NEW YORK (GenomeWeb) — Though most disease-causing genes are expressed throughout the body, they typically have an effect in one tissue, and unraveling that effect can reveal the molecular basis of the disease, researchers from Ben-Gurion University of the Negev in Israel reported in PLOS Computational Biology yesterday.
By superimposing gene and protein expression data from more than a dozen tissues with protein-protein interaction data, the researchers led by Esti Yeger-Lotem found that disease-causing genes are expressed at higher levels in the tissues where they have their effect and are involved in a higher number of protein-protein interactions within those tissues.
"We demonstrate that knowledge of the tissue-specific [protein-protein interactions] of genes causing hereditary diseases can be used to highlight disease-related mechanisms," said Yeger-Lotem and her colleagues in their paper. "Therefore, comparison between tissue interactomes can serve as an efficient strategy for illuminating the molecular basis of these diseases."
The researchers analyzed data on tissue-specific gene and protein expression from three datasets: the Genomics Institute of the Novartis Research Foundation, the Human Protein Atlas, and the Illumina Body Map 2.0. These datasets were generated using DNA microarrays, an immunohistochemical approach, and RNA-seq, respectively. The tissues they examined included brain, breast, heart, lung, skeletal muscle, and more.
From these three datasets, the researchers extracted the set of proteins or protein-coding genes expressed per tissue.
They noted a bi-modal distribution of expressed genes, meaning genes were mostly either expressed globally or in specific tissues. Globally expressed genes were enriched for basic cellular processes like RNA splicing, while tissue-specific genes were enriched for locally specific process like spermatogenesis in the testis, they added.
To build an interactome, Yeger-Lotem and her colleagues drew on an additional four public databases housing recently reported, experimentally detected protein-protein interactions in humans.
They then filtered this interactome using that set of proteins or protein-coding genes they developed, resulting in a tissue interactome covering more than half of the global human interactome.
All the tissue interactomes, the researchers noted, shared a common core sub-network of nearly 5,000 proteins and more than 26,000 protein-protein interactions. The expression levels of these proteins, they found, remains mostly stable across tissues, and these proteins are also enriched for basic cellular processes.
Most proteins, they noted, interacted with at least five others, though some, which they designated as hubs, had more than 45 interacting partners.
Yeger-Lotem and her colleagues assembled a list of 303 hereditary diseases that affect at least one of the 16 tissues they studied and their 233 germline causal genes. Though these genes cause disease in mostly one tissue, the researchers reported that more than 80 percent of those causal genes were expressed in 10 or more tissues.
However, they found that these causal genes had a higher median level of expression in their disease tissues than in their non-disease tissues. Additionally, these causal genes, when expressed as a protein, also had more protein-protein interactions in their disease tissue. Further, they have a higher tendency, the researchers said, to be involved in tissue-specific protein-protein interactions, particularly in their disease tissues.
"Given that mutations leading to diseases were shown in some cases to disturb the physical interactions of disease proteins, the higher tendency for potentially disturbed PPIs in disease tissues may underlie the increased vulnerability of these tissues," Yeger-Lotem and her colleagues said.
For example, the researchers reported that, in breast tissue, the BRCA1 protein is involved in one tissue-specific protein-protein interaction with ESR1, an estrogen receptor that initiates cellular proliferation. Previous reports found that the BRCA1 protein interacts with and inhibits ESR1 and cell proliferation, but with mutated forms of BRCA1, this inhibitory effect is lessened.
Additionally, the epidermal growth factor receptor EGFR is involved in a tissue-specific protein-protein interaction in the lungs with its ligand-protein epiregulin (EREG). EREG, the researchers said, has been shown to impart invasive properties in an EGFR-dependent way.
Yeger-Lotem and her colleagues also noted tissue-specific interactions for causal genes DAG1 and EIF2B1 that also hint at disease mechanisms, in this case for muscular dystrophy and leukoencephalopathy, respectively.
"The distinct features we identified provide a starting point for elucidating the molecular basis of many hereditary diseases and can be further applied to filter the wealth of data being generated by large-scale disease-associations studies," the researchers said.
They also suggested that this sort of comparative tissue analysis and interactome resource "should become a standard framework for interpreting the wealth of disease-related data and for enhancing our understanding of the etiologies of hereditary diseases."