Back

Inflammation-driven reprogramming of goblet cells underlies the onset of serrated adenomas in colon cancer

Verhagen, M.;Guerra, E.;Joosten, R.;Sacchetti, A.;Sun, W.;Lila, K.;Nigg, A.;Ahmed, E.;Saadeh, H.;Bosch, T.;Hendrickx, W.;Roelands, J.;Vries, A.;Doukas, M.;Fodde, R.

2026-06-23 Cancer Biology
10.64898/2026.06.22.733748 bioRxiv
Show abstract

In the context of inflammation, fully committed and post-mitotic cell lineages can initiate intestinal tumorigenesis in the mouse through dedifferentiation and acquisition of revival stem cell (RSC) features1,2. Likewise, by means of machine-learning analysis of whole-genome mutation spectra, the secretory goblet cell was predicted as the most frequent cell-of-origin (COO) of colon cancer in inflammatory bowel disease (IBD) patients2. Of note, even among sporadic (non-IBD) colon cancer patients, goblet cells were predicted as the most common differentiated COO of cancer in the colonic epithelium2, suggestive of the main role played by diet-induced inflammation in the onset of a large proportion of malignancies of the large bowel. However, how goblet cells respond to inflammation and reprogram their identity to become potential tumor-initiating cells remains unclear. Here, by taking advantage of publicly available single-cell RNAseq data from colonic tissues of ulcerative colitis (UC) patients3, we have characterized the inflammation-driven reprogramming of goblet cells. By means of an RNA velocity-based computational approach, we show that the mucus-producing goblet cells acquire an aberrant proliferative state earmarked by MUC5AC+ expression. Trajectory analysis of serrated adenoma cells4 reveal aberrant goblet cells as an intermediate state in the transition to revival stem cells, notably more common in BRAF-mutant cases. In support of these findings, COO predictions using whole-genome mutation spectra from a cohort of sporadic colon cancers5 connect tumors with a predicted goblet origin to BRAF mutations and reveal transcriptional remnants of the aberrant goblet cell state.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.