Lerner Research Institute News
Read about the latest advances from Lerner Research Institute scientists, including new findings, grant awards, innovations and collaborations.
While specific genetic mutations are known to increase disease risk, the assumption of a linear genotype-phenotype connection often proves overly simplistic. Research has found that genes and their products interact with each other to form a system of complex networks within a cell. Known as the human protein-protein interactome, these molecular networks influence cellular function and can lead to system-wide changes in cells that influence disease mechanisms and treatment resistance.
Harnessing these networks to accelerate drug discovery for genetically heterogeneous diseases, such as cancer, is the goal of Cleveland Clinic’s Feixiong Cheng, PhD, Genomic Medicine Institute. In two recent studies, Dr. Cheng and colleagues employed computational tools to map the genomic profiles of human diseases (i.e., cancer) to the human interactome in order pinpoint druggable targets and therapeutic agents.
Identifying a therapeutic agent for triple-negative breast cancer
In a study published in Cell Chemical Biology, a team of researchers led by Dr. Cheng found that wogonoside, a flavonoid derived from a common Chinese herb, effectively inhibits angiogenesis in triple-negative breast cancer (TNBC).
Accounting for 15-20% of breast cancer diagnoses, TNBC is a rare and aggressive disease characterized by the absence of the three most common receptors (estrogen, progesterone and HER2 protein) known to fuel breast cancer. Because TNBC lacks these actionable targets, therapeutic options are limited.
Previous studies demonstrate that TNBC tumors have increased levels of the proteins SMO (Smoothened) and Gli1 (glioma-associated oncogene homolog 1), which mediate the Hedgehog (Hh) signaling pathway, a key regulator of angiogenesis. Drug therapies that target the Hh signaling pathway, therefore, could effectively impede TNBC tumor growth.
Dr. Cheng’s team developed a computational model based on TNBC genetic variations and known drug-target interactions to identify agents with the potential to disrupt Hh signaling in TNBC cells from a human interactome network model. With this model, they predicted that wogonoside would most effectively inhibit Hh signaling – and thus angiogenesis – in TNBC. In vitro and in vivo experiments validated this prediction and demonstrated that wogonoside’s anti-angiogenic effect stems from its ability to inhibit Gli1 by promoting SMO degradation in a proteasome-dependent mechanism (see Figure 1).
Developing a network algorithm for precision cancer medicine
Although in its infancy, the use of genomics to inform drug discovery and development has proved promising for the advancement of precision medicine. In the genomics era, the drug development process involves a highly integrated systems pipeline that utilizes multi-omics (an analytical approach that integrates data from multiple omics, including genomics, transcriptomics, proteomics, metabolomics, etc.) and computational methods. Recent advancements in these methods have allowed researchers to rapidly identify new drug targets for a genetically heterogeneous disease and investigate if a drug already approved for a different disease might be an effective therapy. However, whether these methods can be generalized across various disease types remains unclear.
In a study published in Nature Communications, Dr. Cheng and colleagues developed and validated an algorithm that integrates large-scale patient genomic data with the human protein-protein interactome to (1) identify cancer type-specific disease modules, or networks of cancer-causing proteins, and (2) predict drug responses in those modules (see Figure 2). Called the Genome-wide Positioning Systems network (GPSnet), this novel network-based methodology found multiple potential anticancer biomarkers for 140 approved drugs across 15 cancer types.
To further examine the accuracy of GPSnet, the researchers experimentally tested an approved cardiac and heart failure drug (ouabain) that was strongly predicted to have an anti-cancer effect in lung adenocarcinoma, a form of non-small cell lung cancer. They determined that ouabain downregulates the expression of HIF1α (hypoxia-inducible factor 1-alpha), a protein that potentially activates the transcription of a gene called LEO1 (RNA polymerase-associated protein LEO1), which is correlated with poor survival rates in lung adenocarcinoma.
Based on these findings, Dr. Cheng and his team conclude that GPSnet has the potential to quicken the pace of target identification and drug development in cancer and other diseases. If broadly applied, the methodology can minimize the translational gap between genomic medicine studies and patient outcomes, thereby achieving a significant goal of precision medicine.
Top: Figure 1. A proposed mechanistic model of wogonoside inhibiting angiogenesis in TNBC. This image was prepared by Cleveland Clinic Center for Medical Art & Photography.
Bottom: Figure 2. A diagram illustrating the concept of GPSnet algorithm for precision cancer medicine. This image was prepared by Cleveland Clinic Center for Medical Art & Photography.