First-ever human colon model identifies six unknown cancer driver genes

Researchers engineered a first-of-its-kind ex vivo model of the human colon, then put it to the test to find genes associated with progression of colon cancer. The model identified 17 known cancer driver genes as well as 6 genes that had never been associated with colon cancer progression.

This test demonstrated that their colon cancer model is highly useful for studying cancer development, the researchers concluded in their July 11, 2016 article in the journal Nature Biotechnology.

Currently, scientists routinely use genome-wide association studies to identify mutations in tumors; however, detecting the roles of these genetic alterations in tumorigenesis is still a challenge. So these researchers set out to develop a 3-D model that physiologically replicates the colon.

“You can’t really do experiments very well on human tissue, so having a human system, which allows you to look at the genetics in the context of a controlled environment, is a fairly powerful technique,” said the study’s co-senior author Michael Shuler, PhD, the Samuel B. Eckert Professor of Engineering at Cornell University’s Meinig School of Biomedical Engineering, in Ithaca, NY.

The model that Dr. Shuler and colleagues created goes well beyond a 2-D cell culture because it includes not only tissue from the colon, but also extracellular matrix. In addition, the model preserves the tissue’s normal 3-D architecture. Combined, this provides a more true-to-life environment in which to test and examine tumor pathogenesis, the researchers explained.

Their model is much smaller than an actual colon, though—only about 5 cm³. To engineer it, they started with a sample of normal human colon tissue from which they completely removed the cellular components while retaining intact the tissue architecture, main vasculature, and the muscularis mucosa of the original organ.

Next, they repopulated the tissue with epithelial cells, endothelial cells, and myofibroblasts obtained from colonoscopy patients and from commercial samples. “What you’re really trying to do is provide a micro-environment that encourages the appropriate expression of the genes in the system,” Dr. Shuler said.

Once they had a ready model of the colon, it was time to test how well it demonstrated cancer progression. To do so, they recellularized the colon model with epithelial cells carrying gene muta­tions known for colorectal cancer progression. Malignancy occurred quickly, appearing within 6 to 7 weeks after recellularization (which further showed the value of this ex vivo model as a complement to current cancer models, the authors noted).

Next, the investigators performed a mutagenesis screen to identify and track the genetic changes that occurred inside the colon model, which they found were consistent with typical early stage colorectal cancer.

Of course, there’s no way to tell for sure whether the model provides an exact replica of colorectal cancer progression, but “it gives you a human-based system to characterize some of the key steps in advance-stage colon cancer, and that is something that hasn’t been possible,” Dr. Shuler noted.

Further testing with DNA sequencing identified 38 invasion-driver genes from 21 invasive neoplasias. Of these 38 genes, 17 had already been implicated in colon cancer progression while 6 had never been associated with the progression of colorectal cancer.

“These results demonstrate the utility of our organoid model for studying cancer biology,” the investigators concluded.

“The recellularized human colon provides an exciting new model for identifying genes that are mutated during the earliest step in tumor metastasis,” said co-senior author Nancy Jenkins, PhD, Professor of Oncology at Houston Methodist Research Institute, in Houston, TX.

“Our hope is that a better understanding of the genetics of tumor metastasis will lead to better molecular targeted therapies and/or biomarkers for the treatment of colon cancer,” she added.