In our latest work published in the journal Neuron, we used powerful analytical tools to map what is known as the “epigenetics” of the retina. We analyzed thousands of genes in mouse and human retinal cells—both normal and cancerous—at different developmental stages, to trace the changes as they progressed from immature progenitor cells into specialized retinal nerve cells.

Epigenetics refers to the chemical changes that the cell makes to DNA to switch genes on or off. We’ve long known that the DNA sequences in genes constitute a “genetic code.” But now we know that there’s also an “epigenetic code” for each gene in each type of cell that is established during development.

Figuring out that epigenetic code for the body’s cells and tissues is critical, because we now know that mutations in the epigenetic machinery can contribute to many cancers. For example, it’s known that mutations in genes can cause inherited adult eye diseases such as retinitis pigmentosa. But clinicians have encountered patients who don’t have any known mutations in genes. So they wonder whether the cause was a mutation in a region beyond the genes that acts as an enhancer for the genes. The epigenetic profiling that we performed helps us to identify those enhancers in the retina.

An illustrated eye with a DNA double helix strand and chromosomes reflected in the retina.

Chemical changes deep within cells of the retina hold the keys to unlocking the causes of retinoblastoma and other cancers.

Our epigenetic mapping also adds significantly to our understanding of retinoblastoma—the most common eye cancer in children. It’s been hotly debated for decades which retinal cell gives rise to retinoblastoma. This understanding is critical if we’re to improve treatment. The problem is that you can’t just examine the tumor and say it looks like one cell type or another, because it may no longer resemble the cell it came from. The origin cell might just be a biological wolf in sheep’s clothing—changing its appearance as it becomes cancerous.

When we traced the epigenetic development of retinoblastoma cells, we discovered what, to me, was a real surprise. I thought the retinoblastoma cells would show the first signs of turning cancerous during early development, when they are immature progenitor cells that are dividing rapidly to make new tissue. But we discovered that they became tumors during a very specific window in the middle of development, when the cells are transitioning between rapid growth and differentiating into mature retinal neurons. This finding is important, because it hints that this is a moment when the cell is vulnerable to cancerous mutation.

What’s exciting is that we can apply this same kind of epigenetic profiling to other cancers to trace the cell of origin. We have already begun to explore the epigenetics of rhabdomyosarcoma, a childhood cancer of the muscle. We’re comparing the tumors with normal muscle cells to trace the cancer’s developmental origin. As with retinoblastoma, this understanding will help us identify pathways that were altered in the cancer and that we could target with drugs.

The retina also offers us an incredible window into neural development. Retinal cells are ideal for studying how nerve cells develop because they are so accessible, unlike the neurons deep in the brain. Over the past hundred years or so, scientists have amassed an enormous wealth of knowledge about how the retina is formed during embryonic development. We know when each cell type is made during development, how many cells are proliferating at each stage and how many cells are differentiating.

Michael Dyer, Beth Stewart, and Alberto Pappo talking at a table.

Michael Dyer, PhD, speaks with Beth Stewart, MD, of the St. Jude Department of Oncology, and Alberto Pappo, MD, director of the St. Jude Solid Tumor Division.

As an example of what we can accomplish, in the Neuron paper we reported mapping the retinal epigenome in three dimensions. Other researchers had discovered that retinal rod cells organize the genes in their nucleus into a tight central core, with two concentric outer rings.

What we found for the first time was which genes are in that condensed, presumably inaccessible, central core and which are in the outer rings. We found that the organization is not random; it’s the same in any rod cell, but it’s different in neighboring cells. So there’s a very clear program for how cells organize their nuclei in three dimensions. And we’re trying to understand the functional significance of this organization. For example, a cell may need to keep a tumor suppressor gene handy in an outer ring, in case the cell gets a cancer mutation and needs to stop the cell from becoming a tumor.

The analogy I use is like packing a suitcase. You can’t take all the clothes in your closet on a trip, so if you’re going somewhere cold, you pack your warm clothes and leave the summer clothes in your closet. Similarly, we’re trying to figure out how neurons decide where to pack their genes in their nucleus.