Deep models of integrated multiscale molecular data decipher the endothelial cell response to ionizing radiation

Abstract

Summary The vascular endothelium is a hot spot in the response to radiation therapy for both tumors and normal tissues. To improve patient outcomes, interpretable systemic hypotheses are needed to help radiobiologists and radiation oncologists propose endothelial targets that could protect normal tissues from the adverse effects of radiation therapy and/or enhance its antitumor potential. To this end, we captured the kinetics of multi-omics layers—i.e. miRNome, targeted transcriptome, proteome, and metabolome—in irradiated primary human endothelial cells cultured in vitro . We then designed a strategy of deep learning as in convolutional graph networks that facilitates unsupervised high-level feature extraction of important omics data to learn how ionizing radiation-induced endothelial dysfunction may evolve over time. Last, we present experimental data showing that some of the features identified using our approach are involved in the alteration of angiogenesis by ionizing radiation.

Profiling mRNA, miRNA and lncRNA expression changes in endothelial cells in response to increasing doses of ionizing radiation

“…Several other studies have looked at transcriptional and other molecular responses of the endothelium to radiation using human umbilical vein endothelial cells 27 – 29 . In an elegant study published by Morilla et al in 2021, authors dissected out multi-omics comprising miRnome, transcriptome, proteome and metabolome to identify consistent changes evolving over time in response to radiation 30 . The group identified alterations in angiogenesis in addition to DNA damage response, apoptosis, senescence, immune-response and fibrosis 30 .…”

Treatment-associated remodeling of the pancreatic cancer endothelium at single-cell resolution

“…Interestingly, Lineage Program 5 (reactive EndMT) ECs were associated with malignant cells with a high expression of neural-like progenitor, neuroendocrine-like, and mesenchymal programs, which have been linked to treatment-resistant phenotypes with poor clinical outcomes (7,14,15). We further validated the treatment-associated enrichment of Lineage Program 5 (reactive EndMT) EC expression and depletion of Lineage Program 4 (lymphatic) EC expression in primary human umbilical vein endothelial cells (HUVECs) undergoing ionizing radiation in vitro (8). Taken together, these results motivate further investigation into the potential collaborative interactions among Lineage Program 5 (reactive EndMT) ECs; neural-like progenitor, neuroendocrine-like, and mesenchymal malignant cells; and various immune cell types that may mediate therapeutic resistance in PDAC.…”

“…Targeted transcriptome data from cultured primary human umbilical vein endothelial cells (HUVECs) were obtained from Morilla et al (8). HUVECs were divided into groups: irradiated (n = 24) and non-irradiated (n = 24) conditions.…”

“…We further validated the identification of a treatmentassociated reactive EndMT phenotype by comparing Lineage Program 5 expression in non-irradiated (n = 24) versus irradiated (n = 24) primary human umbilical vein endothelial cells (HUVECs) with targeted transcriptome data (8). Lineage Program 5 (reactive EndMT) expression is significantly enriched in irradiated HUVECs in vitro (Bonferroni-adjusted p value = 6.79 × 10 -6 ; two-sided Wilcoxon signed-rank test) (Figure 2C).…”

Proteome dynamics of COVID-19 severity learnt by a graph convolutional network of multi-scale topology

“…Surprisingly, we could confirm (see Fig.4) that neither was highly connected nor played an important modular role in the graphs. Next, we monitored the behavioural regulation of the nodes’ graph aggregation on semi-supervised learning on a community composed by known and unknown protein interaction with ACE2 and TMPRSS2 Morilla et al 2022 (see the movies in format .avi provided in SI, namely: mild / intermediate / acute_fcD (resp. outlier)_severity_covidp_candidates_anim .).…”