A multi-cellular 3D model of human breast tissue to replace rodent xenograft models in breast cancer research.

The cells that make up the tissues of your body are surrounded by a matrix of proteins and sugars. Cells interact with this matrix environment helping organs fulfil their function reacting to the elasticity and stiffness of the matrix and secreting messenger molecules into the matrix leaving specific information for other cells. This complex communication is crucial during development of an embryo driving the segmentation of the body into organs and to divide organs into functional regions. The matrix also plays an important role in disease progression where cells digest proteins or sugars within the matrix to release cryptic fragments that can instruct cells to multiply or to migrate, both features that are exaggerated in the progression to cancer.

Scientists have specialised techniques that can interpret messages within cells, these are often carried out in human cells grown on plastic (2D) which would not include matrix interactions. To understand 3D matrix-driven signalling they often rely on animal-derived artificial matrices that can be hard to work with due to high variability or use animal models including the growth of human cells transplanted as a xenograft into a mouse. Although these 3D assays are an improvement from 2D culture neither fully reproduce the complex human tissue matrix. There is a need to improve the 3D growth of cells within the laboratory and reduce the need for animal models that do not represent human tissue. To this end we have developed a fully synthetic, highly reproducible gel that can be decorated with proteins and sugars to mimic the matrix of human tissues. Cancer cells, along with other cells types found in native tissues, can then be encapsulated in the gels and easily grown in the lab.

The development of our bespoke human matrix will provide scientists with cheap, functional and robust test environments where cancer cell behaviour can be studied in a human body mimick. This allows researchers to test theories of how cancers develop, discover new targets for intervention and additionally will provide more realistic test environments for screening therapeutics. To test our bespoke gel environments we are using breast cancer as a model system, which can we investigated in its normal, pre-invasive and invasive cancer forms. We will characterise the differences in matrix between dense and non-dense breast tissue (normal and cancer) as evidence shows that dense breast tissue is a major risk factor for breast cancer. We will then use this information to decorate gels with specific proteins and sugars to mimic these distinct matrix environments, by encapsulating cancer cells and pre-cancerous cells in the gels (along with other cell types typical of the tissue) we hope to better understand why this is and what role is played by specific proteins and sugars in the matrix. This project brings together an interdisciplinary team of cancer biologists, materials scientists and clinicians to develop a new solution that we hope will have impact for the study multiple cancer types once proven as a robust model for breast cancer we anticipate significantly reducing the numbers of animals used in xenograft studies across the world.

We aim to reduce the number of rodent models investigating the biology of the breast by developing a superior in vitro matrix incorporating bespoke human breast extracellular matrix (ECM) components, providing a robust culture model as the go-to platform for exploring the biology of the breast and breast cancer. Drs Merry and Meade have developed a simple peptide hydrogel system originally optimised for the culture of stem cells. In collaboration with Dr Farnie and Prof Howell, who provide expertise in primary breast cell culture, ductal carcinoma in situ (DCIS) and breast ECM/density, we have pilot data demonstrating that the hydrogel is suitable for the growth of typical in vitro 3D normal and DCIS breast structures. As with the vast majority of cancers, the interaction of DCIS cells with their local microenvironment, the surrounding stroma and ECM, is fundamental in defining their proliferation, transformation to malignancy and invasion. Our hydrogel is well positioned to replicate aspects of the complex mixture of proteins and glycans that embeds and supports cells, and can also be manipulated to mimic ECM stiffness (breast density) which is a key predictor of DCIS recurrence and primary invasive breast cancer development. We will use a combination of proteomics and glycomics to identify the key ECM components defining dense and non-dense breast tissue in normal, DCIS and invasive breast cancer conditions. The proteins and glycans will then be combined to generate a panel of 6 bespoke hydrogel environments modelling the range of tissue types. We will validate the model using multicellular 3D culture and assay for ECM remodelling by encapsulated cells, directly assessing the ability of our in vitro model system to replicate human breast tissue. Finally, we will incorporate immune cells into the gels, thereby addressing a key feature of animal models that currently separates them from in vitro systems.

We envisage a time in the not too distant future where the research community will advance humane, modern methodologies eliminating the need for animal experimentation and improve the quality of science. To this end we aim to develop an in vitro multicellular human breast model to REDUCE and eventually REPLACE xenograft models for investigating breast cancer development, with an ultimate goal of REDUCING and REPLACING the use of rodent xenograft models in all organotypic and cancer tissues. Despite rodent xenograft models having major limitations they remain the 'go to' validation model in breast cancer research. This reflected by an estimated publication of 300 papers/year involving mammary fat pad and/or subcutaneous xenografts in mice resulting in 12,000+ animals per year (~40 animals/paper). The need for a superior in vitro alternative is more pressing considering the dominance of patient-derived xenograft (PDX) models as a tool for personalized therapy, a EurOPDX consortium reporting a panel of ~1,500 PDX subcutaneous and orthotopic models illustrates this aspiration (2). Such 'Xeno-patient' cohorts are considered the most clinically relevant animal models in cancer drug discovery. However the successful establishment of a PDX can be a little as 24% (1) and maintenance, involving successive passages, requires vast numbers of mice. We anticipate that our technology can fulfill the requirement for a robust, cost-effective and high-throughput in vitro drug screening system and our recent funding for a CellJet live cell printer allowing nanoliter gel dispensing would facilitate such investigations, to compliment or fully replace PDX models. In addition, 600 mice are sacrificed each year (~10 papers/year, 60 animals/paper) addressing the role of extracellular matrix (ECM) in breast cancer progression, not including the 960 mice/year generating genetically engineered mouse models, which are currently crucial for investigating involution. The growing interest in ECM either for modelling breast involution, a key target for breast cancer prevention, and its role in cancer progression (e.g. search term 'breast cancer matrix' returns 3x more publications between 2000-2013) will inevitably increase the use of rodent models. Our hydrogel model would have the versatility to be tailored to mimic these specific breast research questions. Once developed we anticipate our breast hydrogel models to initially serve as a pre-screening tool prior to in vivo experimentation. Utilised in this manner, we envisage a 50% reduction in animals used. As confidence in the scientific community grows, our technology will fully replace the need for mouse xenograft models as a superior alternative. To deliver this impact we will disseminate and promote our findings at scientific meetings, through the media and manuscripts including a methods paper detailing the creation and application of tissue-specific bespoke hydrogels. Moreover, our patented hydrogel technology is currently under development as a commercial product through the University of Manchester (UMIP), which will enable the broad dissemination of data and information from this project. The production of a reliable, cost-effective and commercially available product accepted by the scientific community and replacing xenografts in breast cancer research, would reduce the use of rodents by 12,000+ animals per year. However the versatility of our hydrogel allows the molecular composition and mechanical properties to be manipulated with ease supporting the extended application of our technology to mimic all organotypic or cancer tissue types. The implications for animal testing in this respect are extensive, in the uk alone there were 3.75 million non-toxic scientific procedures in 2013 (3).
1 Tentler et al. Nat Rev Clin Oncol. 2012;9(6):338-50 2 Vinolo et al. Mol Cancer Ther. 2013;12:A8 3 Home Office. Annual
Statistics of Scientific Procedures on Living Animals Great Britain 2013