This article is part of the supplement: Proceedings of the 2011 International Conference on Bioinformatics and Computational Biology (BIOCOMP'11)
Short and long term gene expression variation and networking in human proximal tubule cells when exposed to cadmium
1 Department of Pathology, School of Medicine and Health Sciences, University of North Dakota, Grand Forks, ND 58202, USA
2 Bioinformatics Core, School of Medicine and Health Sciences, University of North Dakota, Grand Forks, ND 58202, USA
3 Rush University Cancer Center, Chicago, IL 60612, USA
BMC Medical Genomics 2013, 6(Suppl 1):S2 doi:10.1186/1755-8794-6-S1-S2Published: 23 January 2013
Cadmium (Cd2+) is a known nephrotoxin causing tubular necrosis during acute exposure and potentially contributing to renal failure in chronic long-term exposure. To investigate changes in global gene expression elicited by cadmium, an in-vitro exposure system was developed from cultures of human renal epithelial cells derived from cortical tissue obtained from nephrectomies. These cultures exhibit many of the qualities of proximal tubule cells. Using these cells, a study was performed to determine the cadmium-induced global gene expression changes after short-term (1 day, 9, 27, and 45 μM) and long-term cadmium exposure (13 days, 4.5, 9, and 27 μM). These studies revealed fundamental differences in the types of genes expressed during each of these time points. The obtained data was further analyzed using regression to identify cadmium toxicity responsive genes. Regression analysis showed 403 genes were induced and 522 genes were repressed by Cd2+ within 1 day, and 366 and 517 genes were induced and repressed, respectively, after 13 days. We developed a gene set enrichment analysis method to identify the cadmium induced pathways that are unique in comparison to traditional approaches. The perturbation of global gene expression by various Cd2+ concentrations and multiple time points enabled us to study the transcriptional dynamics and gene interaction using a mutual information-based network model. The most prominent network module consisted of INHBA, KIF20A, DNAJA4, AKAP12, ZFAND2A, AKR1B10, SCL7A11, and AKR1C1.