Skip to main content

Collagen in Cancer: from the regulatory fibril forming function of collagen V in development to its implication in tumor progression

Periodic Reporting for period 1 - CoCa (Collagen in Cancer: from the regulatory fibril forming function of collagen V in development to its implication in tumor progression)

Reporting period: 2016-06-01 to 2018-05-31

Cancer has been classically considered as a heterogeneous set of (epi)-genetic diseases where a clone with a differentiation blockage proliferates in an uncontrolled manner to give rise to a tumor. However, an increasing number of reports have highlighted the essential role of the extracellular matrix (ECM) in tumor cell functions and tumor-related processes. Disruption of ECM network, composition or organization, contribute to cancer development and metastasis and overall disease progression. A large number of matrix proteins are dysregulated during cancer process making data analysis and interpretation very complex. Most of these proteins are inter-connected and perturbation in expression level of one of the constituents can impact the network architecture. This is the challenge of the project, where a candidate approach was preferred. We chose to study type V collagen (COLV) because it has a central role as a regulator of fibril formation and many different domains essential for cell signaling. This protein, which is quantitatively poorly represented in the ECM of healthy tissues, has an aberrant deposition in several types of tumors such as ductal infiltration carcinomas, non-small cell lung cancer, colorectal cancer and pancreatic ductal adenocarcinoma. We focused on lung carcinomas, which is one of the most frequent cancer in the world and it is also the first killer in adult patients. I studied it not from the tumoral cell but from the microenviroment perspective, which is in fact the matrix that surround those malignant cells.
My research program builds on recent conceptual advances in how the physical and structural properties of the cell microenvironment contribute to tumor progression. While there is evidence that a slight increase in the hardness of the surrounding ECM perturbs tissue function, the influence of ECM topography on cancer progression remains largely unknown. Integrating molecular biology, transcription analysis, biochemistry, high-resolution imaging and mouse and cell models, the project investigated the role of COLV in tumor progression and the underlying mechanisms. This project is original and innovative for the following reasons: (1) it will comprehensively map COLV expression patterns in cancer. Central to this part of the project are two important findings: the ECM is a dynamic structure that influences tumour progression, and multiple cell types within the tumor can contribute to ECM production. It might provide the background for potential use of COLV as a biomarker and clues for understanding its function in cancer progression, helpful to clinicians. (2) Tumoral ECM topography related to the collagen V deposition. This burgeoning field of research focuses on how physical properties of ECM impact on cell behaviour during development. Investigating how disrupted ECM topography may contribute to cancer is even newer. We aimed to connect COLV overdeposition with a specific tumoral ECM topography to crucial aspects of the tumor phenotype and gene expression. (3) Upstream regulation of collagen V by hypoxia. Recent studies have established a direct link between hypoxia and the composition and the organization of the ECM. Blocking tumor hypoxia might provide a strategy to reduce COLV overdeposition and opens a possible avenue for anti-cancer therapy. I see this part of the project as a fundamental building block that will pave the way for future research to develop new therapeutic options offered to patients.
The main objective of this proposal was to decipher the mechanisms by which deregulation of COLV expression and subsequent extracellular deposition impacts cancer development. My aim was thus to provide a more complete picture of its role in tumor ECM topography, and how these changes impacts on cancer progression with a view to exploit this knowledge for cancer diagnosis and possibly in therapy. I had 3 more specific questions to address:
1. Establishment of a compilation data of COLV expression and localization in different tumor-type categories and its the cellular sources
2. Lay down the relationship beween Collagen V overdeposition and tumor ECM topography
3. What are the upstream mechanisms responsible for dysregulation of COLV expression in cancer? Does hypoxia regulate COLV overexpression in tumors in vivo?
I started this project doing classical immunohistochemistry (IHC) for COLV and COLI in a collection of different mouse epithelial tumors. In parallel, a total of 154 lung carcinoma tumors from human patients were examined by IHC for COLV and I, but also CA9 and HIF-1 as hypoxic makers, since we hypnotized that hypoxia could be a possible cause of the COLV overdeposition observed in tumors. This gave the first insights into the cartography of COLV depositions in tumors, the cell that produce it and its relationship with hypoxia.
For the second objective I developed human lung tumor model into immunodefficient mice inoculating A549 tumoral cells. First classical IHC for COLV, I, CA9 and HIF-1 was performed. The studied tumoral sections by SHG microscopy allowed us to see the topography of the matrix fibres, quantify and measured its dimensions. We also performed Raman spectrometry studies which gave the main biochemical composition of the tissue, being able to compare the matrix between the tumor and surrounding healthy tissue.
For the third objective in silico analysis of COLV proximal promoter revelled several putative regions susceptible of hypoxic regulation. This bioinformatic study found a very good candidate motive located 201 before the beginning of the COL5A1 gene. After, I established in vitro culture of different cell types that are present in the tumor (HUVEC, human Fibroblasts and A549 tumoral cells) to compare matrix production under hypoxic and normoxic conditions. We used Q-PCR and Immunofluorescence studies to quantify the production of matrix at transcriptional and protein level. In parallel, I performed a similar mouse experiment as in the objective 2, but in this case submitting the tumoral development to hypoxic conditions. Tumoral growth was followed by bioluminescence and samples were taken at different time points to establish the influence of hypoxia in the matrix structure and disease progression. At last, AFM microscopy was performed in hypoxic and normoxic samples to see the impact of hypoxia in the stiffness of the tumoral and healthy tissue.
This was a very ambitious project, we developed different mice experiments that take many months to organise, develop and analyse and I used many different techniques that needed to be set up each of the experiments. After the 2 years of fellowship were finished, I received an additional contract of an additional month (with Florence Ruggiero, leader of the host lab) to finish some of the experiments. Altogether, I generated a huge amount of data that is currently under final analysis and preparation of one or two manuscript. The most interesting results may not be specified at this time due to consideration of IP issues and further dissemination. There are many promising preliminary results with very interesting potential applications in numerous sectors (public, clinical, pharmaceutical) therefore the finalization of the analysis and publication of the results is the most exciting perspective.