Examples of our commitment to the 3Rs

 Zebrafish help researchers understand how cancer spreads in the body

zebrafish Ray

Zebrafish are helping researchers to understand how cancer spreads.

Zebrafish have enabled researchers to use fewer mice during their investigations of how cancer cells spread around the body – a process called metastasis.

A team led by Dr Stéphanie Kermorgant at QMUL’s Barts Cancer Institute, carried out part of their animal research work on zebrafish embryos in order to implement the principle of 3Rs (refine, reduce, replace) on their research in mice. Zebrafish provide a similar tumour microenvironment to humans, meaning fewer tests need to be carried out in mice and any future experiments in mice will have been optimised to have minimal toxicity. They are aiming to reduce the number of mice used by at least 90 per cent and ultimately use zebrafish to completely replace the use of mice.

The researchers examined the changes that occur in cancer cells as they break away from tumours in cell cultures, zebrafish and mice. They found a previously unknown survival mechanism in cancer cells and found that molecules known as ‘integrins’ could be key.

Integrins are already major targets for cancer treatment with drugs either being tested or in use in the clinic. Most integrin inhibitor drugs target their adhesive function and block them on the surface of the cancer cell. The researchers say that the limited success of these drugs could be partly explained by the newly discovered role of integrins within the cancer cell.

The researchers hope that these insights could lead to the design of better therapies against metastasis and more effective treatment combinations that could prevent and slow both tumour growth and spread

Integrins are proteins on the cell surface that attach to and interact with the cell’s surroundings. ‘Outside-in’ and ‘inside-out’ signalling by integrins is known to help the cancer cells attach themselves to their surroundings. But the study suggests that when the cancer cells are floating, as they do during metastasis, the integrins switch from their adhesion role to take on an entirely new form of communication which has never been seen before – ‘inside-in’ signalling, in which integrins signal within the cell.

The researchers discovered that an integrin called beta-1 (β1) pairs up with another protein called c-Met and they move inside the cell together. The two proteins then travel to an unexpected location within the cell which is normally used to degrade and recycle cell material. Instead the location is used for a new role of cell communication and the two proteins send a message to the rest of the cell to resist against death while floating during metastasis.

Using both breast and lung cells, the team found that metastases were less likely to form when β1 and c-Met were blocked from entering the cell together or were prevented from moving to the special location within the cell.

A new therapeutic strategy could be to prevent the integrin from going inside the cell in the first place. The researchers hope that these insights could lead to the design of better therapies against metastasis and more effective treatment combinations that could prevent and slow both tumour growth and spread.

This work was done in collaboration with Dr Caroline Brennan from the School of Biological and Chemical Sciences at QMUL. The study was published in Nature Communications.

 Facility to store frozen mice sperm and embryos is an important contribution to the 3Rs

Unfertilised and fertilised mouse embryos prior to being frozen

A facility to store frozen mice sperm and embryos reduces the number of mice kept for research at QMUL and it also provides a back-up repository for research scientists.

Instead of having to keep alive colonies of mice that have been bred specifically to model human conditions and diseases, researchers in the School of Medicine and Dentistry can choose to freeze and store embryos and sperm. When they need live animals to substantiate their research or to take it forward, the sperm or embryos can be thawed and used to breed more of the same mice with the same genetic modifications (transgenic mice). This initiative is an important part of the university’s commitment to the “3Rs” in animal research: reduction, replacement and refinement.

The freezing or “cryopreservation” facility was launched in October 2016 and is being led by Dr Carles Gaston-Massuet, group leader and senior lecturer in genetics and endocrinology at the William Harvey Research Institute. For the first year the facility will be open to QMUL researchers only and already Dr Gaston-Massuet and his team have a long list of people who are interested in using it.

Dr Gaston-Massuet says: “If researchers have to keep colonies alive, this costs a lot in terms of money, space in the animal unit and numbers of mice that have to be bred but then not used in research. This facility allows embryos and sperm to be frozen and stored. It is also a way of backing up your research, just as you would back up data on your computer. Animals may get old or a colony could be lost due to disease, which may result in losing important transgenic lines and years of research. The cryopreservation unit will allow users to restore the colony using the stored frozen embryos and sperm.”

This initiative is an important part of the university’s commitment to the “3Rs” in animal research: reduction, replacement and refinement

Angelica Gualtieri, a research assistant working with Dr Gaston-Massuet on the project, explains: “Now, researchers will have the option to use the in-house facility to store their complex genetic models instead of outsourcing to other centres, which don’t offer competitive prices. The facility will also have the ability to recover mouse transgenic lines using sophisticated in vitro fertilisation (IVF) techniques to fertilise the eggs of a live mouse with thawed sperm. Once embryos have been thawed or IVF has been performed, embryos can be transferred by non-surgical means into surrogate mothers to recover the transgenic lines in an efficient way.”

Dr Gaston-Massuet and his team are working with the biological facilities manager to develop a database of all the different transgenic mouse lines currently in use. This will enable researchers to ensure there is no duplication of existing lines, to increase collaboration and to reduce the amount of space required in animal units.

The total cost of the facility so far is around £30,000, which has been funded to date by a £17,000 grant from the School of Medicine and Dentistry and by Dr Gaston-Massuet’s own grants. The aim is for the facility to be self-funding and so researchers will be charged to store their mouse embryos and sperm. This will work out much cheaper for them than having to buy transgenic mice, maintain breeding colonies, or outsource to external facilities at much higher prices.

 New laboratory methods could replace mice to develop drugs targeting cancer stem cells in oral cancer

Cancer stem cells grown in the lab

Dr Adrian Biddle is growing cancer stem cells in the lab to develop anti-cancer drugs.?

Researchers at QMUL are pioneering ways of growing a particular type of cancer cell in the laboratory (in vitro) rather than in mice (in vivo) to develop anti-cancer drugs.

Between 2012 and 2015, the National Centre for the Replacement, Refinement and Reduction of Animals in Research (NC3Rs) funded a Fellowship for Dr Adrian Biddle and ongoing support to Professor Ian Mackenzie for their study of the biology of cancer stem cells. Their work has the potential to dramatically reduce the numbers of mice that are used to develop treatments for cancer.

Cancer stem cells (CSCs) are a sub-population of cancer cells that are responsible for the maintenance, recurrence and spread of cancer. They have the characteristics of normal human stem cells in their ability to self-renew and give rise to other cell types, and they can also be more resistant to chemotherapy and radiotherapy than other cancer cells.

When developing drugs to target CSCs, researchers often transplant human cancer cells into mice in order to monitor tumour development and response to anti-cancer drugs. Not only is this time-consuming and costly but it involves using large numbers of mice that can experience discomfort and distress due to tumour growth.

“Our work at QMUL has focused on the role of cancer stem cells in oral cancers,” says Dr Biddle. “Many millions of mice are used each year in cancer research, and so there is a great need for in vitro tests that can replace these animals. Our study, demonstrating a new cell culture model for testing the response of cancer stem cells to therapies, has the potential to increase the accuracy of pre-clinical cancer drug testing and will encourage a switch away from mouse models.”

Professor Mackenzie, Dr Biddle and QMUL colleagues discovered that CSCs could switch between two different identities: one that drives tumour growth and another that drives the spread of cancer to other parts of the body (a process known as metastasis, which is the main cause of deaths from cancer). Then they demonstrated that the type of CSCs that drive metastasis are highly flexible and very drug resistant. They succeeded in growing these cells in vitro, adjusting the growth conditions in the laboratory dish so that they retained their specific stem-like properties and adaptability or “plasticity” – something that is often lost during long periods in culture.

Researchers at QMUL are pioneering ways of growing a particular type of cancer cell in the laboratory ( in vitro) rather than in mice ( in vivo) to develop anti-cancer drugs

“Using this lab model we have been able to show that the plasticity of cancer stem cells is crucial for their resistance to drugs, and we have identified biological markers that identify this group of highly plastic cancer stem cells,” says Dr Biddle. “We have also identified corresponding cancer stem cell sub-populations with comparable drug resistance in specimens from patients, confirming that it is an important aspect of cancer biology with relevance to human tumours. This replacement for the mouse model has great potential for studying the mechanisms of resistance to treatments such as chemotherapy, radiotherapy and hormone therapy, and eventually for testing new treatments in a fast and accurate way.”

Dr Biddle is now a lecturer at QMUL and is funded by a grant from the Dr Hadwen Trust, the UK’s leading non-animal medical research charity. The NC3Rs is maintaining its support for this work with a grant to Professor Mackenzie, who is Professor of Stem Cell Science at QMUL. The researchers are continuing to develop lab models that will be better than mice for rapidly assessing response to drugs aimed at killing CSCs and controlling metastasis.

 Non-invasive imaging to reduce the number of animals used in scientific research

Non-invasive imaging

Imaging allows researchers to study tumours in mice in a non-invasive way.?

After doing as much work as possible using cells or tissues, some studies then require testing a hypothesis or treatment in living animals. Therefore, some QMUL researchers are working to reduce and refine use of animals using non-invasive imaging. These imaging techniques are the same as those used in humans, for example ultrasound, MRI or PET imaging, but the researchers use specialist equipment for small animals such as mice and rats.  

The most effective way of reducing animal numbers through non-invasive imaging is to carry out longitudinal studies, in which the same animals are studied over a period of time. Cancers develop deep inside the body (sometimes over long periods of time) and it is difficult to assess the size of the tumours by examination, through feeling, or “palpation”. So if researchers want to see how a mouse model of a human cancer, such as pancreatic cancer, is responding to a new, experimental treatment they have two options.

If imaging is not used, groups of animals that are large enough to represent the normal variability seen between different animals would be killed at different timepoints throughout the study. The relevant tissues and organs would be removed for analysis to get a measure of whether or not the treatment is working relative to the control group.

If the size of the group is 10, with a corresponding control group of 10 animals, and the study dictates that you need to assess responses weekly for five weeks, it would be necessary to start the study with 100 animals. However, if tumour response measurements can be made using non-invasive imaging that does not require the animals to be killed, the numbers used can be reduced to 20. The researcher would start with a group of 10 animals plus 10 for a control group and they would be non-invasively imaged each week for five weeks. This is similar to the way we discover if people are responding to treatment in human studies.

QMUL researchers are working to reduce and refine use of animals using non-invasive imaging. These imaging techniques are the same as those used in humans, for example ultrasound, MRI or PET imaging

Many studies done this way use tumour volume as a measure of response. However, we know that there are also biochemical changes in tumours that occur long before any change in tumour volume is seen. The challenge is to measure these biochemical changes – such as tumour metabolism – non-invasively to get functional information about the behaviour of the cancer in response to therapy, without physically removing the tissues for analysis.

In addition to reducing animal numbers using in vivo imaging, there are other important ways imaging can reduce stress and suffering. These are often multi-modality (they use more than one type of imaging at once). For example, nuclear imaging techniques such as single-photon emission computed tomography (SPECT) and positron emission tomography (PET), which rely on the injection of radioactive tracers to image biochemistry or tissue function, are combined with X-ray computed tomography (CT) scans. These give us anatomical information about the location and size of bones, tissues and organs, and they are integrated in one piece of equipment (SPECT/CT or PET/CT) allowing researchers to get both types of information (functional and anatomical) in one scanning period, which reduces the number of procedures carried out on each animal.

Even in the case of single modality cameras this is possible. At QMUL, magnetic resonance imaging (MRI) and SPECT/CT are next to each other, so researchers can transfer anaesthetised animals between scanners, again reducing the stress and risk of repeated anaesthesia.

Imaging techniques like ultrasound and MRI are invaluable in screening animals that are developing tumours and are key to refining experimental techniques to reduce suffering. Being able to detect tumours earlier when the animals are not suffering allows researchers to begin studies at earlier, more clinically relevant stages of the disease and means the experiments can end earlier before the disease is severely advanced. Using non-invasive imaging to gauge disease stage before the beginning of an experiment means researchers can study animals with similar disease burdens, reducing variability of the experimental data. If variability is reduced, the numbers of animals in each study group can also be reduced and experiments are less likely to need repeating.

 Sharing resources to avoid duplicating research

Researchers at QMUL’s Barts Cancer Institute (BCI) are involved in several projects to share information and biological samples from cancer patients and research animals in order to avoid duplicating research.

Professor Claude Chelala, leader of the Bioinformatics Unit at the BCI, is a co-investigator of SEARCHBreast, a searchable online database that aims to facilitate sharing of archived from animals, mainly mice, that have been bred to provide models of breast cancer for researchers to work with.

Breast cancer research involving animals is a key part of thousands of studies published each year. Mice are mainly used as models of the human disease; they may have tissue transplanted that could develop into cancerous cells and tumours, or they may have been altered to include genetic material from other organisms, such as humans, so that researchers can understand the role played by genes in the disease, discover targets for treatment and test potential new drugs.

SEARCHBreast (“Sharing Experimental Animal Resources: Coordinating Holdings in Breast Cancer) was developed to reduce the use of animal models. It encourages sharing between researchers to improve efficiency in animal research and to avoid duplication. Scientists can query, share, or upload materials related to animal models of breast cancer, including genetic and transplant models. The database provides descriptions of each animal model available and associated tissue and materials, which are available for sharing between academics within the breast cancer research community on a collaborative basis.

The database provides descriptions of each animal model available and associated tissue and materials, which are available for sharing between academics within the breast cancer research community on a collaborative basis

SEARCHBreast also promotes the use of alternative models of breast cancer by aligning with another distinct resource, the Breast Cancer Now Tissue Bank (BCNTB). The BCNTB provides human breast cancer and normal breast tissue for research, and specialises in providing primary cell cultures derived from these cells. SEARCHBreast encourages scientists to consider developing 3D cell culture models as an alternative to animal models. Researchers who apply for human breast material from the BCNTB will, via SEARCHBreast, have a mechanism for matching their laboratory work to the most relevant animal models.

BCI is one of five institutes in the UK that are part of the BCNTB. Unveiled to researchers in the UK and Ireland in 2012, the BCNTB is designed to be a ‘one-stop-shop’ from which researchers are able to obtain the specimens they need. Currently, it provides access to range of specimens, comprising over 41,000 samples from breast tissue and blood/blood derivatives, and offers a bespoke cell line service. In addition, the BCNTB is also a source of professional advice for the cancer research and biobanking community.

As part of the BCNTB a unique network of bioinformatics resources has been developed (BCNTBbp). The ultimate goal of the BCNTBbp is to provide researchers with the bioinformatics tools they require to query, analyse and integrate findings from the literature, from publicly-available data, from in-house data and from experimental data generated using BCNTB-supplied samples. It serves to smooth connections to the multi-dimensional data between the BCNTB and the research community by providing a centralised information gateway to explore breast cancer data in a transparent manner. This will ensure that all breast cancer data are used to their full potential, irrespective of any bioinformatics barriers, thereby reducing both duplication of effort and the use for animal models.

Professor Chelala says: “These are two important initiatives that we are involved in at QMUL, which contribute to enabling scientists to employ the principle of the 3Rs – replacement, reduction and refinement — as a framework for humane animal research.”

Pioneering project to grow the first 3D human tumour in the laboratory


CANBUILD aims to grow the first 3D human ovarian tumour in the lab.?

Meanwhile, in another project, QMUL researchers are using bioengineering techniques to grow the first 3D human tumour in the laboratory.

The project, called CANBUILD, is led by Professor Fran Balkwill from BCI and it aims to build an ovarian tumour and its microenvironment from scratch. If successful, the project may provide better ways of testing new drugs that target the tumour microenvironment and could lead to a reduction in the use of animals for this purpose.

Even at their earliest stages, human cancers are more than just malignant cells. Other cells and chemicals that normally support and protect the body are co-opted by the malignant cancer cells to actually help them grow and spread. Research is increasingly pointing to the importance of not just targeting the malignant cancer cells but also this wider ‘tumour microenvironment’ for long-term cancer treatment.

In the CANBUILD project the multi-disciplinary team of scientists are using the latest advances in tissue engineering, biomechanics, imaging and stem cell biology to engineer a complex 3-dimensional human tumour in which the different cell types of the tumour microenvironment will communicate, evolve and grow in vitro (outside the body, in the laboratory).

Professor Balkwill says: “About half the cells in a tumour are not cancer cells, but ‘healthy’ cells such as immune cells and fibroblasts which the cancer is somehow corrupting to help it grow and spread.

“It seems logical that the best long-term treatments will come from combining both therapies that target the cancer cells with something aimed at the wider tumour microenvironment which, while not cancerous cells themselves, are supporting the cancer’s growth.

“Growing an in vitro model, which contains all these types of cells, will allow us to watch how the cells communicate and how the tumour grows, teaching us more about what is going on in this complex system and hopefully giving us a model we can test new drugs on.”

 App helps to optimise tissue sharing

Mouse in lab

A tissue-sharing app could help to reduce the number of mice needed in research. Photograph: Jorge Duarte Estevao

Researchers at QMUL and University College London are collaborating to develop an online tool called OptiShare - a web and mobile app that will help reduce the numbers of animals used in scientific research.

At present there are few systems in place to facilitate the sharing of animal tissues between researchers, even where colonies of mice are duplicated within one institution. OptiShare is designed to facilitate this type of sharing.

“It works a bit like an online dating app,” explains Dr Jean Marie Delalande. “It links researchers, ‘matching’ users who have common interests in their research (since they are using the same mouse models). It creates opportunities for research collaborations.”

Once researchers have registered with OptiShare, the software pinpoints research groups that have ongoing experiments, it creates a "match" with those planning to humanely kill the mice, and sends notifications to the groups to arrange sharing of tissues. In practice, the app flags up the availability of scientifically valuable tissues that would otherwise be discarded. OptiShare is designed to optimise the use of research animals, increasing the scientific value of these research animals and preventing other animals from being killed to obtain these tissues. It should accelerate the rate of new scientific discoveries while also reducing the cost of animal research.

The idea first came to Dr Delalande and his colleagues when he was working at UCL and needed a mouse with a particular genetic mutation to study the development of the gut when the development of blood vessels was impaired. He only needed the gut tissues, so the other tissues would have been discarded. However, he knew colleagues who were studying heart and brain development in the same type of mouse. He contacted them and they agreed to share the organs from the same animals. Dr Delalande thought this was ethically sound, scientifically and financially efficient, and that animals should be used like this whenever possible.

After discussing this with Dr Alan Burns, a senior lecturer at UCL, the idea for the sharing app was born. They then teamed up with Enrique Rodríguez Morón, a software developer, who created the OptiShare website and app to bring the idea to reality.

OptiShare has the potential to ensure the best use of animals by avoiding this needless duplication, as well as saving on the costs associated with keeping animals

“London is a great place to develop OptiShare because there are so many research institutions here,” says Dr Delalande. “We can use the power of London and its many research institutions, so that not only can researchers within the same institution share tissues, but they can also share tissues between institutions. Eventually, we want it to become available nationally and internationally.”

The app helps researchers to meet the requirements of EU and UK legislation on the use of animals in research. At European level, Directive 2010/63/EU states that "Member States shall facilitate, where appropriate, the establishment of programmes for the sharing of organs and tissues of animals killed". In the UK, the Home Office guidance says researchers should "collaborate with other users at your establishment and other places to share animals and tissues wherever feasible". However, to date, progress on sharing tissues has been limited.

“The current situation whereby there can be wasteful duplication of mouse colonies, even within one institution, is not sustainable, neither ethically nor financially,” says Dr Delalande. “OptiShare has the potential to ensure the best use of animals by avoiding this needless duplication, as well as saving on the costs associated with keeping animals.”

Now the researchers are seeking funding to help them to develop the app further, so that it is slick and user friendly. To participate in a trial please contact Dr Burns for UCL, or Dr Delalande for QMUL. Eventually, the team wants to spread the system further afield throughout London and to European centres, such as Erasmus Medical Centre (Rotterdam, The Netherlands) where they already collaborate.