 2). For example, the research team found that a multi-enzyme inhibitor called bonatinib is a potent cytokine inhibitor that has been approved by the FDA for use in certain leukemia patients to cope with the inflammatory response in SARS-CoV-2 and its emerging variants.
Figure 2 Systems biologist Dr. Taran Gujral (Source: [2])
In addition, a cancer study by Gujral in collaboration with Dr. Pete Nelson's team was titled "Computational modeling identifies multitargeted kinase inhibitors as effective therapies for metastatic, castration-resistant prostate cancer" The paper, published in the Proceedings of the National Academy of Sciences, shows that this AI-based drug is equally effective in a metastatic castration-resistant prostate cancer (Figure 3) [3].
Figure 3 Relevant research results (Source: [4])
To figure out the specific pathways, the team decided to focus on kinases— an enzyme that coordinates molecular pathways within cells that regulate the entire process of cytokines from release to cell growth. Using the action of cytokine-modulated kinases, Vijay in Gujral's lab developed algorithms to mimic how cells respond to 428 individual kinase inhibitors.
The "multipharmacology" AI approach developed by Gujral combines mathematics with kinase inhibitors. Each inhibitor acts on a different set of kinases, and the researchers then use complex computational methods to explore the answers. In this way, the researchers were able to narrow down the range of kinases that contribute the most to the release of cytokines mediated by spike proteins, and discover some other kinases that are relevant to the cytokine storm pathway and are useful for research.
Gujral said the novel pharmacological approach has broad application potential because it can identify all downstream kinases that may have an important impact on cytokine release in a relatively objective manner. The "cytokine storm" phenomenon in this study is not specific to COVID-19 infection, but can be more in-depth study of how immune cells should respond to viral attacks, bacteria or cancer threats, which is of high research value.
Title image source: Sohu.com, only for academic exchanges.
Written by | Luna
Typography | Qiao Weijun
End
Resources:
[1]https://www.news-medical.net/health/What-is-Cytokine-Storm.aspx
[2] Chan M, Vijay S, McNevin J, et al. Machine learning identifies molecular regulators and therapeutics for targeting SARS-CoV2-induced cytokine release. Mol Syst Biol. 2021 Sep;17(9):e10426. doi: 10.15252/msb.202110426. PMID: 34486798; PMCID: PMC8420181.
[3]https://www.fredhutch.org/en/news/center-news/2021/09/hutch-scientists-using-ai-identify-potential-covid-cancer-therapies.html
[4] Bello T, Paindelli C, Diaz-Gomez LA, et al. Computational modeling identifies multitargeted kinase inhibitors as effective therapies for metastatic, castration-resistant prostate cancer. Proc Natl Acad Sci U S A. 2021 Oct 5;118(40):e2103623118. doi: 10.1073/pnas.2103623118. Epub 2021 Sep 30. PMID: 34593636; PMCID: PMC8501846.
This article is a biological exploration original, welcome to forward and share. If any other media or website needs to be reprinted, the source of biological exploration must be indicated in front of the main text.