Back

A distinct fingerprint of inflammatory mediators and miRNAs in Plasmodium vivax severe thrombocytopenia

Santos, M. L. S.; Coimbra, R. S.; Sousa, T. N.; Guimaraes, L. F. F.; Gomes, M. S.; Amaral, L. R.; Pereira, D.; Fontes, C. J. F.; Hawwari, I.; Franklin, B. S.; Carvalho, L. H.

2020-08-25 microbiology
10.1101/2020.08.20.260463 bioRxiv
Show abstract

BackgroundSevere thrombocytopenia can be a determinant factor in the morbidity of Plasmodium vivax (Pv), the most widespread human malaria. Although immune mechanisms may drive Pv-induced severe thrombocytopenia (PvST), the current data on the cytokine landscape in PvST is scarce, and often conflicting. The analysis of the bidirectional circuit of inflammatory mediators and miRNAs would lead to a better understanding of the mechanisms underlying PvST. MethodsWe combined Luminex proteomics, NanoString miRNA quantification, and machine learning, to evaluate an extensive array of plasma mediators in uncomplicated Pv patients, whose blood platelet counts varied from reference values to PvST. ResultsUnsupervised clustering analysis identified PvST-linked signatures comprised of both inflammatory (CXCL10, CCL4, and IL-18) and regulatory (IL-10, IL-1Ra, HGF) mediators. As part of PvST signatures, IL-6 and IL-8 were critical to discriminate Pv subgroups, while CCL2 and IFN-{gamma} from healthy controls. Supervised machine learning spotlighted IL-10 in Pv-mediated thrombocytopenia, and provided evidence for a potential signaling route involving IL-8 and HGF. Finally, we identified a set of miRNAs capable of modulating these signaling pathways. ConclusionsThe results place IL-10 and IL-8/HGF in the center of PvST and propose investigating these signaling pathways across the spectrum of malaria infections.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.