EZH2, short for enhancer of zeste homolog 2, is best known as a catalytic component of the Polycomb Repressive Complex 2, or PRC2. In that classical role, EZH2 adds methyl groups to histone H3 at lysine 27, a chemical modification commonly designated H3K27me3. Histones are proteins around which DNA is wrapped, and chemical marks on them can influence whether nearby genes are accessible for transcription. EZH2 is often overexpressed or abnormally activated in cancers, where it has generally been associated with the repression of genes that restrain cell proliferation or promote differentiation. Patients whose tumours contain high levels of EZH2 frequently have poorer clinical outcomes. Yet the exact ways in which excess EZH2 strengthens tumour-forming ability have not been fully explained. The new work expands that picture by showing that EZH2 can participate in a noncanonical, or nontraditional, function: rather than acting only as a chromatin-modifying repressor, it helps stimulate a gene-expression programme connected to lipid metabolism.

The key partner in this process is SREBP2, or sterol regulatory element-binding protein 2. SREBP2 is a transcription factor that monitors and controls cellular cholesterol production. When cells require more cholesterol, SREBP2 can become activated and move into the nucleus, where it binds regulatory DNA sequences near genes involved in cholesterol uptake and synthesis. Among its major targets are genes in the mevalonate pathway, a series of enzymatic reactions that converts acetyl-CoA into cholesterol and other sterol-related products. Cholesterol is not merely a structural component of cell membranes. It also contributes to membrane organization, intracellular signalling and the production of steroid-related molecules. Rapidly dividing cancer cells can place unusually high demands on these systems as they build new membranes and adapt to stressful environments. The study indicates that EZH2 and SREBP2 cooperate to drive high expression of mevalonate-pathway genes, effectively linking an epigenetic cancer-associated protein to a metabolic switch that can increase the supply of cholesterol.

The researchers describe a direct molecular connection between the two proteins and the transcriptional machinery that activates cancer-related genes. According to the study, transcriptional activation domains within EZH2 and SREBP2 bind directly to p300, a well-known coactivator that helps turn genes on. p300 can modify histones and other proteins through acetylation, a process that often promotes a more transcriptionally permissive chromatin environment. In this model, the EZH2–SREBP2 partnership is not simply bringing two regulatory proteins into proximity; it is also recruiting a coactivator capable of strengthening gene activation. The result is a functional complex that supports expression of genes in the mevalonate pathway and activates proto-oncogene programmes. Proto-oncogenes normally contribute to controlled growth and survival, but when inappropriately activated they can promote malignant transformation. This mechanism provides a possible explanation for how high EZH2 levels can support cancer even when its tumour-promoting activity does not fit the classic PRC2-mediated model of gene repression.

The implications are especially striking because cholesterol metabolism has often been viewed as a supporting feature of cancer biology rather than as a central output of EZH2 activity. Tumours rewire metabolism to obtain energy, construct cellular components and survive conditions such as nutrient limitation or low oxygen. Increased cholesterol synthesis may help supply the membrane material required for proliferation, while mevalonate-pathway intermediates can influence signalling and protein modification. The study’s findings place EZH2 near the top of that metabolic control system, where it may help SREBP2 maintain the expression of multiple biosynthetic genes at once. This is different from blocking a single enzyme downstream in the pathway. A regulatory partnership that controls a broad gene set could, in principle, produce a larger effect on tumour biology—but it could also create challenges, because cholesterol production is essential to normal cells. The research therefore points to a vulnerability, not yet a finished treatment strategy, and further work would be needed to determine how selectively the pathway can be disrupted in cancer.

To test whether this noncanonical function could be targeted, the researchers used proteolysis-targeting chimeras, widely known as PROTACs. These are engineered molecules designed to bring a target protein into contact with an E3 ubiquitin ligase, part of the cell’s protein-disposal system. Once recruited, the target can be tagged with ubiquitin and sent to the proteasome, a large molecular machine that breaks down proteins. Unlike conventional inhibitors, which generally occupy a functional pocket and block activity, PROTACs can remove a protein from the cell and may continue acting catalytically as long as the degradation machinery remains available. In the study, independent EZH2-targeting PROTACs degraded EZH2 and, notably, also reduced SREBP2. This dual effect suppressed SREBP2-associated gene-expression programmes, including those linked to cholesterol biosynthesis, and inhibited tumour growth. The result suggests that eliminating EZH2 may dismantle the regulatory partnership more effectively than simply blocking one of its biochemical activities.

The observation that EZH2-targeting PROTACs affect both proteins is central to the study’s therapeutic significance. If EZH2 supports tumour growth through several distinct functions, an agent that removes the protein could potentially block more than a single catalytic activity. Degradation of EZH2 may weaken its association with SREBP2, reduce the availability of the coactivator p300 at relevant genes and collapse the transcriptional programme that sustains the mevalonate pathway. The accompanying loss of SREBP2 would further limit the cell’s ability to activate cholesterol-biosynthesis genes. Together, these effects could explain why the PROTAC strategy inhibited tumour growth in the researchers’ experiments. However, the findings do not establish that such compounds are ready for clinical use, nor do they show that every cancer with high EZH2 depends on the same mechanism. Tumours are genetically and metabolically diverse, and cholesterol production is also vital in healthy tissues. The future challenge will be identifying cancers most reliant on the EZH2–SREBP2 axis while minimizing damage to normal metabolism.

The work ultimately shifts the way scientists may think about EZH2 in cancer. Rather than treating the protein solely as an epigenetic repressor that silences protective genes, the study presents it as a versatile regulator capable of joining a transcriptional complex that actively promotes metabolic and oncogenic programmes. Its partnership with SREBP2 creates a bridge between chromatin biology, gene activation and lipid metabolism, revealing how a cancer-associated protein can influence the supply of molecules needed for tumour expansion. The findings also illustrate why protein degradation strategies are attracting attention: destroying a regulatory protein can expose vulnerabilities created by its interactions, not just those associated with its best-known enzymatic function. By identifying cholesterol biosynthesis as a downstream output of EZH2–SREBP2 cooperation, the researchers offer a new framework for understanding tumour metabolism and a potential route for therapeutic development. The axis is not yet a proven universal weakness, but it may represent a molecular Achilles’ heel in cancers that depend on elevated EZH2 activity and SREBP2-driven cholesterol production.

Anaplastic thyroid carcinoma, or ATC, is rare but exceptionally aggressive. Unlike many differentiated thyroid cancers, ATC can grow rapidly, invade nearby structures, and spread before treatment has had time to take effect. The BRAF V600E mutation is one of the most important genetic alterations in a subset of these tumors. It changes the BRAF protein so that it remains abnormally active, driving the MAPK signaling pathway, a molecular chain that regulates proliferation, survival, and cellular behavior. Dabrafenib is designed to inhibit mutant BRAF, cutting off this signal at a critical point. Yet cancer cells are not passive targets. Under therapeutic pressure, they can adapt, rewire signaling networks, and alter the way they process nutrients.

The central insight of the new work is that drug resistance is not solely a matter of changing receptors or acquiring additional mutations. It can also involve a shift in the cell’s internal economy. Cancer cells require a continuous supply of ATP for energy, reducing equivalents to control chemical reactions, and carbon-based molecules for constructing DNA, proteins, membranes, and other cellular components. Glucose is a major source of these materials, but its carbon can be redirected into multiple biochemical pathways. PHGDH sits at the entrance to one such branch: the serine synthesis pathway. It diverts the glycolytic intermediate 3-phosphoglycerate away from energy production and toward the generation of serine and related metabolites.

Serine is more than a simple amino acid. It contributes to the production of glycine, nucleotides, phospholipids, and glutathione, an important antioxidant system. Through these connections, the serine synthesis pathway can influence how cells replicate their genomes, build new membranes, and withstand oxidative stress. PHGDH also participates in the cell’s management of redox balance, helping shape the availability of molecules required for maintaining chemical stability during rapid growth. In tumors exposed to a BRAF inhibitor, these functions may become particularly valuable. If dabrafenib suppresses the signaling program that normally supports proliferation, a resistant cell may compensate by strengthening metabolic pathways that preserve survival and biosynthetic capacity.

Kim and You’s study identifies PHGDH inhibition as a means of disrupting that compensation. The reported effect is described as metabolic rewiring: rather than simply adding another brake to the same signaling pathway, PHGDH inhibition changes the flow of nutrients through the cancer cell. This distinction matters because resistant tumors often survive by routing around a blocked pathway. A second drug that targets the same route may have limited impact if the cancer has already activated a parallel mechanism. By interfering with serine-related metabolism, however, PHGDH inhibition may remove the raw materials or redox support that resistant BRAF-mutant cells require to remain viable under dabrafenib treatment.

The approach also illustrates why combination therapies are increasingly being designed around the relationship between signaling and metabolism. Oncogenic BRAF signaling can alter glucose uptake, mitochondrial activity, amino-acid use, and the expression of metabolic enzymes. At the same time, metabolic changes can feed back into signaling by modifying the cellular energy state and the chemical environment in which proteins operate. A tumor treated with dabrafenib may therefore not be understood as a static collection of mutated cells. It is a dynamic system responding to stress. The combination of BRAF blockade and PHGDH inhibition aims to attack both the growth command and the adaptive fuel network that can help cancer cells survive its suppression.

From a technical perspective, the strategy may produce several layers of stress at once. Restricting PHGDH activity can reduce the ability of cells to generate serine through glucose-derived intermediates. That shortage may affect nucleotide synthesis, limiting the materials needed for DNA replication and repair. It may also weaken phospholipid production, which is essential for expanding cellular membranes during division. At the same time, reduced support for glutathione production could leave tumor cells more exposed to reactive oxygen species. These chemically reactive molecules arise naturally during metabolism and can increase when cells are under therapeutic pressure. A cancer cell that can no longer buffer oxidative damage may cross a threshold leading to growth arrest or cell death.

The findings are especially notable because dabrafenib resistance remains a major obstacle even when a tumor carries a mutation that appears directly druggable. Targeted therapies can produce striking responses, but those responses may be temporary. Some resistant cells reactivate MAPK signaling through alterations elsewhere in the pathway, while others engage receptor tyrosine kinases, change cell identity, or enter a slower-growing state that tolerates treatment. Metabolic rewiring adds another dimension to this problem. It suggests that the cells surviving therapy may not merely be genetically different; they may also be physiologically different, using nutrients in a way that makes them less dependent on the pathway originally targeted.

The study’s implications extend beyond thyroid cancer, although its immediate focus is BRAF V600E anaplastic thyroid carcinoma. PHGDH is overexpressed or relied upon in several cancer contexts, and the serine synthesis pathway has attracted attention as a potential therapeutic target. Still, an experimental result in a cancer model does not automatically establish a safe or effective treatment for patients. Serine metabolism is also important in normal tissues, and the therapeutic challenge will be to determine whether PHGDH can be inhibited strongly enough to affect resistant tumor cells without causing unacceptable toxicity. Future work will need to clarify which tumors are most dependent on PHGDH, identify biomarkers that predict response, and establish the appropriate dose, timing, and sequence for combining a PHGDH inhibitor with dabrafenib.

The broader message is that cancer resistance may be defeated by targeting not only what tumors signal, but also how they survive. In BRAF V600E anaplastic thyroid carcinoma, the work by Kim and You positions PHGDH as a metabolic pressure point that becomes important when dabrafenib pushes cancer cells into an adaptive state. If these findings are confirmed in additional models and ultimately in clinical studies, the combination could help convert a resistant tumor’s survival strategy into a liability. The research offers a vivid example of modern oncology’s expanding battlefield: the fight is no longer confined to mutated genes and signaling proteins, but includes the intricate metabolic networks that keep malignant cells alive.

Subject of Research: PHGDH inhibition and metabolic rewiring as a strategy to overcome dabrafenib resistance in BRAF V600E anaplastic thyroid carcinoma.

Article Title: PHGDH inhibition overcomes dabrafenib resistance through metabolic rewiring in BRAF V600E anaplastic thyroid carcinoma.

Article References: Kim, S.Y., You, MH. “PHGDH inhibition overcomes dabrafenib resistance through metabolic rewiring in BRAF V600E anaplastic thyroid carcinoma.” Cell Death Discovery (2026). https://doi.org/10.1038/s41420-026-03293-7

Image Credits: AI Generated

DOI: https://doi.org/10.1038/s41420-026-03293-7

Keywords: PHGDH, dabrafenib resistance, metabolic rewiring, BRAF V600E, anaplastic thyroid carcinoma, serine synthesis pathway, targeted therapy, cancer metabolism