New understanding of stem cells in human colon lesions

Zoe Leech Posted in Publications 13 March 2015

cancer research uk 482Dr Ann-Marie Baker with the Wright and Graham labs in our Centre for Tumour Biology has a new study published in Nature Scientific Reports.

They have found that the number and location of stem-like cells in the human colon - those that can potentially make copies of themselves as the gut renews itself – differs between pre-cancerous lesion types.

Ultimately, if we can understand "stemness" and its role in cancer better, it should lead to improvements in diagnosis and treatment.

Understanding colon cancer

In the colon, carcinogenesis tends to follow a staged process, with abnormal growth sometimes leading to malignant transformation. People are offered screening when at risk of cancer, if they have family history or show signs of lesions being present.

When a biopsy (tissue sample) is taken from a patient to see how abnormal it is and whether they will require further treatment, it is viewed under the microscope by pathologists. There are different types of lesion that can occur in the gut, both those that do not lead to cancer and those that can, including classical adenomas and serrated adenomas. However, when different types of tissue lesion look very similar, it can be hard to establish a diagnosis.

While it is known that the stem-like cells that tend to be found at the base of healthy colon crypts, this organisation tends to break down as cancer develops, and tracking stem cells may help in finding out how aggressive it is, or will become in the future.

How reliable are models?

Stem-like cell characteristics have been studied in mice previously, but working out how much they have in common with human cells is important; the more characteristics they share, the more likely it is mouse model results will translate to humans – for example when testing new drug treatments.

The Graham lab has previously shown that bowel cancer evolution seems to operate similarly in both mice and people. In this work, Dr Baker has confirmed that a marker for stem-like cells in mice, a gene called LGR5, is also present in the same locations in human tissue.


Left: “conventional” adenoma sample. Right: serrated adenoma sample. In situ hybridisation staining with LGR5 in red-purple. While the tissue architecture is very similar, it is clear that LGR5-positive cells are restricted to crypt bases in serrated lesions. (Images by Ann-Marie Baker)

While other groups have looked at late-stage cancer in mouse models, this is a more comprehensive study of many types of early tumours. This discovery of a marker that shows cells behaving differently in similar lesions can help pathologists make accurate diagnoses.

One issue with finding a marker as useful as LGR5 has been the lack of antibodies available. In tissue studies, antibodies are used to highlight proteins of interest, but they are not yet available for every human protein.

As the end-products of genes, proteins tell us how active the gene is in a cell. If we cannot detect the proteins directly, we can also look at mRNA – the intermediate message created from the original DNA code (a process called transcription), before it is translated into proteins.

In this study, a technique called in situ hybridisation was used to detect LGR5 mRNA in human gut tissue samples. They found that the stem cell niche is maintained in all serrated lesions tested but in other types, the structure is lost.

In the clinic

This distinction could have implications for treatment, once we understand more about stem-cell behaviour in colon and have treatments that could target or regulate them and cancer progression.

There are also questions of over-diagnosis, as with most screening programmes; the stem cell organisation in a tumour could help to inform which patients are recommended for which treatments, once we can link lesion type to the likely disease progression paths. In other words, measuring the stem cell organisation may help answer the question, "will cancer develop from this lesion, and will it be aggressive or slow-growing?"

The group is planning to use similar techniques to relate stem cell architecture to cancer progression risk.

This work was funded by Cancer Research UK. With thanks to Dr Manuel Rodriguez-Justo and the patients at University College London Hospital for providing archived tissue samples


Characterization of LGR5 stem cells in colorectal adenomas and carcinomas
AM Baker,TA Graham,G Elia, NA Wright, M Rodriguez-Justo, Sci Rep (2015) 5:8654, PMID: 25728748

Leave a comment

You are commenting as guest. Optional login below.

This site uses cookies in order to function properly. By continuing to browse, you agree that we can save them on your device. Privacy Policy.