Investigating the trigger for fibroblast senescence in colorectal cancer: a novel role for extracellular vesicles

Bowel cancer is the second most common cause of cancer death in the UK and most deaths occur when cancer spreads to other parts of the body. Although there have been marked improvements in life expectancy following bowel cancer in the past 20 years, there is still scope for better detection, risk-stratification and treatment. One way to achieve this is to precisely understand differences between normal and cancerous bowel at a molecular level. If certain molecules are more or less abundant in cancer, then these differences can be utilised to develop diagnostic tests, predict the likelihood of cancer coming back after treatment and design more effective drugs, which destroy cancer cells with less collateral damage.

An important concept to understand is that cancer is more than cancer cells. A tumour consists of cancer cells, supporting cells and structural molecules. These components interact to allow the tumour to expand and spread. One type of supporting cell, known as a fibroblast, is well known for regulating the architecture of a tumour. Fibroblasts are strongly linked with tumour development, growth and spread. As fibroblasts age, it is though that they produce different molecules, which accelerate tumour growth. There are several proposed mechanisms for fibroblast aging but the trigger is unclear.

Our laboratory is interested in how cancer cells affect supporting cells and what influence this has on the tumour as a whole. Our initial experiments show that certain bowel cancer cells transmit messages to fibroblasts in tiny packages called vesicles. The effect of this is to bring about aging in the fibroblasts, which we know leads to cancer progression. One of the transmitted messages has been identified as a molecule called miR-143.

The aim of this project is to find out how bowel cancer cells cause fibroblasts to age. We think that: (i) cancer cells transmit miR-143 to fibroblasts in vesicles; (ii) miR-143 causes fibroblasts to age; (iii) miR-143-related aging causes fibroblasts to release particular molecules which make the tumour grow and spread. To investigate this, miR-143 levels will be artificially altered in fibroblasts and aging will be assessed. Molecules produced by fibroblasts with artificially high miR-143 will be identified. These molecules will be fed back to cancer cells to see if they cause increased growth and greater spread.

This work has the potential to identify the trigger for fibroblast aging in bowel cancer. With this information, it may be possible to identify patients who are at higher risk of death so that they can have more intensive treatment. Equally, it makes it possible to design a drug which can prevent fibroblast aging, potentially reducing tumour growth and spread.

Our research group has a good balance of clinicians and scientists and the unique blend of skills to make this project feasible. As a surgical trainee, I bring a perspective which ensures that my research is focused on areas of clinical need so that any findings can be translated efficiently into beneficial outcomes for patients.

Background: Fibroblast senescence is a poor prognostic factor in several solid cancers but the trigger for this fundamental process is unknown. E-cadherin-expressing colorectal cancer (CRC) cells produce extracellular vesicles (EVs), attenuating ERK activity in fibroblasts. This corresponds with a proliferation defect, G1 arrest and a higher proportion of senescent cells. EVs from these CRC cells are rich in miR-143, which represses KRAS.

Aims and Objectives: This project aims to identify a pathway which leads to fibroblast senescence in CRC through the following objectives: 1. Determine whether miR-143 regulates fibroblast senescence via Ras-inhibition. 2. Ascertain whether extracellular vesicles (EVs) from CRC cells deliver a functionally relevant amount of miR-143 to fibroblasts. 3. Identify changes in the secretory profile of miR-143-modulated fibroblasts and how this alters behaviour of CRC cells. 4. Demonstrate findings in vivo using an orthotopic colorectal cancer mouse model.

Methodology: MiR-143 will be overexpressed and inhibited in fibroblast lines. Proliferation, cell cycle profile and senescence will be assessed. Similarly, E-cadherin-expressing CRC cells will have miR-143 levels manipulated, to alter the miR-143 content of their EVs. EVs will be delivered to fibroblasts and senescence assessed. MiR-143-modulated fibroblasts will have their conditioned media profiled for cytokines. CRC cells will be exposed to conditioned media and proliferation, migration, invasion and chemoresistance measured. Finally, an inducible miR-143 fibroblast line will be co-injected with CRC cells intra-caecally in nude mice. Animals will be followed up by in vivo imaging to assess effects of miR-143, including response to chemotherapy.

Scientific and Medical Opportunities: Understanding the molecular drivers of fibroblast senescence in colorectal cancer is critical to identifying novel prognostic biomarkers and actionable therapeutic targets.

The focus of my research to date is the dynamic ecosystem which represents the tumour microenvironment (TME) and the interplay between cancer and stromal compartments which results in cancer progression. One important observation in the TME of solid cancers is fibroblast senescence, which is a poor prognostic marker, partly due to the ability of the senescent phenotype to increase tumour growth and metastasis. There are several proposed mechanisms for fibroblast senescence but it is unclear which is most important in cancer and moreover, the trigger for these mechanisms has not been elucidated. Furthermore, a detailed understanding of the pathways involved in this process is likely to uncover a host of dysregulated molecules which could potentially develop our armoury of prognostic biomarkers and therapeutic targets.

This proposal represents an innovative and feasible research strategy to investigate the trigger for fibroblast senescence in colorectal cancer (CRC), dissecting out a critical aspect of tumour biology with significant translational promise. The translational aspects of this work are very exciting. Currently we are desperate for biomarkers which can predict recurrence in intermediate-risk (stage II) patients, as these individuals are not routinely offered adjuvant treatment, and as many as 25% develop recurrent disease after surgery. Clearly there is a balance to be met and the blanket prescription of cytotoxic chemotherapy is inappropriate. However, if there were a molecule associated with the senescence pathway, such as miR-143, which could predict higher-risk patients who needed more intense treatment, this could have a significant public health impact in the medium term. Used in combination with other prognostic miRNAs such as miR-21, this could form the basis of a prognostic algorithm for intermediate-risk colorectal cancer, exemplifying the shift towards personalised cancer medicine.

Similarly, the identification of dysregulated molecules in this pathway could facilitate pharmaceutical companies to develop targeted stromal cancer therapy, which is less likely to cause cytotoxic side effects. This would clearly impact on oncological practice in the medium to long term. If our hypothesis proves to be correct, then miR-143 within extracellular vesicles (EVs) is the trigger for fibroblast senescence. This would be fortuitous because microRNA inhibitors are already commercially available and pharmaceutical companies would be able to focus efforts on drug delivery. Furthermore, advances in our understanding of tumour biology have been mirrored by technological advances, which allow for use of synthetic vesicles and even autologous vesicles. This momentum has the potential to create a new generation of anti-cancer drug, which are better targeted, with a more acceptable side effect profile.

On a scientific level, this project has and will continue to generate pioneering models which help to answer fundamental scientific questions. For example, I stably transfected and cloned a CRC cell line which produces fluorescent EVs (HCT116-CD63-GFP). This model was used to demonstrate in vivo transfer of vesicles, which is central to this proposal. Another innovative approach entails generating a stably inducible fibroblast line, in which miR-143 overexpression can be stimulated exogenously. I am happy to share all these models with my collaborators and other researchers in the field, in order to widen their impact on the scientific community.