Positive feedback regulation between glycolysis and histone lactylation drives oncogenesis in pancreatic ductal adenocarcinoma

Pancreatic cancer and sugar metabolism seem like they belong to different conversations. One is a disease, and the other is biochemistry. But Li and colleagues have found that in pancreatic ductal adenocarcinoma, or PDAC, these two things operate in a feedback loop, each one accelerating the other. The more fuel the cancer burns, the more aggressively its genes get rewired. And the more those genes are rewired, the faster it burns. Researchers traced exactly how that loop works, and it points to a vulnerability that had not been named before. PDAC accounts for more than eighty percent of pancreatic tumors, and by the numbers, it is among the deadliest cancers we know. The five-year survival rate is grim, and despite a decade of therapeutic effort, mortality rates have not considerably declined. What makes PDAC so hard to kill is the combination of two properties that have mostly been studied separately: metabolic reprogramming and epigenetic rewiring. On the metabolic side, PDAC cells rely on glycolysis at a furious pace — even in the presence of oxygen, a behavior called the Warburg effect — and they produce enormous amounts of lactate as a result. Lactate was long considered a metabolic waste product. What this paper establishes is that lactate is something more: a chemical signal that reaches directly into the cell nucleus and changes which genes are active. The mechanism is called histone lactylation. Histones are the proteins that DNA wraps around, and chemical tags on histones act like switches, turning genes on or off. Twenty-eight lactylation sites have been identified on histones, and Li and colleagues focused on one in particular: H3K18la, which means lactylation at lysine 18 on histone H3. When the team measured H3K18la levels in PDAC tissue samples and compared them to normal pancreatic tissue, the mark was clearly elevated in tumors. When they stratified patients by H3K18la level using a Kaplan-Meier survival analysis, high H3K18la correlated with poor overall survival. This isn't just a molecular curiosity — it's a prognostic signal written in the chromatin itself. Every epigenetic mark needs an enzyme to install it and one to remove it. Li and colleagues identified P300, an acetyltransferase, as the writer for histone lactylation in PDAC, and HDAC2 — histone deacetylase 2 — as the eraser. The experimental logic is clean. When P300 was knocked down using small interfering RNA, or siRNA, or inhibited with the compound C646, both global pan-lysine lactylation and H3K18la dropped significantly in two PDAC cell lines, MIA PaCa-2 and AsPC-1. Critically, these reductions happened even when cells were flooded with extra sodium lactate at ten millimoles per liter — meaning P300 activity, not substrate availability, is the limiting factor. P300 inhibition also suppressed proliferation and migration in functional assays. On the other side, broad HDAC inhibition with the compound trichostatin A raised lactylation, and when specific HDAC subtypes were tested, only HDAC2 overexpression reliably reduced H3K18la. HDAC1 and HDAC3 overexpression had no effect. Two enzymes, two roles, cleanly separated by experiment. So P300 writes the lactylation mark and HDAC2 removes it. But the bigger question is: what does the mark actually do? To answer that, the team ran a genome-wide search using CUT-and-Tag — a technique that maps where a specific histone modification sits across the entire genome — combined with RNA sequencing. They applied the glycolysis inhibitor Oxamate to MIA PaCa-2 cells and looked for genes whose promoters lost H3K18la and whose transcripts fell at the same time. About fifty-three percent of H3K18la peaks were located in promoter regions, right where you'd expect an active transcriptional mark to sit. The intersection of the chromatin data, the RNA sequencing data, and a public database of genes overexpressed in PDAC produced fourteen candidate targets. Four mapped to the cell-cycle pathway. Two of those four — TTK and BUB1B — sat upstream of M phase, the stage of cell division where chromosomes must be accurately pulled apart. TTK, also called MPS1, and BUB1B, also called BUBR1, are mitotic spindle assembly checkpoint regulators. Think of them as quality-control inspectors: they monitor the machinery that separates chromosomes during cell division and signal the cell to halt if something is wrong. In a healthy cell, this is protective. In a cancer cell, these kinases are hijacked to push division forward. ChIP-qPCR confirmed that H3K18la is genuinely enriched at the TTK and BUB1B promoters, and that enrichment drops when cells are treated with multiple glycolysis inhibitors — dichloroacetate, Oxamate, and two-deoxyglucose. Protein and transcript levels of both genes fell accordingly. When the team knocked down TTK or BUB1B directly with siRNA, cell viability dropped and migration was impaired in both PDAC cell lines tested. Conversely, TTK overexpression partially blunted the anti-tumor effects of glycolysis inhibitors, consistent with TTK functioning downstream of H3K18la to promote malignancy. These are drivers, not bystanders. Then comes the part that makes this loop genuinely remarkable. TTK and BUB1B don't just drive cell division — they feed back to accelerate the very glycolysis that activated them. Li and colleagues showed that TTK and BUB1B elevate expression of P300, the lactylation writer. More P300 means more H3K18la, which means more TTK and BUB1B, which means more P300. That's one arm of the loop. The other arm is more direct. Using co-immunoprecipitation assays — a technique that captures proteins physically bound to each other — the team demonstrated that TTK binds directly to lactate dehydrogenase A, known as LDHA, the enzyme that converts pyruvate to lactate at the end of glycolysis. TTK knockdown reduced phosphorylation of LDHA at tyrosine residue Y239, which lowered LDHA enzymatic activity, which reduced lactate levels in culture supernatant, which decreased pan-lysine lactylation and H3K18la. Pull on one thread and the whole circuit responds. The full loop runs like this: elevated glycolysis raises lactate, lactate drives H3K18la, H3K18la activates TTK and BUB1B, those kinases elevate P300 and directly stimulate LDHA through Y239 phosphorylation, and LDHA boosts glycolysis and lactate further. Each cycle reinforces the next. This is not a linear pathway you can block at one point and stop. It's a ring with no obvious entry point — which is part of why PDAC is so aggressive. Li and colleagues then took this loop out of cell culture and into mice. MIA PaCa-2 cells were implanted subcutaneously, and once tumors reached roughly seventy cubic millimeters, animals received daily Oxamate at seven hundred fifty milligrams per kilogram by injection for thirty days. Tumor volume was significantly lower throughout the treatment period, and tumor weight showed a dramatic decline compared to vehicle. A parallel experiment used cells with stable LDHA knockdown — the shLDHA group — which also produced smaller, lighter tumors over twenty-six days. In both cases, the anti-tumor effect tracked directly with metabolic and epigenetic changes: intratumoral lactate fell, pan-lysine lactylation dropped, and H3K18la levels declined on both western blot and immunohistochemistry. The proliferation marker Ki-67 was reduced, and liver histology showed fewer metastatic foci. The feedback loop observed in dishes held up in living tumors. What this work ultimately delivers is a named, dual-axis therapeutic target: the H3K18la–TTK/BUB1B axis. To interrupt PDAC growth, you don't have to choose between attacking metabolism or attacking the epigenome — this circuit connects them. LDHA is the metabolic entry point, P300 is the epigenetic one, and TTK and BUB1B sit at the junction where both converge. The authors are careful to note that further investigation is needed to fully define the scope of this epigenetic reprogramming. But the in vivo data move this beyond a mechanistic curiosity. A self-reinforcing loop that links a cancer cell's fuel consumption to its gene regulation, validated in mouse tumors — that's the shape of a druggable target. For a disease where almost nothing has worked, naming the loop is the first step toward breaking it. This lecture was created by ennepō. Go to https://ennepo.ai to Discover, Create and Follow the latest research in your field. Read when you can. Listen when you want to.