Date Published: October 3, 2018
Publisher: Public Library of Science
Author(s): Markus Heinonen, Fabien Milliat, Mohamed Amine Benadjaoud, Agnès François, Valérie Buard, Georges Tarlet, Florence d’Alché-Buc, Olivier Guipaud, Wataru Nishimura.
The vascular endothelium is considered as a key cell compartment for the response to ionizing radiation of normal tissues and tumors, and as a promising target to improve the differential effect of radiotherapy in the future. Following radiation exposure, the global endothelial cell response covers a wide range of gene, miRNA, protein and metabolite expression modifications. Changes occur at the transcriptional, translational and post-translational levels and impact cell phenotype as well as the microenvironment by the production and secretion of soluble factors such as reactive oxygen species, chemokines, cytokines and growth factors. These radiation-induced dynamic modifications of molecular networks may control the endothelial cell phenotype and govern recruitment of immune cells, stressing the importance of clearly understanding the mechanisms which underlie these temporal processes. A wide variety of time series data is commonly used in bioinformatics studies, including gene expression, protein concentrations and metabolomics data. The use of clustering of these data is still an unclear problem. Here, we introduce kernels between Gaussian processes modeling time series, and subsequently introduce a spectral clustering algorithm. We apply the methods to the study of human primary endothelial cells (HUVECs) exposed to a radiotherapy dose fraction (2 Gy). Time windows of differential expressions of 301 genes involved in key cellular processes such as angiogenesis, inflammation, apoptosis, immune response and protein kinase were determined from 12 hours to 3 weeks post-irradiation. Then, 43 temporal clusters corresponding to profiles of similar expressions, including 49 genes out of 301 initially measured, were generated according to the proposed method. Forty-seven transcription factors (TFs) responsible for the expression of clusters of genes were predicted from sequence regulatory elements using the MotifMap system. Their temporal profiles of occurrences were established and clustered. Dynamic network interactions and molecular pathways of TFs and differential genes were finally explored, revealing key node genes and putative important cellular processes involved in tissue infiltration by immune cells following exposure to a radiotherapy dose fraction.
Half of patients with tumors receive radiotherapy (RT) at some point during the course of their disease . In combination with surgery and chemotherapy, RT achieves good results in terms of long-term survival and tumor cure in a variety of tumors. Although the latest generation devices deliver doses more and more precisely to the tumors, the therapeutic ratio of RT is still limited by normal tissue injury in organs at risk and by the radiation resistance of some tumors . The vasculature plays a crucial role in tumor progression and in tumor sensitivity or resistance and is considered as a target in attempts to destroy tumors . It also orchestrates wound healing in the case of radiation injury . In the vasculature, the endothelium is considered as a promising target to improve the differential effect of RT in the future [4, 5].
We experimented with the proposed kernels and the clustering method and then applied them to real data. To gain insights into the biological relevance of the clustering as regards the response of endothelial cells to a conventional RT dose fraction (2 Gy), (i) we clustered the gene expression curves by the OVL kernel and the spectral clustering, (ii) we searched for putative TFs associated with the clustered differential genes and (iii) we searched for pathway relationships between TF, gene entities and the term “radiation”. Fig 1 shows the overall methodology used in this work and Fig 2 displays the workflow of data analysis, from irradiation of cells to clustering and network interaction analysis of genes and TFs. The new method and the results (both on simulated data and real data) are presented below.
To gain insights into the mechanisms involved in the molecular response of endothelial cells to ionizing radiation, we applied a new GP-kernel-based clustering to gene expression time series of irradiated HUVEC cells. This method exploits the results of the previous analysis we performed by establishing a new method that combines GPs and a novel Bayesian likelihood ratio test . In this previous work, we demonstrated that the method could well highlight phenomena already described in the response of cells to irradiation. Using the new approach, we go further in exploiting gene expression data. The novel proposed method introduced similarity measures for comparing GPs, allowing kernel-based supervised and unsupervised learning methods to be utilized on GPs. We additionally introduced an outlier-resistant variant of spectral clustering, which is particularly suitable for kernel-based clustering approaches. We evaluated the proposed novel kernels on a simulated clustering dataset. Using the real experimental data we generated and published earlier , temporal clustering over time windows, enrichment analyses and molecular pathway analysis indicate the temporal activation of biological entities and TFs by expression profile clusters.