How does the gene transcription scheme make us different? How is it a precursor to disease? How is this knowledge applied to personalized genomics?
Dave Burbank
Dave Burbank


Using a combination of innovative techniques, the Charles Danko lab looks at regions on the genome to determine how they control gene expression.
Dave Burbank
Dave Burbank

Danko Collage_edit.jpg

The Danko lab took a Cornell technology to a new extraordinary level, achieving a groundbreaking technique that lets them work with solid tissue not just cell cultures, which could have a tremendous impact on precision medicine.
Beatrice Jin; Dave Burbank
Beatrice Jin; Dave Burbank


Referring to his lab’s study of glioblastomas in brain tumor patients, “You can pick out the regions of the genome that control gene expression and then differences in gene expression among the different patients.”
Dave Burbank
Dave Burbank


“I really do this for that moment when you’ve just finished an analysis and things come together. You understand something about the world that nobody else does in quite the same way.”
Dave Burbank
Dave Burbank

Our Genome, Managing Gene Regulation

by Caitlin Hayes

How do our different cells, with their varied morphologies and functions, arise from the same DNA sequence? Charles G. Danko, Biomedical Sciences/Baker Institute for Animal Health, points to variations in how our genes are used or how that use of genes is regulated.

“I’m interested in everything that relates to the molecular biology of the nucleus,” Danko says. “If I had to choose one theme for our lab, it’s understanding how our genomes encode various programs of gene regulation.”

Specific regions of our genome, Danko explains, are dedicated to controlling which genes are expressed—converted into mRNA by a process known as transcription—in each cell type. Variation in the levels of mRNA leads to the different characteristics of each cell, each person, even each species. Up to now, however, researchers have had trouble locating the position of regulatory areas in order to study them. Danko’s lab uses innovative techniques, both molecular tools and machine learning methods, to find these regions on the genome and ask how they control gene expression.

Brain Tumors—Where the Cells Go Wrong During Transcription

An assay called PRO-seq (Precision Run-on Sequencing), developed by John T. Lis and Hojoong Kwak, Molecular Biology and Genetics, allows researchers to detect where transcription of DNA into mRNA is happening in the genome at a given time. Until recently, the method was somewhat limited, working best in cell cultures. Danko’s group has been able to adapt the technique to work better with solid tissue samples, which is invaluable for the study of disease.

With the new technique, ChRO-seq (chromatin run-on and sequencing) and complementary machine learning tools to analyze the data, Danko can now take limited clinical samples of diseased tissue and find where transcription is happening. By comparing this data to that of non-diseased tissues, he can possibly pinpoint where the behavior adopted by cells goes awry.

Danko’s group has recently used these techniques to make groundbreaking observations in a group of 20 sample glioblastomas, deadly brain tumors. They found that the tumors were more similar than they expected to the tissue of origin, or normal brain tissue. “Basically, it’s a normal brain with a few regulatory programs activated that give it the malignant behavior,” Danko says.

Impressively, Danko’s team was able to identify these destructive regulatory programs. “You can pick out the regions of the genome that control gene expression and then differences in gene expression among the different patients,” Danko says.

“There are a lot of applications for this kind of analysis in personalized genomics and medicine that I think are really important.”

These regions in the genome, called enhancers, can be traced further to associated proteins called transcription factors. In tightly coiled DNA, the transcription factors bind to DNA in the enhancer region and increase transcription in the gene that the enhancer activates. “Specific transcription factors therefore drive groups of genes that then change the behaviors of cells in ways that affect the outcome for the patient,” Danko says. “Eventually I think we’ll be able to target these regulatory programs, and just being able to find them is a big advance in this field. There are a lot of applications for this kind of analysis in personalized genomics and medicine that I think are really important.” Danko’s group is now working to apply these methods to a larger cohort of glioblastoma samples—70 in all.

Manipulating Gene Regulatory Programs to Better Understand Them

With the ability to locate these regulatory programs in samples, Danko now wants to manipulate them in order to see precisely how they work. To do this, his group is using CRISPR epigenome editing technologies, where a non-active Cas-9 enzyme binds to DNA and changes the activity of that local region of the genome.

“One way to think about these dead Cas-9 tools is that you’re designing a new transcription factor, and you can tailor the activity of that transcription factor in any way you want by changing which functional region it’s associated with,” Danko says. “It’s an extremely powerful tool to get at what different transcription factors do and how they work in concert to achieve a specific outcome for different genes.”

In a new set of experiments, Danko is working toward a better fundamental understanding of how the interplay of transcription factors and enhancers impacts gene expression. “There are a lot of regions that bear this characteristic signature. They look like they control gene expression, but some are very far from genes; and we’re not sure in many cases what their function is,” he says. “What are they doing? How do they come to actually have a function? Those are the kinds of things we want to address.”

Up to now, the only way to study these elements was to isolate sections of the DNA and move them to an artificial context. With that method, researchers lost access to how the regions of DNA are interacting in the native environment within the genome. Using the CRISPR epigenome editing tools, along with another adaptation of PRO-seq, Danko will be able to knock out the enhancers to see what patterns emerge. He’ll even be able to tune specific regulatory programs to modify the levels of gene expression. This allows his team to track the impact and function.

The Evolution of Gene Regulation

Danko has also looked at how gene regulatory systems—responsible for many of the differences between species—evolved. “If you look at natural variation between species, or even within species, a really important source of that variation is changes in gene expression levels rather than changes in amino acid sequence,” Danko says. “So basically, the questions here are these: How do DNA sequence changes to these regulatory regions arise? How do those changes actually affect the transcription of genes located nearby?”

Together with Lis and Adam C. Siepel (Cold Spring Harbor Laboratory), Danko has found that the sites far from genes—distal enhancers—and some enhancers close to genes—are quickly evolving. They often differ between humans and chimpanzees, which are identical in more than 99 percent of DNA. This rapid change may mean that they have no significant effect on the fitness of the organism. “But evolution is a random process, so you almost expect there to be a test bed of new functional elements that maybe don’t do anything in that particular species. They are sort of like tryouts for a new function,” Danko says.

The team found that the rate of change in these enhancers is much faster than the rate of change in the transcription of protein-coding genes. There’s also redundancy, where multiple enhancers target the same gene and compensate to some degree for changes in one another. This is the ideal model for natural selection. Gene expression is highly constrained by redundancy but can also change rapidly. This allows organisms to quickly adapt if new mutations in regulatory sequences are advantageous.

“This is the perfect physical regime to be working in if you want to set up a system where you can undergo quick evolutionary changes, but you still need to constrain the level of specific genes in most cases,” Danko says.

Inquiries and the Rewards of Analysis

Danko has always been interested in computers and learned to program at a young age. The questions that led him into biology were about the neuroscience of the mind and what makes consciousness. “Those questions are starting to be answerable, but when I got into this area, they were not tractable,” he says. “But we had just sequenced the human genome. We had draft assemblies of the mouse, and the rat was on the horizon, and I started to wonder about the related questions: What are the physical properties of an organism? How is that controlled at a molecular level?”

A broad range of inquiries have stemmed from these initial questions and intersected with Danko's ability and interest in computing. While Danko loves asking the questions, he says the biggest rewards are the fruits of analysis. “I really do this for that moment when you’ve just finished an analysis and things come together. You understand something about the world that nobody else does in quite the same way. You’ve made a change in your own thinking about the way the world works. Nurturing that in students and postdocs, I find very rewarding, too.”