More than 200 metabolic enzymes sit directly on DNA, and cancer types each have their own pattern

Seven percent of all proteins found clinging to human DNA are metabolic enzymes, molecules traditionally assigned to the mitochondria and cytoplasm where they generate cellular energy. That figure, drawn from a systematic survey of 54 cell lines, suggests that the boundary between metabolism and genome regulation is far more porous than textbooks describe.

The study, published in Nature Communications, comes from the Centre for Genomic Regulation in Barcelona. Led by Dr. Sara Sdelci, the team examined 44 cancer cell lines and 10 healthy cell types from ten different tissues, cataloging every metabolic enzyme physically attached to chromatin, the complex of DNA and proteins that constitutes the working form of the genome.

A nuclear metabolic fingerprint

The most striking finding was not just the presence of metabolic enzymes in the nucleus, but their specificity. Different cell types, tissues, and cancer types each displayed a distinct pattern of which enzymes appeared on their chromatin, in what quantities, and in which combinations. The researchers call this pattern a nuclear metabolic fingerprint.

The fingerprints differed dramatically between cancer types. Oxidative phosphorylation enzymes, the machinery responsible for generating most of a cell's energy, were common residents of the nucleus in breast cancer cells but largely absent in lung cancer cells. When the team examined tumor samples from patients rather than cell lines, the same tissue-specific patterns held.

What are they doing there?

That is the question the study raises more than it answers. Several possibilities exist. The enzymes could be catalyzing metabolic reactions locally, producing or consuming small molecules right next to the genes that need them. They could be moonlighting as gene regulators, directly turning genes on or off through protein-protein interactions. Or they could be providing structural support to chromatin organization.

The researchers investigated one group of enzymes involved in producing nucleotides, the building blocks of DNA. These enzymes clustered around chromatin when DNA was damaged, suggesting they participate in genome repair by supplying raw materials at the site of injury.

One enzyme, IMPDH2, showed particularly revealing behavior. When the researchers confined it to the nucleus, it helped maintain genome stability. When confined to the cytoplasm, it affected different pathways entirely. Same protein, different location, different function.

Implications for drug resistance

This dual identity of metabolic enzymes creates a potential problem for cancer treatment. Many chemotherapy drugs target either a tumor's metabolic activity or its DNA repair mechanisms, treating these as separate vulnerabilities. If metabolic enzymes are physically present on chromatin and participating in DNA repair, those two categories may be more intertwined than clinicians assume.

That overlap could help explain a persistent clinical puzzle: why tumors from different tissues, even when carrying identical mutations, often respond very differently to the same chemotherapy. If breast cancer nuclei are loaded with oxidative phosphorylation enzymes while lung cancer nuclei are not, the same DNA-damaging drug may encounter fundamentally different repair environments.

The size problem

There is also a basic cell biology mystery here. Many of the metabolic enzymes found on chromatin are large, bulky protein complexes that should not fit through nuclear pores, the channels that control traffic between the cytoplasm and the nucleus. Nuclear pores have well-characterized size limits, yet enormous enzyme complexes are somehow getting through. Whatever mechanism cells use to bypass those limits has not been identified, and finding it could itself provide a therapeutic target for controlling nuclear metabolic activity in cancer.

One enzyme at a time

The study provides the first global map, but not yet a functional atlas. As first author Dr. Savvas Kourtis noted, each of the 200-plus enzymes may have its own unique nuclear function, and these will need to be investigated individually. Some may be actively catalyzing reactions, others may be dormant passengers, and still others may serve roles that have no precedent in current biochemistry.

In the long run, mapping the function of each enzyme on chromatin could reveal new biomarkers for cancer diagnosis and new vulnerabilities for drug targeting. But that work is just beginning.