For decades, retinoic acid has been a key part of the arsenal against the childhood cancer neuroblastoma. For just as long, scientists have wondered exactly how it works.

The answer, Dana-Farber researchers have found, is by reprogramming the activity of two crucial pairs of genes with such precision that the drug almost seems to have been designed expressly for that purpose. The findings, reported in a new study in the journal Science Advances, not only solve a longstanding puzzle about retinoic acid’s mode of action in neuroblastoma but also reveal why the drug is effective in some patients but not others.

Neuroblastoma is a pediatric tumor of the peripheral nervous system, the network of nerves outside the brain and spinal cord. About 30 years ago, researchers found that compounds derived from retinoic acid, called retinoids, slowed the growth of neuroblastoma cells in laboratory dishes and caused them to differentiate — to look and act more like normal nerve cells. The discovery led to clinical trials of retinoic acid in patients with neuroblastoma and to its use, along with intense chemotherapy, in patients with high-risk forms of the disease.

Some of the hardest neuroblastomas to treat are those with surplus copies of MYCN, a gene whose overabundance prevents the tumor cells from growing and maturing normally. For Dana-Farber scientists Mark Zimmerman, PhD, and A. Thomas Look, MD, and colleagues at St. Jude’s Children’s Research Hospital and other institutions, this knowledge provided an entryway for an investigation of the deeper mechanism by which retinoic acid brings neuroblastoma cells to heel

                                                                  Mark Zimmerman, PhD, and A. Thomas Look, MD.

“MYCN is very appealing as a therapeutic target because its overexpression inhibits neuroblastoma cells from differentiating. So far, however, efforts to develop drugs that specifically target the gene haven’t been successful,” says Zimmerman, the first author of the new paper. “In our study, we show that retinoids suppress MYCN by rewiring the circuitry that controls transcription of the gene” — the process of converting genetic information from DNA to RNA.

Like all genes, MYCN is transcribed when a chromosome forms a loop, bringing a region known as an enhancer alongside a section known as a promoter, which sits close to the gene itself. The point where the loop closes — where the enhancer and promoter are now neighbors — becomes the gathering spot for a variety of proteins, including those known as transcription factors. The transcription factors and other proteins entice an enzyme to the promoter site, where the process of converting the gene’s DNA to RNA begins.

In the new study, researchers found that most neuroblastoma cells with an overabundance of MYCN genes depend for their growth and survival on a small group of genes that hold the code for certain key transcription factors. And they found that treating such cells with all-trans retinoic acid (ATRA) had a very specific effect on those transcription factor genes — an effect exquisitely calibrated to be detrimental to the continued growth and immature behavior of the cells.

“We found that treatment with ATRA suppressed the expression of PHOX2B and GATA3 — genes for transcription factors that are essential for MYCN expression and neuroblastoma cell proliferation,” Zimmerman states. “It also led to an increase in expression of the genes MEIS and SOX4, which code for transcription factors that cause cells to differentiate.”

If the human genome is thought of as a control room with levers to raise or lower the activity of thousands of genes, ATRA is able to adjust at least four of those levers in such a way that is particularly deleterious to neuroblastoma cells with excess copies of MYCN.

The researchers went on to show how ATRA exercises such control over these critical genes.

The genome contains regions where multiple enhancers are clustered like beads on a necklace. These regions, known as super-enhancers, deliver an extra jolt of activity to the genes they regulate.

Zimmerman, Look, and their colleagues discovered that ATRA alters the way super-enhancers position themselves with PHOX2B, GATA3, MEIS1, and SOX4, such that the first two are suppressed and the latter two are more highly expressed.

The findings also shed light on why some neuroblastomas with overactive MYCN are impervious to treatment with ATRA. “We showed that neuroblastomas that activate MYCN through genomic structural arrangements, called enhancer hijacking, cannot be reprogrammed by ATRA to differentiate into a more stable state,” Zimmerman says.