Tuning into epigenetic programming in brain development and disease

Imagine your genome as a musical score — a set of instructions for notes to be played by different instruments. However, instruments and sheet music themselves are not enough to make a performance; musicians bring technique and flair, and conductors coordinate the synchronicity, volume and tempo that create a beautiful symphony. Epigenetics are the conductor and musicians of our cells, coordinating the genes that play the melody or keep the rhythm.

Epigenetic regulation of development and disease

During development, cell identity and fate are determined through epigenetic programming — chemical modifications to chromatin, the building block of chromosomes, comprising DNA and DNA-binding proteins. Three-dimensional (3D) chromatin organization is dependent on histones, a protein complex serving as the spool upon which DNA is wound. By changing properties of chromatin, epigenetic programming dictates which factors can access and bind the DNA, including RNA transcription machinery, subsequently regulating whether the gene is expressed or silenced. Genes that are unimportant for cell function are turned off, as are genes specifically needed during development to coordinate immature cell types. Maintaining this epigenetic programming is crucial for differentiated cell fates and specialized cellular functions. Disruptions to the epigenome have devastating consequences ranging from neurodevelopmental disorders to pediatric cancers.

While adult cancers often result from accumulating mutations in the DNA itself (i.e., playing the wrong notes), many pediatric cancers instead arise by epigenetic dysfunction, or how DNA is regulated. These cancers typically involve mutations in genes described as “writers,” “erasers,” and “readers,” which add, remove or recognize chemical marks on DNA and histones. These modifications control chromatin organization and which genes are active.

Pediatric tumors can also arise from chromosomal rearrangements, which change the proximity of genes to regulatory elements where transcription factors and transcriptional complexes bind to regulate expression. Epigenetic reprogramming can also occur through unknown mechanisms that shift the boundaries between open and closed chromatin, changing what binding sites are available, which transcriptional repressors or activators can access the DNA, and altering gene expression.

In our symphony, cancer-associated changes to the epigenome would be a cacophony of missed or false cues, a lack of volume control, and swapped music. These changes in gene regulation and expression impact cancer biology in ways that include tumor initiation, progression and evolution. Tumor evolution creates significant clinical challenges, contributing to treatment resistance, metabolic adaptability and metastasis.

Because of this reliance on epigenetic mechanisms, researchers are changing their tune and facing the music rather than just reading the score. In other words, science is shifting to study, diagnose and treat pediatric cancers as both a genetic and epigenetic disease, rather than simply a genetic one. To address this challenge, investigators at the St. Jude Center of Excellence in Neuro-Oncology Sciences (CENOS) are investigating epigenetic mechanisms in pediatric brain tumors compared with normal development to understand the pathobiology of these diseases and develop new therapies to treat them.

Histone modifications underlying disrupted development in pediatric gliomas

Suzanne Baker, PhD, Comprehensive Cancer Center deputy director and Department of Developmental Neurobiology member, is tackling epigenetic mechanisms of pediatric brain tumors head-on. Her research focuses on pediatric high-grade gliomas (pHGG), one of the most clinically challenging childhood brain tumors, including diffuse midline glioma (DMG).

Early in her career, Baker co-led efforts that identified that mutations to a histone (H3) cause the majority of DMG cases, affecting certain chemical modifications. As histone protein modifications change how tightly DNA is wrapped, creating open or closed chromatin regions, this could have profound impacts on gene expression.

To understand how this chromatin regulation is important for normal developmental cell fate and tumor formation in the brain, the Baker Lab is studying the epigenetic changes that occur in DMG in diverse patient-derived models.Baker and others have demonstrated that the mutation causing 80% of DMG (H3K27M) occurs at a key histone modification site. This change to the histone impairs a methylation “writer,” which blocks the addition of a methyl group (CH3) at a site important for closing chromatin and silencing neighboring genes. This histone methylation is crucial for development and cell differentiation by silencing early genes and enabling expression of the genes that enable functionally mature neurons and glial cells in the brain. This epigenetic change, therefore, causes aberrant expression of key developmental genes in developing glial cells, creating a less differentiated, stem cell-like epigenetic state that could potentially increase the risk for cancer transformation and tumor formation.

A second histone mutation (H3G34R/V) is responsible for around 30% of a different pHGG, diffuse gliomas in the cerebral hemispheres. Like H3K27M, this mutation blocks a methylation writer, but at a different site (H3K36). This methylation modification has a different biological mechanism, with important functions in recruiting transcriptional co-factors, DNA damage repair and regulating histone modifications at other sites. Rather than creating a stem cell-like state, disrupted H3K36 methylation creates a unique neuronal state and susceptibility to DNA damage. Understanding mechanisms for how this epigenetic dysregulation contributes to tumor biology and identifying opportunities for new therapies is an active research focus for the Baker lab.

Despite impacting the same histone protein, these two genetic variants cause very different epigenetic states and tumor biology, including location, developmental timing and clinical properties.

Now, Baker and her lab are addressing outstanding questions. Why do mutations in histone H3, which can occur in the whole body, only cause gliomas in specific parts of the brain? How do these mutations disrupt glial cell fate? What makes these cells vulnerable to tumor initiation?Future work aims to understand clinical implications of epigenetic changes, such as poor responses to therapies, and apply these research insights to develop more effective therapies for pHGG and improve patient outcomes.

Tuning into the epigenome to diagnose and monitor treatment response in pediatric brain tumors

For Paul Northcott, PhD, CENOS director and Department of Developmental Neurobiology member, tapping into epigenetics is both an opportunity to study mechanisms of pediatric brain cancers and a powerful diagnostic tool. His past research identified how changes to enhancer regulatory elements (i.e., noncoding DNA) can cause medulloblastoma, a tumor in the developing cerebellum.

Since many pediatric tumors lack genetic changes associated with them, researchers and clinicians needed another method besides DNA sequencing to identify tumors. The methylation of DNA by adding a methyl group (CH3) to the cytosine nucleotide held the answer. DNA methylation is crucial for regulating cell identity, both in development and for tissue specification. Patterns of DNA methylation undergo major alterations in cancer, but often retain features of their cell and origin.

Efforts by the Northcott Lab and others described unique DNA methylation patterns for each tumor entity and demonstrated that they are highly accurate for predicting tumor type. “DNA methylation classification has become the gold standard molecular approach for establishing a diagnosis in brain tumors and other cancers,” Northcott said. “The DNA methylome is a more reliable indicator of tumor type and subtype than any other tool at our disposal, making it widely used across both clinical and research settings in neuro-oncology.”

However, a major limitation of this approach has historically been the need for tumor tissue, which requires surgery or invasive biopsies, before diagnostic information is available. Northcott and his team set out to develop a less invasive approach that could work for tumors that are difficult to remove surgically, and to use as a powerful monitoring tool during treatment and follow-up.

Unlike other solid cancers, where blood tests that detect tumor biomarkers are already available, tumor markers for pediatric brain cancers do not reach the blood in detectable quantities due to the blood–brain barrier, a physical structure separating the brain from the rest of the body. Cerebrospinal fluid (CSF) can contain circulating tumor cells and fragmented tumor cell components, such as cell-free DNA, but often at very low levels and mixed with healthy cell material, making tumor-derived DNA challenging to detect. To classify pediatric brain tumors using CSF, Northcott and his team had to develop a way to “turn up the volume” to detect ultra-low levels of tumor material using small fragments, rather than the whole genome, of cell-free DNA.

In a 2026 Nature Cancer paper, Northcott and a team led by co-first authors Katie Han and Kyle Smith, PhD, Department of Developmental Neurobiology, described a tool to overcome these challenges called Methylation-based Predictive Algorithm for CNS Tumors (M-PACT). M-PACT is an AI tool trained on 5,000 DNA methylation profiles across nearly 100 brain tumor entities, which molecularly classifies tumors based on the pattern of DNA methylation released from circulating tumor cells (ctDNA). The tool fills in the gaps of the genome, which are often missing due to DNA degradation outside of the cell, providing accurate diagnostics of tumor signatures from liquid biopsies.

Beyond the diagnostic capabilities of this method, the team noticed that when comparing samples collected from a patient over time, the tool captured how the epigenome evolved during treatment. Currently, they are expanding cohorts and testing the tool for predicting patient outcomes, such as tumor recurrence or failure to respond to specific interventions. In the future, the team hopes to expand M-PACT to cover all types of childhood cancers and study tumor and immune biology and treatment response.

“With further development, we anticipate that the M-PACT workflow developed in the research lab will be used clinically to track treatment response, remission and relapse, making liquid biopsy the new gold standard in pediatric brain tumor diagnostics,”Northcott said.Research from CENOS investigators is driving the field forward by describing how epigenetics causes tumor formation, cancer behavior and disrupted developmental processes in pediatric brain tumors. “These rapid and continual advances are illuminating the contribution of epigenetics to brain tumor biology,” Baker said. “We are developing our understanding of fundamental mechanisms and creating new opportunities for novel therapies and earlier detection.”