Research

Project duration

5 years 0 months

Granted 19/12/2018

Project purpose

• (a) Basic research
• (b) Translational or applied research with one of the following aims:
  • (i) Avoidance, prevention, diagnosis or treatment of disease, ill-health or abnormality, or their effects, in man, animals or plants.

Keywords

Aging, bladder overactivity, oxidative stress, NADPH oxidase, urothelium

What's the aim of this project?

The overall aim of this project is to investigate the role of NADPH oxidase (Nox) and its derived reactive oxygen species (ROS) in controlling bladder function and the mechanisms whereby Nox proteins act in aging bladders leading to dysfunction.

Age-related diseases present huge health and societal challenge to our aging society. In UK, 40% NHS budget is spent on patients aged over 65. One class of highly prevalent but under-studied aging disease is overactive bladder disorders (OAB). OAB seriously reduces the quality of life and is the main reason for entering into care-homes. It costs 3% NHS budget with additional financial and social burden to the community. Why OAB occurs is largely unknown and factors that predispose aging bladders to over-activity are poorly understood.A recent advance is recognition of the urothelium, the inner mucosal lining of bladder wall, as a new sensory structure that detects bladder fullness and hence controls bladder function. Urothelium can be selectively targeted with fewer side effects on normal muscle function.The bottleneck for further progress is to identify urothelium-derived factors that make aging bladders prone to overactivity. Reactive oxygen species (ROS), the oxygen-containing and chemically active molecules, are fundamental signalling molecules in many pathological processes.Oxidative damage is associated with aging and many age-related abnormalities.Pilot data indicate that ROS has a significant role in the urothelium and in bladder aging and dysfunction.However, the source of excessive ROS generation in the bladder is unknown. Nox are the unique enzymes that produce ROS as their sole function while other ROS-generating enzymes are required for physiological oxidation. Nox enzymes have generated intense interest as they can be specifically inhibited by drugs to limit excessive ROS production with little side effect. Nox has several subtypes, each with different function. Pharmaceutical companies begin to synthesize subtype-specific Nox inhibitors with an aim for treatment in patients. We will find out 1. Whether Nox subtypes exist and are functional in bladder; 2. How these Nox and associated ROS affect urothelium and bladder function during aging; 3. By what mechanisms these Nox pathways work at cell and molecular levels and the ways to therapeutically target these molecules. The outcomes from this research will provide new targets – Nox, for the development of novel drugs to control and prevent bladder overactivity and dysfunction in the elderly.

What are the potential benefits that will derive from this project?

This is the first study to examine the role of Nox and ROS in urothelial function. The outcome of this project is expected to identify a novel regulator, Nox molecules, in the newly-recognized sensory structure – the urothelium, and provide new insight into bladder aging and bladder dysfunction in the humans. It will also provide added translational value for the use of novel specific Nox inhibitors in improving aging bladder function and the potential for treating bladder disorders. Providing scientific basis for instillation of Nox inhibitors in the bladder in clinical practice.

What types and approximate numbers of animals will you use over the course of this project?

C57BL/6J mice genetically modified. The project is expected to use 1646 mice over a 5 year period.

In the context of what you propose to do to the animals, what are the expected adverse effects and the likely/expected level of severity? What will happen to the animals at the end?

To breed and maintain these mice to different ages in high standard full-barrier housing conditions with environmental enrichment and obtain the tissue from these mice after schedule-1 procedure for ex vivo experiments. None of the strains to be bred under this licence is expected to exhibit harmful effects when housed appropriately. One strain has reduced immune function against certain types of bacteria but under full-barrier environment there is no increased rate of infection. Aging mice may have some potential welfare issues due to long period of maintenance required to achieve the objectives of understanding aging under the project. Common problems are lump on the skin (incidence 7-17%), ocular problems (4-12%), and wounds (5-9%). The severity is mild and humane endpoints will be used. One strain has learning and memory defect which will only manifest when being subject to learning and memory tests and lives normally under laboratory animal maintenance conditions. This strain will only be maintained for short period of time (under 10 weeks) while being used for experiments. The severity level is mild. Genotyping and mouse identification procedures can cause transient and mild discomfort and pain. Appropriate anaesthetic/analgesic procedures and skin disinfection will be carried out to limit these momentary discomfort or pain. The severity level is mild. All the mice will be euthanized by schedule 1 procedure and the tissues from these mice will be used for ex vivo experiments. There will be no intervention to live animals.

State why you need to use animals and why you cannot use non-animal alternatives:

Major goal of this project is to characterize mammalian ROS regulation and bladder function in the process of ageing. There is no non-animal alternative for mammalian ROS regulation and bladder function. The bladder has several mutually interacting tissue layers with both electrical communications and release of chemicals to influence adjacent cells. These are in turn under voluntary and involuntary nerve controls with both sensory and motor nerves and reflex circuits which again relay to the central nervous system under brain cortex control. In particular the urothelium has been discovered to have unique “neuronal-like” function, which has never been described in other mucosal system in the body and there are many unknowns to yet to be explored in this native tissue. The body aging and aging bladder add more complexity. For these reasons no other non-animal or in vitro alternatives are suitable for this project.

Explain how you will assure the use of minimum numbers of animals:

The number of animals used has been estimated by power calculation. Data will be analysed statistically before repeating each type of experiment and the experiment will be stopped when a statistical significance is reached. In ex vivo experiments, tissues obtained from in vivo model will, where possible, be used for different experiments, for example, several inflammatory mediators to examine the ROS responses in one bladder, considering constraints of cross-interactions between different interventions, reversibility of a response, duration of the experiments, quantity of the tissue required for a given experiment; this will reduce the number of mice required. In some sets of experiments, we will also use cells isolated from ex vivo tissues and cells in primary cell culture. This will also reduce the use of mice. Tissues we don’t use will be made available to other researchers to reduce the total number of animals used.

We use power calculation of the minimal number of mice required to reach a statistical significance. From our experience, the coefficient variation about 20% for bladder tissue ROS measurements in response to neurotransmitters. To detect a 20-25% (23%) mean difference, at alpha error =0.05, beta error =20%, power 0.8, one-sided hypothesis, a sample size about 12 for each experiment is required. The total number of mice required is estimated by the total number of groups of separate experiments under different aims of the whole project and the sample size per group.

Explain the choice of species and why the animal model(s) you will use are the most refined, having regard to the objectives. Explain the general measures you will take to minimise welfare costs (harms) to the animals:

Mice are the most suitable animals to study ROS signalling and bladder function with defined genetics and gene knockout models are readily available. C57BL/6 mice have been chosen for the following reasons: 1. Well-characterized genetic background and the available transgenic modifications suited to the scientific questions asked in the project; 2. extensively- studied Nox expression, ROS signalling and bladder physiology which show great similarities to humans. 3. All the Nox subtypes under study have knockout strains on this background, but not from other species. The preliminary results are also based on this species. Young and aging mice have been chosen to study the effect of aging. 22 months are the youngest age to study the aging bladders in C57BL/6 mice. Several Nox subtype knockout mice will be used in this project to dissect the molecular mechanisms underlying Nox enzyme regulation of bladder function. The number of groups is determined by the types of separate experiments.

All the mice, including the model where the immune status of the mice (Nox2 knockout) might compromise body defence again certain bacteria, will be protected against any potential infection as they will all be maintained in a full-barrier sterile environment in our recently opened state-of-art new biomedical research facility. Environmental enrichment will provide good living conditions to the mice and reduce the stress. There will be no interventions carried out on live animals except for identification and genotyping (ear pinch) purposes. In latter case, great care will be taken to handle the mice and this will be performed by qualified persons. Appropriate disinfection measures, anaesthesia and analgesia will be taken. Mice will be euthanized as judged by humane endpoints to avoid suffering.

Project duration

5 years 0 months

Granted 22/03/2021

Project purpose

• (a) Basic research
• (b) Translational or applied research with one of the following aims:
  • (i) Avoidance, prevention, diagnosis or treatment of disease, ill-health or abnormality, or their effects, in man, animals or plants

Keywords

Vaccines, bacteria, antibiotic resistance, Outbreaks, immunology, Vulnerable populations

Animal types and life stages

What's the aim of this project?

Our research aims to:

1. Develop and evaluate new vaccines or vaccine components against pathogens affecting humans, for which there is currently no vaccines or vaccines that have insufficient efficacy, and
2. Investigate and understand the mechanisms underlying successful immune responses to infections, in order to support the generation of novel and more efficient vaccine strategies.

Why is it important to undertake this work?

Infectious diseases are a leading cause of death and disease worldwide, in particular dramatically affecting vulnerable populations (children and elderly), and low-income countries. Vaccines represent the best hope for tackling the devastating effects of several dangerous pathogens, because of their cost effectiveness at preventing diseases, as opposed to using diagnostics and treatment, and because of the ability to tackle epidemics, pandemics and outbreaks, as the COVID-19 pandemic has clearly shown Three categories of pathogens are considered in this project:

1. Antibiotic-resistant bacterial diseases, or diseases contributing to the rise of antibiotic resistance.
2. Bacterial diseases that have the potential to cause outbreaks and epidemics
3. Pathogens affecting the vulnerable (paediatric and aging populations).

Therefore, developing new and improved vaccines, that may need fewer doses, be cheaper or can be given needle-free is supporting the goal for healthier lives.

What outputs do you think you will see at the end of this project?

The potential benefits for the 5-year duration of this project are to:

• Discover new vaccines or vaccine components against the selected infectious diseases, suitable for the vaccine platforms we are using (Neisseria species, antibiotic-resistance bacteria, bacterial pathogens susceptible to cause outbreaks). We aim to develop or investigate several vaccine candidates for each of the diseases, including novel formulations (solid dose, needle-free).
• Establish the proof of concept and the mechanism of protection induced by these vaccine candidates.
• Provide sufficient data to support the progression of the successful candidates to clinical trial.
• Compare and identify different mechanisms by which vaccines induce the desired immune responses.
• Confirm the impact of factors such as genes, identified during clinical studies, in the vaccine-induced responses and side effects.

Who or what will benefit from these outputs, and how?

The long-term aim is to develop vaccines that will in the future be included in worldwide human vaccination programs. The diseases we are targeting affect primarily babies and young children, and particularly vulnerable populations in developing countries that can benefit most from vaccination (including the elderly). Our ultimate objective is to prevent patients from suffering and dying of these diseases. We expect to discover new vaccines and regimens that will be safe and protect humans from a number of major diseases. Our discoveries, if successful will be tested in human clinical trials and could be included in vaccination programs. In addition, a major expected benefit is the new knowledge that we aim to bring not only into the vaccinology field, but also in the immunology of each of these diseases, through publication of our research.

In addition, our program of work with new vaccines and delivery methods, such as needle-free parenteral and mucosal delivery, and new technologies such as understanding the genes involved in successful vaccine-induced immune responses may lead to novel knowledge that will significantly and lastingly improve vaccine development programs, efficacy, uptake and safety.

How will you look to maximise the outputs of this work?

To maximise the outputs of this work, we follow different strategies:

• Publication: we publish the successful as well as the unsuccessful approaches.
• Conferences: we present our data to the relevant conferences (specific for the diseases we are targeting or conferences focusing on vaccines and vaccine development).
• We look out for potential collaborators for all our programs. We have an expertise in vaccine delivery technologies, and how these can be applied to different pathogens. We thus aim to collaborate with experts in the pathogen/disease itself, which allows fruitful collaborations as the expertise and networks are different.
• Protection of intellectual property, in order to attract a commercial company to take up the further vaccine development and potentially its commercialisation.

Species and numbers of animals expected to be used

• Mice: 9,500

Explain why you are using these types of animals and your choice of life stages:

We do not use animals during vaccine design and production; however, we have to test the vaccines initially in animals before we can trial them in humans to ensure they are safe and efficacious. Responses to vaccines are complex and at present there is no other way of testing them than using animals, there is no non-animal system that recapitulates the function of the immune system.

The project will use mouse models to evaluate the immunogenicity of vaccines, and also models of human diseases to evaluate if the vaccine can efficiently protect against the diseases.

We use mice because they are the less sentient specie studied, with an immune system that is well characterized, and there is an understanding of how responses in mice may translate into what we would find in humans. There are extensive sets of reagents available for analysing vaccine-induced immune responses in mice. The project also involves genetically altered animals (GAA) in order to investigate the role of specific genes in the immune response, and to identify which part of the immune system provides the immune responses and protection.

The project involves evaluation of the vaccines immunogenicity at ages where the immune system is less reactive (the elderly).

Typically, what will be done to an animal used in your project?

For the immunogenicity studies, mice receive the vaccines by injection twice, and blood samples are taken 5 times to evaluate the response.

For the efficacy studies, mice are immunized twice, blood samples are performed twice prior to challenge. Diseases are induced through experimentally exposing the mice to the bacteria, by injection, or orally, or through spraying the cage with a solution containing the bacteria. Within 2 to 3 days, mice may become ill, and are killed before showing pain and distress, or recover if the pathogen induces a mild disease. For colonization / carriage models, mice don’t become ill.

Very rarely, other explorations of the immune responses may be carried out, involving further injections or using genetically modified mice or aged mice.

What are the expected impacts and/or adverse effects for the animals during your project?

For the immunogenicity studies, the adverse effects expected are mild, these procedures are similar to what would be performed to a human or a baby, except that mice are sedated during the injection to avoid the stress or pain from the manipulation and the injection. Blood samples are performed without sedation, as these are so quick that sedation and short drowsiness would likely induce more stress.

For the challenge studies, mice may become ill and show signs of discomfort (less mobile, loss of body weight). Mice will be monitored daily at the peak of infection and not allowed to suffer discomfort for more than 48 hours. Then they will either recover or be immediately killed. The models of infections are not expected to cause severe pain because the experiments will be stopped before mice become sick, as the effect of the vaccine can be observed by measuring the number of bacteria or virus in the body before it becomes too high.

What are the expected severities and the proportion of animals in each category (per animal type)?

Most experiments will be immunogenicity testing, with a mild severity limit (63% of mice or more). For these immunogenicity studies, mice receive the vaccines by injection, and blood samples are taken to evaluate the response. The adverse effects expected are mild, these procedures are similar to what would be performed to a human or a baby, except that mice are sedated during the injection to avoid the stress or pain from the manipulation and the injection. Blood samples are performed without sedation, as these are so quick that sedation and short drowsiness would likely induce more stress.

In up to 21% of mice, we will explore correlates of protection or the mechanism by which the vaccine induces immune responses, and the mice may be subjected to injection of substances other than vaccines. This is not painful, but the increase in amounts of procedures, or old age, increases the stress to the animals, and these studies can thus be classed in the moderate severity limit.

Finally, up to 16% mice may be subjected to a bacterial challenge, to assess the capacity of the vaccines to protect against the disease. In these studies, mice may become ill and have discomfort for up to 48hours. This is thus a moderate severity limit. The diseases are induced through experimentally exposing the mice to the bacteria or virus. The injections are performed under anaesthetic to avoid pain. Within 2 to 3 days, mice may become ill and show signs of discomfort (less mobile, loss of body weight). Every effort will be made to reduce the welfare cost to these animals by the most refined husbandry methods and providing mashed up food and water on the floor. Mice will be monitored daily at the peak of infection and not allowed to suffer moderate discomfort for more than 48 hours. Then they will either recover or be immediately killed to avoid suffering. The models of infections are not expected to cause severe pain because the experiments will be stopped before mice become sick, as the effect of the vaccine can be observed by measuring the number of bacteria or virus in the body before it becomes too high. Every effort will be made to ensure protocols are continuously refined – in particular by identifying challenge doses and routes of administration that cause reduced animal suffering and distress, and by identifying early timepoints after challenge that allow the evaluation of the vaccine effect without letting mice become ill. Control measures and humane endpoints are used so that any adverse effects experienced by animals are moderately severe at the maximum. All animals will be humanely killed at the end of a specific set of procedures. They will not be kept alive and re-used for other experiments

What will happen to animals at the end of this project?

• Killed

Why do you need to use animals to achieve the aim of your project?

The program of work involves research on live animals because there is no alternative: the evaluation of the vaccine candidates’ immunogenicity and efficacy is the essential part of the research, and there is currently no in vitro system available that can mimic the complexity of the entire immune system. In particular this work proposes to develop vaccine for human use, and thus requires use of a mammal in order to mimic as closely as possible the human immune system. Mice are the standard species used for immunogenicity testing of almost all vaccines. There is a unique body of research on vaccines in mice allowing comparisons with previous work and there are uniquely extensive sets of reagents available for analysing vaccine-induced responses in mice.

Which non-animal alternatives did you consider for use in this project?

There is to date no suitable non-animal alternative to evaluate the immunogenicity of a vaccine. However, several steps of the project benefit from non-animal work:

• When a correlate of protection is known for a particular disease (for example the serum bactericidal activity for meningococcal disease), we establish the in vitro assay in house or with collaborators, and routinely use this to avoid the need for challenge experiments in animal models.
• Evaluation of antigen expression in vitro: We quality control and verify the antigen expression or composition / appearance of the vaccines in vitro appropriate techniques, prior to injection into animals: for viral vectors by infection of mammal cells and immunofluorescence or flow-based assays, for outer membrane vesicles by SDS-PAGE and electron microscopy.

Why were they not suitable?

While non-animal alternatives are available to establish some vaccine's potency (such as measuring the amount of antigen), these are not suitable to measure the immune response a vaccine can induce. The in vitro methods do not mimic the complexity of the entire immune system, from antigen presentation, detection and response to the danger signals, migration of immune cells to lymph nodes, generation of germinal centers and generation of antigen-specific B and T-cell responses.

How have you estimated the numbers of animals you will use?

The number of animals is based on:

• The number of projects we have currently running
• The number of projects for which we are planning to apply for funding.
• Our experience of how many animals are used within each of our past project.

What steps did you take during the experimental design phase to reduce the number of animals being used in this project?

We aim to follow the ARRIVE guidelines, use statistical power calculations and the NC3R's Experimental Design Assistant to inform our experimental design.

What measures, apart from good experimental design, will you use to optimise the number of animals you plan to use in your project?

Combining studies together to minimise the number of control mice required: Many experiments necessitate the inclusion of control groups such as unvaccinated animals or control vaccines. To minimise the repeated use of control groups during the project, several test conditions are included simultaneously in each experiment, so that one control group is used as a comparator for many vaccines, rather than assessing one vaccine candidate per experiment which would mean that a control group is required in each experiment.

Sequential sampling: Across a time course, we perform tail-bleeds rather than terminal bleeds, and have developed protocols allowing the evaluation of the immune responses in small blood volumes. This allows the reduction of the number of animals needed to conduct these studies, as it allows data from multiple time points to be generated from the same group of mice rather than requiring a separate group of animals for each time point. The same approach is used for S. aureus carriage, where simple non-invasive techniques are used (swabs, collection of faeces) to measure the bacterial load, thus allowing long longitudinal studies.

Collaborations: We have established collaboration with groups or companies who have a pre-optimised challenge models to test vaccine efficacy. This allows us to perform challenge experiments without having to use extra mice for optimisation of the model. In addition, these challenge models are well characterized, reproducible and performed by experts, thus allowing a minimum number of animals per group and a minimum stress to the animals.

Pilot studies: When initiating a new protocol, as a challenge study, we perform pilot studies, thus using fewer mice to establish a predictable and reproducible model and then expand its use to assess the scientific question. Similarly, for a new vaccine composition, a short pilot study is performed to assess the reactogenicity and immunogenicity, and if suitable then we expand to more extensive protocols (including comparator groups, long term responses).

Sharing tissues: We collect several organs for assessment of the immune response, and whenever possible provide tissues not relevant to our protocols to colleagues.

Which animal models and methods will you use during this project? Explain why these models and methods cause the least pain, suffering, distress, or lasting harm to the animals.

Choice of mouse model: Mice have been chosen as they are the standard model species used for initial immunogenicity testing of almost all vaccines.

The project mainly involves the use of wild-type mice (BALB/c, C57BL/6, NIH), including outbred mice (CD-1). Well-characterized strains are used so that the discomfort and sensitivity to challenge agents is already known (from previous publications), for example, BALB/c for S. aureus challenge; sensitive to the challenge but in a highly predictable pattern, allowing a humane end-point to be used prior to severe disease but allowing a robust scientific readout. For enteric bacteria, C56BL/6 and BALB/c are susceptible to challenge and have a predictable pattern of infection, while 129Sv/Ev are resistant and thus allow studies of other aspects of infection.

The project also involves genetically altered animals (GAA) in order to investigate the role of specific genes in the immune response (for example cytokines, TLR receptors, chemokines, T Helper markers). We will use only established GAA strains with known phenotype, so the appropriate care and monitoring can be assigned prior to their arrival. Mice with immunocompromised phenotype (for example knock-out of a TLR receptor), the risk of infection should be absent due to the housing in IVCs.

We also explore alternatives delivery routes for vaccines that are needle-free: this includes mucosal routes (the vaccine is deposited as a drop in the nose or mouth), and parenteral delivery but without needle (gas-propelled).

We use anaesthesia for some parenteral injections, such as intramuscular injections, as these can be painful.

We also use and refine humane endpoints.

All new staff performing these injections will be trained on cadavers and then animals under anaesthesia.

Increased monitoring is used upon detection of a side effect, and during challenge studies, and pilot studies allow us to have a predictable pattern of side effects.

Refinement of bacterial challenge models: We have use and when possible improve the challenge models described in the literature, including the severity scoring system (used as a scientific readout). We have established challenge models that are reliable, reproducible and inducing low variability. Close observation and monitoring of the mice allowed us to stop experiments earlier post injections, without compromising the reliability of the model. Only experienced staff monitor the animals during the challenge studies, and we perform pilot studies of any new dose, new strain of microbe or new strain of mice, to identify humane endpoints by close monitoring of the side effects and their timing, before starting experiments in vaccinated animals.

Infectious agents that do not cause systemic illness will be used in preference to pathogen strains that cause systemic illness, where possible, as well as non-invasive exposure routes, for example to establish carriage (by simple contamination of the food and bedding, the bedding and cage is sprayed with a liquid preparation containing the bacteria, the mice become infected simply by nibbling on it or breathing the contaminated air, and this is used instead of handling the mice and applying the bacteria in the nose or mouth). This avoids stress. We also propose to perform administration of antibiotics ip over im route.

We propose to use fasting prior to bacterial challenge by oral gavage: this is an alternative instead of the depletion of gut commensal bacteria with antibiotic treatment prior to enteric infection. Moreover, shortening the fasting period using daytime fasting can be used: a study showed that fasting for 6h gave similar results to fasting for 18 h regarding gastric emptying and intestinal transit time of charcoal (Prior et al., 2009). This is partly due to the fact that mice are nocturnal and eat the major part of their food intake during the night. Therefore, shorter daylight fasting (6 to 10 hours) can reduce the discomfort to the animals. This is an alternative instead of the depletion of gut commensal bacteria with antibiotic treatment prior to Salmonella infection. A milder model of Salmonella infection without antibiotic treatment was developed (Del Bel Belluz et al., PLoS Pathog. 2016 Apr 7;12(4):e1005528), including fasting the animals overnight. Food is provided again after the infection. The course of disease is milder: and thus, is a relevant model to vaccine studies. Controlled access to food prior to infection by oral gavage will be performed only once for that purpose.

When mice are bought aged, request will be made for well-established social groups, so as to induce the least disturbance possible due to transport and new environment.

Why can’t you use animals that are less sentient?

There is a unique body of research on other vaccine types in mice allowing comparisons with previous work, including our own work, which is used to inform our studies. There are extensive sets of reagents available for analysing vaccine-induced immune responses in mice. Although mice have not proven to be perfect predictors of immunogenicity in humans, they continue to provide extremely valuable data, and are thus considered the least-sentient species adequate for this type of work.

How will you refine the procedures you're using to minimise the welfare costs (harms) for the animals?

We use the most refined route for the specific vaccine formulations, in keeping with the scientific end point. We use short anaesthesia with isoflurane with most routes when anaesthesia is appropriate (for example not used for oral gavage). The routes and maximum volumes follow the code of practice and guidelines set out in The Handbook of Laboratory Animal Management and Welfare, Sarah Wolfensohn and Maggie Lloyd, Blackwell publishing, Third Edition (2003). We also explore alternatives delivery routes for vaccines that are needle-free: this includes mucosal routes (the vaccine is deposited as a drop in the nose or mouth), and parenteral delivery but without needle (gas-propelled).

We use anaesthesia for some parenteral injections, such as intramuscular injections, as these can be painful.

We also use and refine humane endpoints.

All new staff performing these injections will be trained on cadavers and then animals under anaesthesia.

Increased monitoring is used upon detection of a side effect, and during challenge studies, and pilot studies allow us to have a predictable pattern of side effects.

Refinement of bacterial challenge models: We have use and when possible improve the challenge models described in the literature, including the severity scoring system (used as a scientific readout). We have established challenge models that are reliable, reproducible and inducing low variability. Close observation and monitoring of the mice allowed us to stop experiments earlier post injections, without compromising the reliability of the model. Only experienced staff monitor the animals during the challenge studies, and we perform pilot studies of any new dose, new strain of microbe or new strain of mice, to identify humane endpoints by close monitoring of the side effects and their timing, before starting experiments in vaccinated animals.

Infectious agents that do not cause systemic illness will be used in preference to pathogen strains that cause systemic illness, where possible, as well as non-invasive exposure routes, for example to establish carriage (by simple contamination of the food and bedding, the bedding and cage is sprayed with a liquid preparation containing the bacteria, the mice become infected simply by nibbling on it or breathing the contaminated air, and this is used instead of handling the mice and applying the bacteria in the nose or mouth). This avoids stress. We also propose to perform administration of antibiotics ip over im route.

We propose to use fasting prior to bacterial challenge by oral gavage: this is an alternative instead of the depletion of gut commensal bacteria with antibiotic treatment prior to enteric infection. Moreover, shortening the fasting period using daytime fasting can be used: a study showed that fasting for 6h gave similar results to fasting for 18 h regarding gastric emptying and intestinal transit time of charcoal (Prior et al., 2009). This is partly due to the fact that mice are nocturnal and eat the major part of their food intake during the night. Therefore, shorter daylight fasting (6 to 10 hours) can reduce the discomfort to the animals. This is an alternative instead of the depletion of gut commensal bacteria with antibiotic treatment prior to Salmonella infection. A milder model of Salmonella infection without antibiotic treatment was developed (Del Bel Belluz et al., PLoS Pathog. 2016 Apr 7;12(4):e1005528), including fasting the animals overnight. Food is provided again after the infection. The course of disease is milder: and thus, is a relevant model to vaccine studies. Controlled access to food prior to infection by oral gavage will be performed only once

When mice are bought aged, request will be made for well-established social groups, so as to induce the least disturbance possible due to transport and new environment.

What published best practice guidance will you follow to ensure experiments are conducted in the most refined way?

The Handbook of Laboratory Animal Management and Welfare, Sarah Wolfensohn and Maggie Lloyd, Blackwell publishing, Third Edition (2003).

Published literature on challenge m models: regular checks ensure we consider the shortest and mildest possible challenge models.

How will you stay informed about advances in the 3Rs, and implement these advances effectively, during the project?

We receive the regular NC3R's newsletter. One of our team members is also a member of an animal ethics committee, and this regularly exposed to the improvements made by others working with animals on science. We also keep up with the literature not only specific to the vaccine platforms we are using, but also specific to the diseases and pathogens we are targeting. We have the flexibility to implement these new advances within projects, or when initiating a new one. Most of the times, 3R's advances also improve the workload of experiments, or the reproducibility, and ultimately results in less work, thus benefiting not only the animals but the science too.

Project duration

5 years 0 months

Granted 27/07/2021

Project purpose

• (a) Basic research
• (b) Translational or applied research with one of the following aims:
  • (i) Avoidance, prevention, diagnosis or treatment of disease, ill-health or abnormality, or their effects, in man, animals or plants

Animal types and life stages

What's the aim of this project?

Neglected tropical skin diseases (skin NTDs) are a group of skin infections caused by different types of organisms. These diseases, defined by the World Health Organisation, affect the world’s poorest people and are an important factor in trapping communities in a cycle of poverty and disease due to their association with stigma, disability and mental health problems. The aim of this project is to better understand the processes that occur for one particular skin disease caused by a bacterial pathogen. By understanding the disease better, we can achieve our long-term goal of designing better treatments that might help the infections heal more quickly. At the end of this project we aim to have identified at least one drug, ideally already licenced for other diseases (so-called "drug repurposing"), that could be tested in the future.

Why is it important to undertake this work?

Skin NTDs include diseases that can cause extremely serious skin ulcers that may grow to cover large areas of the body, such as an entire arm or leg. Affected individuals are usually found in remote farming communities in developing countries, where access to healthcare is difficult. Current treatments take several months, may require hospitalisation and often leaves patients with large wounds that need either skin grafts or even limb amputation. Better treatments that improve wound healing are urgently needed as this will reduce the burden on patients (reducing deformity and long-term disability) and their families (since many patients are young teenagers requiring treatment far from home). This project will contribute to a fuller understanding of disease development and allow better treatments to be found.

What outputs do you think you will see at the end of this project?

The main outputs of this project will be increased knowledge of how skin disease caused by this organism occurs. This is very important as relatively little is known about NTDs. Outputs will mainly be in the form of peer-reviewed scientific publications and conference proceedings.

Who or what will benefit from these outputs, and how?

In the short- and medium-term, the expected beneficiaries of this programme will be different parts of the academic community such as those working on the skin NTDs, and scientists interested in how specific genes are involved in disease formation. It may also be of benefit to those involved in drug development or the pharmaceutical industry as our findings could lead to the identification of new uses for existing drugs, or new drug targets for the disease.

In the long term, this project will benefit the populations living in regions where skin NTDs are found. This includes many countries of West and Central Africa, as well as other parts of South America and Australasia. These benefits are only likely to be seen some time after the project is complete.

How will you look to maximise the outputs of this work?

We will pursue an active publication strategy. New knowledge will be widely disseminated via publication in open access journal articles. Less impactful findings will also be published in Wellcome Open Research, a dedicated open access platform for results of this nature. We will frequently share our results with collaborators and results will also be disseminated at conferences, particularly the World Health Organisation's biennial meeting on skin NTDs.

Species and numbers of animals expected to be used

• Mice: 4300

Explain why you are using these types of animals and your choice of life stages:

We will use adult mice rather than neonates in this project because their skin is most similar to the human skin that we need to mimic. Mice are a well-understood model of bacterial skin infection for skin NTDs, and have been successfully used to develop treatments that are now given to patients. Inbred mice are also available, which helps by greatly reducing the number of variables that could influence the outcomes of our experiments making our results more reliable. Furthermore, mice can be genetically engineered to remove or alter the expression of genes that are also found in humans, therefore genetically modified mice can be used to study the role of these genes in the disease. The genetic models we will use have previously provided important insights into other medical conditions. Mice are the lowest sentient species we can use to achieve our scientific aims.

Typically, what will be done to an animal used in your project?

There are two types of experiments in this project, and each mimics a skin NTD by injecting material into the skin. The first type involves infecting the skin with a small amount of bacteria, by giving one injection either into the base of the foot or into the tail. These experiments last several weeks, up to a maximum of approximately 11 weeks, because the bacteria grow very slowly. The second type of experiment involves giving one injection of a purified bacteria-derived compound that is important for the development of disease, either into the tail or the ear. These experiments are shorter, lasting up to a maximum of approximately 2 weeks. These injections are given while the mouse is asleep, under a general anaesthetic.

In both cases, some of the animals might first be given a drug that causes the cells of a genetically modified mouse to change their DNA content. Alternatively, some of the animals might be given drugs that we think will change the ways the skin will respond to the injected material. The drugs will be given to the mice in a variety of different ways, and sometimes we might surgically implant a device that automatically delivers the drugs, if a vet thinks this will be less stressful to the mice overall.

We will carefully monitor the injection site and might sometimes use a medical device similar to an x-ray machine or an ultrasound, or other “imaging” device to see what is happening inside the skin. We will make sure the mice are asleep when we do this, so that it is less stressful to them.

At the end of the experiment, we will kill the animals at pre-determined points, while the animals are no more than moderately affected by the disease, in line with the severity limits of each protocol. We will kill them by a humane (Schedule 1) method. In some cases, we want to study whether the blood vessels are working properly, and these animals will be given a general anaesthetic first so we can inject a dye into their blood stream.

What are the expected impacts and/or adverse effects for the animals during your project?

In all Protocols we will carefully monitor changes in the general condition of the mice that might indicate that they are affected by the injected material. We will also look for potential side effects of anaesthetics, drugs given and/or the genetic modifications present. However, we are not expecting the mice to show any signs, like weight loss or indications that the animals are in pain, during any protocol. This is because the bacteria make a substance that is known to acts like a painkiller. We know that mice that have had material injected into one footpad may experience changes in the way they move about due to their foot swelling up. However, they will still be able to move around the cage without distress. We are expecting changes in the appearance of skin around the site of injection, but this is necessary to understand the disease. We will monitor all animals carefully to ensure that any harms are kept to a minimum and, if they occur, dealt with promptly.

What are the expected severities and the proportion of animals in each category (per animal type)?

The expected severity for the mice used in this licence is expected to be mild or moderate. Overall, considering all Protocols except the breeding of genetically modified mice, we expect 70% of the mice to experience a cumulative severity of moderate, with the remaining 30% of the mice experiencing a cumulative severity of mild. For the breeding of genetically modified mice, we expect 99% of the mice born on the Protocol to experience mild severity, with a low risk of one of the 10 genetically modified mouse lines we will breed showing moderate severity in 10% of the animals.

What will happen to animals at the end of this project?

• Killed

Why do you need to use animals to achieve the aim of your project?

Animal use is needed to understand the disease in the context of a living being, and in particular live tissues with blood flow and an immune system. Two other important reasons are that the bacteria grow very slowly (much slower than the speed at which human cells divide in the laboratory) and the optimal growth temperatures are different (29-33°C, much lower than human body temperature of 37°C). Moreover, revealing the role of individual genes and/or pathways requires the use of genetically modified mice in which these genes and/or pathways are changed so that we can study them. Finally, pre-clinical screening of compounds that might make good treatments is necessary in live animals before testing in humans can be considered. This is because a “fully body system” is needed in which the drugs are given in the context of a body’s metabolism as well as the immune system and blood circulation. None of these can be completely substituted by other techniques that don’t use live animals.

Which non-animal alternatives did you consider for use in this project?

This project involves a range of non-animal alternatives that run alongside animal use. For instance, we test cells isolated from donated excess human skin in laboratory assays, and examine skin biopsies from patients. This has allowed us to develop ideas about how the infections take hold. We also carry out basic biochemical assays on purified components of cells, in which we can investigate fundamental cellular processes. We have also considered developing a system in which a whole piece of donated skin can be maintained in the lab for several days/weeks.

Why were they not suitable?

The laboratory studies are informative, but tell us about cells bathed in nutrients in artificial conditions, rather than in their natural context. The clinical studies are informative, but present a snapshot at quite a late stage in disease which can mask the molecular "trigger" for the symptoms the patient experiences. The skin explant models do not survive for long enough for infections to establish, and also suffer from the temperature barrier, since they must be maintained at 37°C (not suitable for the bacteria causing this neglected tropical skin disease).

How have you estimated the numbers of animals you will use?

The numbers of animals used are estimated based on the experimental design within each protocol. For pilot studies we used a so-called Resource Equation to determine groups sizes according to the law of diminishing returns. Data available from the scientific literature, or these pilot studies, are used to perform power calculations to determine appropriate group sizes for definitive experiments. We then estimated the total number of animals by adding up the number of animals used for pilot studies, the number of mice needs to perform studies that follow the evolution of the disease as it progresses, and the number used in definitive experiments. In each case this assumes a defined number of experimental endpoints, and that the experiment will be performed for 10 candidates over the course of the licence.

One Protocol describes the breeding of strains of genetically modified mice for use in other Protocols. We used calculators published in the scientific literature to estimate the minimum number of breeding pairs/trios would be required to obtain the required number of animals, and well-established general principles of animal husbandry to estimate the size of the colony needed to maintain this number of breeding pairs/trios for the minimum possible time to meet our scientific aims.

What steps did you take during the experimental design phase to reduce the number of animals being used in this project?

During the experimental design phase, we took advantage of the NC3R’s Experimental Design Assistant tool to consider all aspects of the experimental design. This included performing power calculations for group sizes, blinding and randomisation strategies, and statistical methods for the analysis of the data. This tool was extremely useful to ensure that the experiments will be performed in a way that reduces animal wastage by ensuring that the experimental data collected is robust and reproducible. Advice has also been sought and taken from a statistician on the design of all experiments in this license, and this will be an ongoing process throughout its lifetime.

What measures, apart from good experimental design, will you use to optimise the number of animals you plan to use in your project?

Wild-type in-bred animals will be supplied by a registered breeder. Only the correct number of animals will be purchased. We will also breed genetically modified mice which will be used in the project. Where possible these will be bred as homozygous lines, and compared to wild type animals of the same strain supplied by a registered breeder, in order to avoid wastage by breeding excess animals. We are aware of the issues around genetic drift, and will take this into consideration while maintaining any colony of genetically modified mice. Breeding of these animals is controlled by a breeding plan, devised by the NACWO, NVS and the research team, to minimise animal wastage. Wherever possible, we will re-use tissue that has already been collected in exploratory studies to investigate candidate pathways before embarking on definitive experiments. Moreover, tissue will be shared within local and collaborating research groups.

Which animal models and methods will you use during this project? Explain why these models and methods cause the least pain, suffering, distress, or lasting harm to the animals.

We will use established models of a skin NTD in mice that involves injection of bacteria, or material derived from them, into the skin. This is necessary to mimic how the disease appears in patients. The injection sites are chosen based on established knowledge about how to model human skin diseases in mice. The footpad, tail and ear are all sites with characteristics most similar to human skin, and this means our results will be scientifically meaningful.

In both models we will perform pilot studies to compare within these injection sites to determine the approach which tells us what we need to find out about the disease, whilst causing the least distress to the animals. We will do this by taking measurements in the skin tissue after the mice have been killed, or by using imaging techniques on live mice, and compare them to our findings in patients. Like the infections in humans, these models are extremely unlikely to cause pain to the animals. Since the skin condition will deteriorate the longer the infections last, we have clearly defined humane endpoints to ensure that the animals never experience more than moderate severity. We will also use pilot studies to determine the earliest possible point in the experiment that answers our scientific questions so that these harms are minimised.

In order to test our ideas about the involvement of certain genes in the development of neglected tropical skin diseases, we will use two alternative approaches. These are either genetically modified mouse lines, or drug treatments known to reduce the activity of these genes. Since the genes we are interested in are also involved in other non-tropical diseases, they have been studied previously. This means we have been able to read these previous studies and select strains and/or drugs that cause the least harm to the animals.

Why can’t you use animals that are less sentient?

Skin is a complex organ which has changed throughout evolution, and is very different between mammals (such as humans and mice) and lower vertebrates or invertebrates. Therefore, skin diseases in general require mammalian models. For skin NTDs in particular, some mammals (but not lower vertebrates or invertebrates) have been found to get similar diseases, especially when the infection is picked up from an environmental source that they are exposed to. In addition, mammalian skin changes before and after birth, including changes to the way the blood flows and immune system works. This means that immature life stages are not appropriate for this project. The length of time taken for the infections to develop means that experiments under terminal anaesthesia cannot be used.

Therefore, an adult mammalian model is required to replicate clinical skin infections, and mice are the lowest sentience to meet the scientific aims and get results we can use to develop treatments for people in the longer term.

How will you refine the procedures you're using to minimise the welfare costs (harms) for the animals?

In order to minimise the possible harms, mice will be monitored with health checks that have been tailored specifically to them, and the protocol being followed. We will use anaesthesia and/or painkillers as advised by a vet to prevent and control any potential adverse effects (including stress or pain). We will implement additional care should specific animal husbandry be advised due to the models (for instance, the use of soft bedding/nesting material). We will perform pilot studies that will enable us to further refine doses, injection routes and the length of experiments so that we can use the most refined protocols. Where possible, we will preferentially use the model involving injected purified components, rather than infection with live bacteria, as these experiments are shorter. Our results will be reviewed regularly and integrated with new knowledge/experience/other publications.

In order to reduce the stress to animals, and hence the need for restraints, during the taking of animals' measurements we will previously undertake regular familiarisation handling of the animals and will use positive reinforcement. For instance, we aim to use refined handling techniques such as tunnel handling or cupping to reduce stress. We will also minimise the number of times each animal is handled by combining techniques where possible (such as measurements and drug administration if required on the same day), and using anaesthesia/sedation should this be advised by the NACWO/NVS.

What published best practice guidance will you follow to ensure experiments are conducted in the most refined way?

We will follow guidelines from the NC3Rs and the Code of Practice for the Housing and Care of Animals Bred, Supplied or Used for Scientific Purposes. We will follow the PREPARE guidelines and checklist when planning our experiments.

How will you stay informed about advances in the 3Rs, and implement these advances effectively, during the project?

The 3Rs are regularly discussed and best practice shared during the User Forum for the unit, which the research team attend and participate in regularly. Members of the team also sit on our institution’s AWERB. These animal welfare meetings, advice disseminated by the NACWO/NVS/NIO, and accessing regularly updated on-line sources of information about the 3Rs (such as NC3R, Norecopa, Frame), will allow us to keep up to date with advances in the 3Rs and implement them in this project. Throughout this PPL, we will review our results regularly and integrate any new knowledge/experience from other publications and collaborative network.

Project duration

5 years 0 months

Granted 03/09/2021

Project purpose

• (a) Basic research
• (b) Translational or applied research with one of the following aims:
  • (i) Avoidance, prevention, diagnosis or treatment of disease, ill-health or abnormality, or their effects, in man, animals or plants

Keywords

Cancer, Immunotherapy, Oncolytic virus, HOX genes

Animal types and life stages

What's the aim of this project?

To assess the ability of anticancer agents, either alone or in combination with other anticancer agents, to treat or prevent tumour growth

Why is it important to undertake this work?

To assess new targeted treatments for cancer. To provide data to support new clinical trials for cancer patients.

What outputs do you think you will see at the end of this project?

We will use cancer killing viruses which retarget the patient’s immune system against the cancer. This form of treatment is called viral immunotherapy. Bladder cancer offers intriguing opportunities for viral immunotherapy. Bladder cancer is the seventh most common cancer in the UK, with over 10,000 new cases annually in the UK. Approximately 70% to 80% of patients with bladder cancer are initially diagnosed with an early form of the disease called non-muscle invasive disease. Superficial, non-muscle invasive bladder cancers (NMIBCs) are managed with surgery followed by chemotherapy and/or immunotherapy. The use of bacteria Bacillus Calmette-Guerin (BCG) as an immunotherapy for bladder cancer and its proven effects of reducing recurrence and progression and improving disease-specific survival have revolutionized the treatment of this malignancy. However, the potential for serious side effects of local and systemic bacterial infection as well as the fact that there is a significant (30%) group of non-responder patients to this treatment highlights the need to develop future immune-based therapies that overcome these problems. Our group has developed a novel viral immunotherapy based on a cancer killing virus called Coxsackievirus A21 (CVA21). We have shown that CVA21 has strong immunotherapeutic properties in experimental bladder cancer models and in patients in a clinical trial. A new license would allow us to study a new range of cancer killing viruses expressing bacterial proteins to better target and create a more potent immune response for this disease and for other cancers.

Prostate cancer is the most common non-cutaneous cancer in men worldwide, with an estimated 1,600,000 cases and 366,000 deaths annually.   Despite the high long-term survival in localized prostate cancer, metastatic prostate cancer remains largely incurable even after intensive multimodal therapy. Prostate cancers are generally considered to be ‘cold’ tumours with minimal T cell infiltrates, lacking a type I IFN signature and chemokines and containing immunosuppressive cells such as myeloid derived suppressor cells. This non-inflamed phenotype is thought to be largely responsible for the disappointing lack of sensitivity of prostate cancer patients to immune checkpoint blockade (ICB) therapy. Whilst cancer immunotherapy with checkpoint blockade (ICB) therapy has revolutionized the treatment of patients with certain malignancies, clinical trials in prostate cancer have shown ICB to have very limited efficacy. The potential benefits of these orthotopic models is to allow our group to study a combination of ICB with other immunotherapies, to increase the potential of ICBs seen in other malignancies.

Oesophageal cancer – of which adenocarcinoma (OAC) is the predominant subtype in the United Kingdom - is a highly aggressive malignancy, ranking sixth among all cancers for mortality, with 5-year survival rates of approximately 15%. Surgically resectable disease is primarily treated with chemo+/-radiotherapy and surgery. Systemic chemotherapy is associated with significant side effects and morbidity. While chemotherapy or chemoradiotherapy may modulate the tumour microenvironment (TME) towards tumour rejection, they are unselective treatments with significant systemic toxicity. Hence novel approaches to the treatment of OAC are required. Immunotherapy may achieve more precise and potent immunomodulation with delivery of biological agents. Creating novel immunotherapeutic strategies which utilise multiple agents to achieve the greatest effect offers the opportunity for significant advances in the treatment of OAC. Innovation in therapeutics has been hampered by a lack of animal models which replicate the tumour microenvironment with an intact immune system.

Who or what will benefit from these outputs, and how?

We hope in the short term, to obtain knowledge for the wider scientific community, which will be disseminated through scientific papers and international conference’s. Data from each protocol can be used to help design drug dosing regimens and minimise/eliminate adverse effects in other protocols within this license. Our group has a sustained and proven track record in development of basic science and clinical translation in the fields of cancer killing viruses and therapies that turn the immune system against cancer. Our group has a collaboration with our local NHS Hospital which has a pharmacy, with a dedicated gene/viral therapy facility and a treatment centre with the infrastructure to conduct and manage early phase trials with such biological agents. Thus, in the long term, data generated from this license may translate into clinical trials, which may lead to novel treatments or combinations of treatments for cancer, specifically, for oesophageal adenocarcinoma, prostate and bladder cancer.

How will you look to maximise the outputs of this work?

We will obtain knowledge for the wider scientific community, which will be disseminated through scientific papers and international conference’s.

Species and numbers of animals expected to be used:

• Mice: 3890
• Rats: 600

Explain why you are using these types of animals and your choice of life stages:

Adult mouse and rat cancer models represent a mammalian system that is well characterized and available in defined genetic backgrounds, greatly reducing the number of variables that could influence experimental outcomes.

Typically, what will be done to an animal used in your project?

Tumours will be inserted into the animals by injection or surgery. Animals with a confirmed tumour, will be treated with anticancer agents over maximum of 4-6 weeks. Tumour load may be assessed by ultrasound, IVIS or x-ray imaging. In some instances, treatment may lead to the regression of tumours. These animals may be re-challenged with fresh tumour cells to look for long lasting immunity against the specific tumour type. Alternatively, immune cell components, may be depleted by antibody neutralization, to understand which immune cells are important in the regression of the tumour. Animals will be culled according to scientific or humane endpoints.

What are the expected impacts and/or adverse effects for the animals during your project?

In all tumour models we monitor changes in: hunching, an altered demeanour, tumour dimensions and tumour ulceration. To minimise suffering of the animals timely monitoring and careful observation of the mice will ensure that any harms are kept to a minimum and when they occur, dealt with promptly.

What are the expected severities and the proportion of animals in each category (per animal type)?

In our last licence, using experiment tumour models 97.1% of animals show no signs or symptoms of clinical illness and were culled at scientific endpoints (mild). But 2.2% of animals showed an alter gait, 0.22% showed hunching/low demeanour and 0.33% Tumour ulceration and therefore these animals were culled at Humane end points (moderate).

What will happen to animals at the end of this project?

• Killed

Why do you need to use animals to achieve the aim of your project?

One of the key aspects to our research is to study the role that our novel anti-cancer agents play in modulating the immune system to kill cancer. For this we require a live animal with a fully functioning immune system.

Which non-animal alternatives did you consider for use in this project?

In vitro (test tube experiments): Each treatment is tested in cell culture for tumour killing properties. This data is then compared against well studied anti-cancer agents such as chemotherapeutic agents (eg Cisplatin) on the same tumour cell line. Also, from tissue culture experiments, we can obtain information on the type of cell death caused by anti-cancer agents and whether proteins that will affect the patient’s immune system are produced.

Ex vivo (experiments on patient biopsies): Ex Vivo testing of anticancer agents using the slice culture model using patients tumours.

Our group has developed the tumour slice model which is another way we can screen anti-cancer agents. The model allows us to slice tumours taken directly from human patients and from animal models. Cell lines grown in in vitro culture normally contain a single type of cell which is propagated in a single layer. In contrast tumours are complex and multi-layered structures. The tumour tissue slice cultures are preserved in the native state of the tumour, for further study. Tumours are obtained directly from operating theatres at NHS Hospitals. Tumours are slice using a vibrating blade. Tumour slice cultures are maintained under atmospheric oxygen levels. A tumour can be divided into up to 30 slices which allows multiple drugs and viral doses to be tested at any one time. This system has allowed our group to refine and reduce the number of animals needed for our cancer therapeutic studies by allowing us to look at the novel therapies directly in primary human tumour cells. It can also be used with animal tumour slices.

Why were they not suitable?

Limitations of cell culture (test tube experiments)

1. Cells in tissue culture do not contain morphologic structures seen in the tumour in patients. Cell culture also does not contain the mix of population of cells seen in tumours ie stroma and immune cells.
2. Cell culture assays can only be studied over a number of days whilst animal experiments can give us scientific information over weeks. This limits treatment and response time for studying the anti-cancer agent.
3. The cell culture model does not allow us to examine cancer-immune cell interactions. It is essential for our studies to understand the interaction of the immune system with cancer cells and with anti-cancer agents.

Advantages of slice cultures:

1. Tumour-killing properties of anti-cancer agents can be studied by staining slices with proliferation markers or staining cell death markers.
2. Induction of proteins that may alter the patient’s immune system can be measured Which could suggest whether an anti-cancer agent will prime an immune response.

Limitations of slice cultures:

1. Currently tissue slices can only retain tissue morphology in culture for between 96 and 127 hours. This limits treatment and response time for studying cancer agents.
2. The slice culture model does not allow us to examine cancer-immune cell interactions. It is essential for our studies to understand the interaction of the immune system with cancer cells and anti-cancer agents.

In conclusion, the slice culture model offers a way of pre-screening anti-cancer agents directly onto patient biopsies. It allows us to quantify tumour killing properties of these agents and measure immune response proteins. Unfortunately, however, this does not replace animal studies completely because it doesn’t represent a fully active immune system and therefore, we can’t fully study the effect of anti-cancer agents.

How have you estimated the numbers of animals you will use?

It is difficult to estimate the number of animals that will be used before pre-screening of anti-cancer agents has been carried out in vitro. Protocol-1 is a protocol to standardise tumour growth. Previously in 2018 we used 30 mice to study tumour growth in mice. We also expect to use the same number of mice for this objective in this licence to test tumour growth (up to 30 mice per year and up to 150 mice or rats in total over 5 years). However, overall, we will use approximately 50 mice or rats per year (250 mice in total) because an additional 20 mice or rats per year will be required to grow tumours for use in the tissue slice culture model system described above. Protocol-2 is a rodent cancer heterotopic models. Based on our returns from 2016, 341 mice were used to test two different types of oncolytic virus therapy or/ and checkpoint inhibitor and another 24 mice were used in HOX gene therapeutic peptide studies. A total over 5 years of 1825 mice. Similar data was obtained on our previous project licence. Therefore, we have estimated up to 2150 animals per five years of the project license would be used in protocol 2 of this licence (2000 mice, 150 rats). Protocol-3 is a treatment and prevention vaccination model. Based on our returns from the last set of experiments under this protocol in 2015 (previous project licence) 97 animals were used for four oncolytic virus vaccine lysates experiments. Based on these previous studies we therefore estimate for this licence we will use up to 750 mice, and 250 rats per five years of the project license for protocol 3. Oesophageal adenocarcinoma (Protocol 4/5) is a new area of research for our group. Protocol-4 will produce N= 60 L2-IL-1β (Tg[ED-L2-IL1RN/IL1B) mice that have been bred with wt C57BL/6J mice, for use in protocol 5. To breed this group of animal, we require 9 breeding females, each will have 5 litters. Breeding of these GAA mice is controlled by a breeding plan, devised by NACWO, NVS and members of our lab, to minimise animal wastage (https://www.ncbi.nlm.nih.gov/books/NBK43325/). Detailed calculations are shown in appendix 6. Protocol-5 will use 60x animals produced on protocol-4 to study effect of anticancer agents on oesophageal adenocarcinoma. Protocol 6 is an intra- bone (metastatic) tumour model, which may need 20x animals to set up. Our current design of experiments will use 15x mice per treatment group with two treatment groups in each experiment. We may carry out this protocol 6x times on the license, therefore in total we may use 200 mice per five years of the project license. Protocol 7 is an Orthotopic Prostate Murine Model. This protocol has the same design of and number of experiments as protocol 6, it also will need similar numbers of animals for its set up, therefore we may use 200 mice per five years of the project license. Protocol 8 is an orthotopic bladder tumour model. In the last year this protocol (Previous Project Licence) was carried out we used 97 rats (485 rats/5years) in this procedure. Therefore, we estimate we will use up to 500 rats or mice (per five years of the project license. Protocol 9 It is difficult to estimate the number of animals that will be used before pre-screening of anti-cancer agents has been carried out in vitro. Group sizes will depend on, which protocol the animals enter after protocol 9. We expect to use 50 mice/10 rats per year and therefore 250 mice/50 rats throughout the license.

As described above, our group has developed in vitro and ex vivo pre-screening procedure for anti-cancer agents, to limit the use of animals used in our studies. If an anticancer agent does not show the desired efficacy in vitro then in vivo experiments will NOT be carried out. It is difficult to estimate the number of animals that will be used before pre-screening of anti-cancer agents has been carried out in vitro, but based on our previous licences, we estimate up to 4490 animals will be used over the five year period of the project license. Due to the broad variety of projects being currently investigated in this laboratory the numbers and protocols have been stipulated to cover all eventualities now and in the future. In practice, the actual numbers of animals in use at the moment is below this estimate. It is our intention to continuously reassess in vivo projects and refine them.

The design of the individual experiments will ensure that the animals used are both necessary and sufficient; more animals than necessary would lead to an unethical loss of life with no additional gain in information, whilst insufficient numbers of animals would mean that the animals used are wasted because the scientific question cannot be answered using that data.

Because the tumour models / treatments to be tested are not confirmed before pre-screening of anti-cancer agents has been carried out in vitro, it is difficult to be definitive about the required sample sizes at this stage. However, the following principles will apply, and we offer past data from similar work as examples to help inform the sample size calculations required here. If new relevant data become available before the experiment is to be carried out, preliminary sample size calculations for the relevant experiments will be revisited to ensure the calculations are based on the most up to date information. Due to the exploratory nature of the work, the experiments will be powered at 80%. In general, the significance level will be 0.05, but where multiple comparisons are required (for example experiments with more than two groups) the family-wise-error-rate will be held at 0.05 by making an adjustment to the significance level appropriate to the number of planned comparisons. All sample sizes will be inflated by a factor of 10% to allow for attrition, for example if animals need to be killed at humane endpoints and therefore cannot complete the study. Where we have used standard deviations from previous studies to inform the calculations, we will increase the standard deviation assumed by 10% to allow for differences in variability across studies.

Protocols 2, 3, 5, 6, 7, & 8 involve parallel group design experiments, with tumour volume post-treatment as the outcome measure. Baseline data for tumour volume will also be measured. Most of these experiments will involve two parallel groups, a control group and an active treatment group. However, Protocols 2 and 3 govern the testing of heterotopic models screening a wide range of anticancer drug combinations and may include experiments with three or four groups, typically one control group and two or three active treatment groups. All the experiments will be analysed by ANCOVA (analysis of covariance), with treatment group and baseline measurements as covariates and will output estimates of mean tumor volume, adjusted for baseline. Using ANCOVA adjusts for baseline differences between groups and has greater power than a t-test leading to smaller required sample sizes. The sample size calculations for ANCOVA require an assumption of the correlation beween baseline and outcome tumour volume, if this is not available from previous work. Where this is the case, we have assumed a moderate correlation of 0.5. The sample size will then be the number required to detect a difference in tumour size post-treatment between the two groups using ANCOVA adjusted for baseline, with 80% power, alpha of 0.05 and assumed correlation r between baseline and outcome measures, where r is obtained from previous studies or is set at 0.5. For experiments with more than one treatment contrast of interest, the study will be powered to detect the smallest pairwise difference and adjusted for the number of treatment contrasts being studied. It is likely that some treatment contrasts will require large sample sizes; in this case, the study is not powered to detect such a difference, but the results may serve as signals of efficacy and used to help inform future research.

In the following paragraphs, we outline some examples in further detail and the sample sizes can be seen in the Table 1. Protocol 5 will be testing the effect of anti-cancer agents on orthotopic oesophageal tumours. There are few data published in this field about reductions in tumour size using these treatments, so we will base our calculations on published data on human breast cancer organoids in nude mice from Sach et al 2018. In this study, the control group had an average tumour size of 235 mm3, with standard deviation (SD) of 135 mm3). Average tumour size in the treated group (Afatinib) was 75 mm3 (SD 20 mm3) (appendix 5a). This study will require 9 animals per group, after allowing for 10% attrition. The experiments for Protocols 6, 7 and 8 are similar, and calculations will be conducted as described above, using the most up to date relevant data available to inform the calculations. All assumptions made for such calculations will be captured and available for scrutiny.

Protocols 2 and 3 cover heterotropic models that may require more than two parallel groups. In this case, we will power the study for several independent t-tests and will allow for multiple comparisons by reducing the signficance level for each test to control the total Type 1 error rate.

We will use Protocol 2 as an example of how this will work.

We will use data from a study on the trampC2 prostate model in which the average tumour size in the control group is 208 mm3 (SD 189 mm3). There are three possible treatments of interest; reovirus, PD-1, and reovirus / PD-1 combined. The mean tumour size and SD for each treatment group are 120 mm3 (SD 111 mm3), 143 mm3 (SD 89 mm3) and 21 mm3 (SD 23 mm3) respectively (appendix 5a). There are six potential treatment contrasts here if every pairwise comparison in included, but the potential effect sizes may be so small in some cases that a) there is insufficient signal of efficacy to warrant further research into that treatment comparison and b) the sample size required to detect such a small difference would be very large (see Table 1). The study will therefore be powered to detect the three potential differences that are of greatest interest. In this study, 18 animals per group would have 80% power to detect the difference between reovirus and the combined treatment, if such a difference exists, and if the standard deviations in the two groups are similar to those seen in Annel. It would also be powered to detect differences between PD-1 and combined treatment, and between the control

Protocol 10 will be testing the effect of anti-cancer agents on Orthotopic Mammary Gland Tumour. This this Orthotopic breast cancer model is new to the lab and therefore we have not treated it with Hox gene-related anticancer therapy. But previously we have treated breast cancer SC tumours with these agents. Therefore, we will base our calculations on our publication Morgan et al 2012. In this study, the control group had an average tumour size of 1351mm3, with standard deviation (SD) of 651 mm3) (at day 30). Average tumour size in the treated group (HXR9) was 497 mm3 (SD 158 mm3) (at day 30) (appendix 5a). This study will require 9-10 animals per group, after allowing for 10% attrition and non-tumour growth. An extra five animals per group will be added for gene expression and IHC studies, meaning we will require a total of 15 animals per group.

What steps did you take during the experimental design phase to reduce the number of animals being used in this project?

In order to limit the number of mice used in our experiments we will firstly test our novel anticancer agents both in in vitro cell culture models and using an ex vivo tumour slice model. The in vitro models will allow us to establish the half maximal inhibitory concentration (IC50) for the cell line that we are using and establish the potency and mechanism of action of the anticancer agents that we are studying.

As described above our group has developed a tissue slice culture model system. This system has allowed our group to refine and reduce the number of animals needed for our cancer therapeutic studies. We can obtain tumour tissue from protocol 1 (Standardisation of tumour growth after injection of live tumour cells). After the tumour growth model has been studied and the animal is culled the tumour can be divided into up to 30 slices which allow multiple drugs and viral doses to be tested at any one time. These may give us an indication of possible drug and viral doses for use in protocols 2-3 therefore refining and reducing the number of animals needed for a project. A further source of in vivo tissue material could be obtained from the PBS control tumours from protocols 2-3. This will therefore maximise the amount of scientific information obtained from each animal used in a procedure. A wide spectrum of treatment types and routes described are used to screen a wide range of novel anticancer drug combinations and establish efficacy and toxicity in vivo. We have refined routes of treatments such as IV, IP, injections, infusion, gavage etc based on a comprehensive review of the scientific literature and extensive experience acquired in previous license. Data from this protocol (1, 2 3 and 9) can be used to help design drug dosing regimens and minimise/eliminate adverse effects in orthotopic protocols (4, 5, 6, 7 and 8) within this license. Imaging in vivo will allow us to obtain longitudinal data giving us more information and reducing numbers of animals needed for our studies. Throughout the life of the license, we will analyse the data produced and conduct regular reviews of the literature, enabling us to gain more knowledge to help assess appropriate sample sizes.

Advice has been sought and taken from our University statisticians on the design of all experiments in this license, and this will be an ongoing process throughout the life of the licence.

What measures, apart from good experimental design, will you use to optimise the number of animals you plan to use in your project?

All animals used in 1, 2, 3, 6, 7, 8 and 9 protocols will be supplied by registered breeders. Only the correct number of animals, will be purchased, based on sample sizes which will be set using power calculation’s. Protocol 4 allows us to breed GAA (L2-IL-1β (Tg[ED-L2-IL1RN/IL1B) mice, for the induction and treatment of oesophageal adenocarcinoma tumours in protocol 5. Breeding of these GAA mice is controlled by a breeding plan, devised by NACWO, NVS and members of our lab, to minimise animal wastage appendix 6. Wherever possible tissue, will be shared within research groups within the university and other institutions.

Which animal models and methods will you use during this project? Explain why these models and methods cause the least pain, suffering, distress, or lasting harm to the animals.

The majority of work carried out under previous project licences (70/7347 (XF3112393) and PBE74785E) has be SC tumour flank model such as C57/BLACK mouse /B16.F10 cells a model of melanoma. These heterotopic models, which are generally viewed as benign, so are seen to address researchers’ obligations to minimise suffering. In this new licence we will still use these heterotopic model to obtain scientific data, but also they can be used to help design drug dosing regimens and minimise /eliminate adverse effects in orthotopic protocols within this PPL.

In this new licence a small number of studies will be carried out using IV injection. Other tumour cells that will be used in either SC or IV tumour flank models include K1735, A375 (melanoma), AY-27, EJ, T24, Ku19-19, 5637, RT112, TCCSUP-1, VMCUB, MB-49 (bladder) DU145, LNCAP, Tramp c2. PC-3, (prostate) SKOV-3, PEA1, 2, (Ovarian) MDA-MB-231 ZR-75-1, MCF7, SK-BR-3, (breast), CT-26, CaKi-2, HT29, SW620, HCT-116 (colon) PL45, PANC-1, PSN 1, AsPC1, HPAF II, BxPC3 (Pancreatic) A549, H520 (Lung) LN-18 , U87 (Brain) and Renca (renal). These will be grown in immuno-competent (C57BL/6, BALB/c and C3H/H) or immuno -deficient mice (Nude, Scid). These are examples of possible models to be used, but other models may be investigated either by a comprehensive review of the scientific literature or pilot tumour growth studies (see protocol 1) as recommend by Workman et al 2010 [37]. Pilot tumour growth studies using small numbers of animals can indicate patterns of local and metastatic tumour growth. Any adverse effects due to tumour progression can assess and the clinical score sheet can be adjusted. These studies can help identify the correct humane endpoints. Data from these studies can help define group numbers in order for experimental time frames.

Based on the data from heterotopic models, we will be able to carry out orthotopic models in such areas as oesophageal adenocarcinoma (protocol 4-5), intra- bone (metastatic) cancer (protocol 6), prostate cancer (protocol 7) and bladder cancer (protocol 8). These models can be used to obtain more clinically translatable data on novel anti-cancer agents.

Protocols 4 and 5 Oesophageal cancer – of which adenocarcinoma (OAC) is the predominant subtype in the United Kingdom - is a highly aggressive malignancy, ranking sixth among all cancers for mortality, with 5-year survival rates of approximately 15%. The TME is a complex interaction between the tumour, host immune cells, fibroblasts, chemokines and other factors such as hypoxia and acidosis. An orthotopic model is the best way of replicating the TME, as the tumour develops within the native organ architecture of the cancer to ensure close similarity to the human TME. The L2-IL-1β genetically modified orthotopic mouse model allows for the generation of OAC in-situ at the squamo-columnar junction of the mouse, as seen in OAC in humans. This model will give us a TME that we can trial novel immunotherapies to advance the care of OAC beyond what has been possible to date.

Protocols 6 and 7 will allow us to study the effect of anti-cancer treatments on two forms of cancer: primary and metastatic cancer. Protocol 7 describes an orthotopic prostate cancer model involving the surgical injection off prostate tumour cells into the anterior prostate lobe of mice. Protocol 6 With this orthotopic intra- bone (metastatic) model, we will surgically inject cancer cell lines into the tibia, leading to the development of orthotopic metastatic tumours with a clinically relevant TME.

Protocols 8 To obtain the most appropriate microenvironment for transitional cell carcinoma (bladder) cancer, tumour cells must be allow to attach to the bladder lining and grow into the bladder void. This can only be achieved by damaging the bladder layer using this protocol. Implanting tumours into the tissues of origin within the animal results in development of cancer in an appropriate microenvironment which includes for example stromal cell infiltration, including specific immune cells.  A major component of our research is to study the role that our novel anti-cancer agents play in immunogenic cell death and modulating the immune system.

Why can’t you use animals that are less sentient?

Reliable models of cancer which have been used in peer reviewed studies previously have been proposed [37]. These have led to successful translation into human trials. The mice and rats proposed are representative of the human diseases: the molecular and immunological properties of the tumour cells are assessable in these models, particularly in response to therapy, and closely mirror the human context [37]. The models are also selected on the basis of our previous experience with them, reliability and reproducibility of results, and that the experiments are completed in a short time to minimise distress to the animals (Workman et al 2010).

How will you refine the procedures you're using to minimise the welfare costs (harms) for the animals?

There is a wide spectrum of treatment types and routes described in this licence for screening a wide range of novel anticancer drug combinations and establish efficacy and toxicity in vivo. We will be constantly refining routes of tumour implantation/treatments such as IV, IP, injections, infusion, gavage etc based on a comprehensive review of the scientific literature and extensive experience acquired in previous PPLs. An example of the type of well fare refinement we will try to bring in is discussed in Protocol 8 (Treatment of orthotopic bladder tumours with anticancer agents). A major objective of this protocol (8) is to refine our model by trialling replacements for the acid/alkaline wash such as poly-L-lysine or trypsin treatments, to limit the severity of the model. Imaging in vivo will allow us to obtain longitudinal data giving us more information and reducing numbers of animals needed for our studies. The health checks we propose are based on knowledge acquired in previous PPLs and comprehensive review of the scientific literature, these will be under constant review throughout the life of the PPL.

What published best practice guidance will you follow to ensure experiments are conducted in the most refined way?

The Workman et al. (2010) guidelines for the welfare and use of animals in cancer research are an invaluable resource, setting out appropriate humane endpoints and relevant clinical signs to monitor. We will also use the NC3Rs website which provides a number of resources to help with further refinement to our research. This include advice and guidance on common procedures such as blood sampling, which we have used in this application. This website also have species specific pages dedicated to non-aversive mouse handling and genetically altered mice.

How will you stay informed about advances in the 3Rs, and implement these advances effectively, during the project?

All the members of the Oncology lab, who are carrying out in vivo’ studies, will attend the Biomedical Research users group meeting at university of surrey. Also a researcher from the team is a member of AWERB of Surrey University. These animal welfare meetings will allow our lab to have an understanding of current 3R issues within the research community. Throughout the life of the PPL, we will analyse the data produced and conduct regular reviews of the literature, enabling us to gain more knowledge to help refine protocols and procedures in terms of the 3Rs.

Project duration

5 years 0 months

Granted 05/10/2021

Project purpose

• (a) Basic research
• (b) Translational or applied research with one of the following aims:
  • (i) Avoidance, prevention, diagnosis or treatment of disease, ill-health or abnormality, or their effects, in man, animals or plants
  • (ii) Assessment, detection, regulation or modification of physiological conditions in man, animals or plants

Keywords

heart, arrhythmia, mitochondria, ion channel, electrophysiology

Animal types and life stages

What's the aim of this project?

The overall purpose of our project is to determine the link between the genetic causes of inherited arrhythmic syndromes and more general phenotypes associated with acquired arrhythmias. In doing so, we aim to advance understanding of arrhythmogenic mechanisms in order to aid future therapeutic and clinical intervention.

Our project is defined by the following key elements:

1. The mechanisms underlying arrhythmias caused by genetic perturbations
2. The possibility for drugs restoring the function of ion channels and metabolic regulators affected by such genetic perturbations to rescue atrial and ventricular arrhythmias.

Why is it important to undertake this work?

Understanding mechanisms for arrhythmia particularly in ageing remains a main stay in the diagnosis, treatment and prevention of cardiac arrhythmia. In addressing this global challenge, the use of animal models with targeted disruption of specific physiological pathways provide valuable insights into the elucidation of arrhythmogenic mechanisms. Concerns and lack of fundamental mechanistic understanding with regards to ageing and sodium channel pathologies has led to limitation in how new drugs are develop and an aged arrhythmic patient is managed in clinics.

Our work is fundamental to the understanding of arrhythmia and its propagation, a question of particular importance to electrophysiologists, but also importance to the study of biological pattern generation associated with drug administration and ageing. It will link the underlying pharmacology to cell membrane and intra-cellular currents, bridging research between drug-induced disturbances and physiological effects. It will allow this novel paradigm to be scrutinized as a platform for analysis of electrophysiological abnormalities in a range of pharmacological agents. The work will be of relevance to pharmaceutical researchers and cell biologists, who will gain insight into the role of ageing and selected drugs in cardiac function and correlation with electrical and cellular abnormalities.

What outputs do you think you will see at the end of this project?

Beyond the inevitable short-term output of an enhancement in our foundational scientific understanding of arrhythmogenic mechanisms by the elucidation of candidate molecules and associated signalling pathways, our work will have even more direct clinical implications. Notably, by addressing our objectives outlined previously, our expected long-term outputs are the identification and vetting of: (1) biomarkers for improved diagnoses and prognoses7; and (2) pharmacological targets for the development of efficacious novel drugs8. We aim to publish our data in high-impact academic journals, national and international conferences, and, by extension, disseminate them to the wider, global community of cardiac electrophysiologists – a feat we believe highly likely given our publication of over 60 papers in the last years and our presentation at more than 80 events.

Who or what will benefit from these outputs, and how?

Given that, every year, there are 50-100 Sudden Cardiac Deaths (SCDs) per 100,000 members of the population in Europe and the USA, the translation of our findings based on the laboratory Genetically-Modified (GM) murine model into diagnostic tests and pharmacological or other therapies at clinicians’ disposal will undoubtedly be perceived as worthwhile by the patients who benefit from them. Additionally the pharmaceutical industry or biotech companies may also benefit from this work given that we may potentially identify new ‘drug-able’ targets. A corollary of patient benefit is, of course, the diminished economic and health burden of arrhythmias on the NHS.

How will you look to maximise the outputs of this work?

Our experience from previous projects, as well as through our collaborators, who have held Project Licences in the past, ensures we will minimise pursuit of unproductive lines of enquiry, such as a drug soon identified as producing unacceptable side-effects. Likewise, this will maximise pursuit of novel avenues on signalling molecules and drug targets that exhibit early promise. Importantly, we aim to greatly facilitate the clinical limb of our translational research program through fostering a link with NHS hospitals, using the results of laboratory experiments involving our transgenic mouse model to focus on the electrophysiological assessment potentially in patients.

Likelihood of achieving benefits with these models and methods:

All protocols and techniques necessary for the generation of monogenic perturbations in mouse lines such as those described above have been well established by scientific consensus in the electrophysiology community and refined through our own experience in using similar Genetically-Modified (GM) murine models for 15 years under the previous Project Licences. Given the critical role that ion channels permitting currents play in the waveform of the cardiac action potential, it is highly likely that any metabolic regulator directly or indirectly modulating channel gating properties will precipitate arrhythmogenesis when either absent or overexpressed in our GM mice lines. Furthermore, owing to the function of metabolic regulators such as PGC1 as transcriptional coactivators of genes upregulated by nuclear receptors, coupled with the observed and highly substantiated contribution of an aberrant cardiac transcriptional profile to arrhythmias, our analysis of age-related homeostatic disturbances is also very likely to identify mediators of the link between metabolic dysregulation (as in obesity and diabetes mellitus) and arrhythmias.

Of the completed series of experiments working with previous collaborators, many have provided a basis on which to conduct predictive testing. Importantly, the identification of high-risk patients permits rapid instigation of protective measures, such as the installation of an implantable cardiac defibrillator (ICD), to reduce the incidence of sudden cardiac deaths (SCDs) that would otherwise arise from arrhythmic events. In fact, such diagnostic utility is evident in the implantation of more than 15,000 ICDs globally. Moreover, the use of our GM mice to understand the basis of arrhythmogenic risk in monogenic disorders has already facilitated, and will continue to permit, characterisation of molecular targets for therapeutic, and especially pharmacological, intervention. Our GM mouse model conveniently provides a means for testing not only the efficacy, but also the safety, of any pharmacological agent. We thus consider it highly likely that, with such links, our generation of knowledge on the biochemical structure, biophysics and molecular interactions of putative targets will drive future development of antiarrhythmic drugs specific to patients with metabolic disease, with the long-term aim of reaching phase I clinical trials.

Species and numbers of animals expected to be used

• Mice: 240

Explain why you are using these types of animals and your choice of life stages.

These animals have been validated to provide the most translational results that will closely resemble conditions affecting humans. It is not possible to undertake such experimental interrogations in humans.

We are aiming to study the effects of certain gene mutations in the ageing process. We will use mice up to the age of 24 months old since mice ranging from 18-24 months of age correlate with humans ranging from 56-69 years of age. This age range meets the definition of “old,” which is the presence of senescent changes in almost all biomarkers in all animals. We will not use mice older than 24 months as mice are then considered “very old” and survivorship drops off markedly.

Typically, what will be done to an animal used in your project?

Breeding, ear clipping for identification (ear notching), maintenance up to 24 months of age, drugs administration, ECG recording under anaesthesia and surgery for telemetry under anaesthesia. Humane killing with a Schedule 1 method.

What are the expected impacts and/or adverse effects for the animals during your project?

Pain during injections can occur. Otherwise, any other adverse effects may be related to the normal ageing process. For this reason, animals will be monitored regularly. Moderate to long term administration of substances/drugs can cause repeated incidences of mild pain and therefore stress to the animal. Additionally, this is coupled with the stress of repeated handling during the injection or oral gavage procedure. Insertion of needle ECG, whilst done under anaesthesia, may have some minimal lasting mild soreness at needle entry site. Animals that undergo surgery will have incision wounds that will take time to heal and this can cause low grade inflammation and pain over the healing process. There may be some initial discomfort due to post-surgical anaesthesia or due to the device implanted within the body cavity.

What are the expected severities and the proportion of animals in each category (per animal type)?

Moderate – We predict that approximately 50% of animals to be used in this license will be subjected to Protocol 3 and thus would be subjected to moderate severity. This is estimated at approximately 240 mice.

What will happen to animals at the end of this project?

• Killed

Why do you need to use animals to achieve the aim of your project?

The heart is a complex organ with several processes occurring simultaneously under physiological and pathophysiological conditions (e.g. organelle and cellular interactions , dynamic metabolic changes, hormonal alterations and more). The ageing body, especially when coupled with other conditions such as obesity and diabetes add to the complexity of the studied processes. Hence, compared to cellular or tissue models, an animal model is better suited in studying physiological processes in the heart. Cultured cell lines, although useful for initial non-translational discovery testing, cannot reproduce the complete array of molecular, cellular, neuronal, physiological, and pathological aspects of the mouse model necessary to fully understand the mechanisms investigated in this project, especially regarding ageing and additional co-morbidities. While it is evidently possible to corroborate our data with in vivo studies on human arrhythmia patients, it is not possible to attempt experimental drug rescue on them, creating the need for an animal model. Furthermore, while it is also clearly possible to conduct ex vivo experiments on atrial samples isolated from humans, such experiments would not be translatable to ventricular samples, which are: (1) considerably more difficult to obtain; and (2) unable to maintain an intact arrangement of cardiomyocytes.

This study will be performed using mice because all relevant methods and techniques are successfully established in this species, and because of the availability of genetic alterations in these species. With this approach we will be able to study specific biochemical pathways with a view to understand disease progression. Understanding the effects of interfering with the disease process at specific points in underlying pathways can lead to the development of new treatments for heart disease. All procedures will be conducted by highly skilled, licenced personnel, and non-invasive techniques will be used where possible. Components of procedures will be the minimum required to be consistent with reaching the scientific objectives. Animals will be closely and regularly monitored during the study. Surgical animals will receive pain relief as standard. Any clinical problems will be dealt with in consultation with the veterinary surgeon. There are strict pre-determined 'humane endpoints' in place, based on clinical signs, at which animals are promptly and humanely killed to ensure that any animal does not suffer unnecessarily.

Which non-animal alternatives did you consider for use in this project?

• In silico mathematical models of myocardial conduction
• Human iPSC-derived ventricular and atrial cardiomyocytes

Why were they not suitable?

It should be noted we have considered using our in silico mathematical models of myocardial conduction as an alternative. The basis of such models in the first place is the experimental data collected from our murine model. We then use the computer model to refine our experiments on the GM mice.

Furthermore, since 2018, we have also been developing human models in the form of in vitro experiments on human iPSC-derived ventricular and atrial cardiomyocytes and gradually using them as an alternative to the murine model. However, despite replacing some aspects of the work on GM mice, this in vitro model is largely complementary and thus unable to completely replace our in vivo GM mice model.

How have you estimated the numbers of animals you will use?

The design of individual experiments generally involves factorial design to maximise the information obtained. The majority of measures are quantitative and suitable for statistical analysis. Comparison between groups will be made by 1-way analysis of variance (ANOVA) or a 2-way repeated measures ANOVA followed by an appropriate post hoc test. We have reviewed a range of published studies that use similar techniques being proposed here and observed great variability in the number of animals used per experimental group. Following our discussions with a statistician, we have calculated our sample sizes for our quantitative experiments by power analysis using a significance level of 5% and a power of at least 80% (Ref: http://dx.doi.org/10.1136/emj.20.5.453).

We will need 8 groups of mice due to:

1. 2 mouse genotypes (WT or respective mutant)
2. 2 age groups (young or old)
3. 2 treatments (control vs treatment with relevant drug – please see details above).

The standard deviation within a group is unknown, hence our power calculations are conducted based on effect size. Aiming for power of 80% and significance level of 5%, an effect size of 1.51 (i.e. the difference between the means of the groups is 1.51 multiplied by the standard deviation) is consistent with 8 mice per group. Based on our previous studies, attrition rates in this kind of experiments is higher in aged mice compared to younger mice. For this reason, we need to add 2 more mice to our calculated n=8/group, thus requiring n=10/group. It should be noted that the exact numbers of animals required will vary with specific experiments as well as the estimates of the coefficient of variation for specific outcome measures; nevertheless, the general principle outlined in the example will be followed. In summary, we need 8 groups of mice at 10 mice/group for a total of 80 mice per readout.

The readouts for each experimental condition include surface ECG recordings, telemetry recordings, patch clamp recordings, and Langendorff recordings. Surface ECG and Langendorff recordings will be performed with the same group of mice, whereas separate groups are required for telemetry and patch clamp recordings. The final number of mice is estimated at 80 mice per readout for surface ECG/Langendorff, telemetry, and patch clamp recordings (total of 240 mice).

What steps did you take during the experimental design phase to reduce the number of animals being used in this project?

We have performed pilot experiments in cells to determine a range of appropriate dosage for the drugs we are aiming to use in vivo. Moreover, we have longstanding experience on maintenance and breeding of the appropriate number of animals to maintain the line. Typically, we will always have 2-3 breeding pairs for each line to minimise risks of ‘losing’ the line due to disrupted breeding. From previous experiments, we know when we might need to separate pairs if no experiments are planned. We have also consulted a statistician in our efforts to calculate the appropriate number of animals required in this study. Moreover, as part of our experimental design and good laboratory practice, we will continuously monitor our experimental needs and re-evaluate the number of mice necessary in our project and data analyses. We will measure production and breeding performance for each line and ensure only the minimum numbers of animals required are bred in any case. Finally, based on previous pilot studies we have optimised serial assessment of electrophysiological measurements and, for example, confocal microscopy experiments to generate data from the same animal, thereby reducing the number of mice necessary in our studies.

What measures, apart from good experimental design, will you use to optimise the number of animals you plan to use in your project?

We have recently refined our techniques to include, rather than merely determination of monophasic action potentials and restitution metrics, a multi-electrode array that enables detailed analysis of conduction properties. Importantly, this method is ideal for providing an experimental platform to define not only the role of individual genes in arrhythmogenesis, but also to systematically examine the effects of pro- and anti-arrhythmic interventions on arrhythmogenic substrates.

Which animal models and methods will you use during this project? Explain why these models and methods cause the least pain, suffering, distress, or lasting harm to the animals:

Animal models: C57BL/6 and S129 mice with mutations such as Scn5a+/- sodium channel haploinsufficiency; deltaKPQ sodium channel gain of function; RyRP2328 ryanodine receptor mutation, Atf5 deletion. The GAA we will use in our experiments have been validated in prior experiments and they are useful models for the project.

Methods include breeding and maintenance, ECG recordings, telemetry, patch clamp, Langendorff, drug administration.

Why can’t you use animals that are less sentient?

In principal, mice, despite having a smaller heart that works at much higher heart rates compared to humans, share considerable genetic homology and their hearts have similar cardiac conduction systems to humans. This justifies using mouse models in studies of arrhythmogenic phenomena. Furthermore, genetically modified murine cardiac models potentially fully replicate the genetic changes underlying congenital clinical conditions associated with arrhythmogenesis, without requiring prior pharmacological intervention – we have previously developed a range of monogenic mouse models in our previous Project licences (Scn5a+/- sodium channel haploinsufficiency; deltaKPQ sodium channel gain of function; RyRP2328 ryanodine receptor mutation) and will now use these models, as well as other commercially available GM models in this study. These mice make powerful drug testing models to explore the cardiac electrophysiological events associated with drug administration, as well as the response mechanisms particularly in arrhythmic conditions.

Using a transgenic mouse model will allow us to continue delving into the mechanisms by which genetic alterations coincide with complex disease fostering a proarrhythmic environment. No other species less sentient is available at the moment that will allow obtaining these results.

Large animals models are not suitable for this study when it comes to ageing as they require longer durations to age and thus lead to more welfare implications. Whilst pigs maybe an option (ignoring ageing component) the preclinical perspective is not entirely captured by the pig model. An equally valid and suitable GM model is not available in pigs and the mouse model purposed offers also the advantage of being already a validated model for these conditions and drug testing.

Given the well-established contributions of sedentarism, obesity, diabetes mellitus and metabolic syndrome to the presently increasing prevalence of arrhythmic phenotypes in the UK population, such an opportunity is of great relevance to the future clinical treatment of NHS patients.

How will you refine the procedures you're using to minimise the welfare costs (harms) for the animals?

Animals used in the experiments described herein are not expected to exhibit any harmful phenotype. For older animals, special consideration will be given to their environmental enrichment. Enrichment for aged mice will follow best-practice guidelines as recommended by the Experimental Biology Unit. We will be monitoring our mice on a daily basis. After initial recovery from surgery, adverse effects may include wound infection, which is extremely rare. We will prevent that using good aseptic technique in our procedures, perioperative analgesia and implementing a monitoring regime. In the event of any animals showing any signs of mild discomfort (e.g. swelling, redness or discharge at the operation site), but are otherwise well will treated by minimally invasive methods following advice from the NVS. The animal will be culled by a Schedule 1 method if no improvement is observed within 24 hours of treatment. In all cases, we will work closely with the NVS and seek advice on animals whose welfare is giving reasons for concern. We will use gas anaesthesia which is safer than injectable anaesthesia and we will monitor our animals during anaesthesia and keep them on heating pad. Whenever possible we will try to carry out more techniques during the same anaesthesia event to reduce them (ex. Imaging, drug administration, sampling during the same anaesthesia). We have implemented a specific welfare monitoring method for ageing signs and a details scoresheet with actions to better assess old ageing animals welfare.

What published best practice guidance will you follow to ensure experiments are conducted in the most refined way?

The most effective means of elucidating the function of genes whose perturbations are implicated in cardiac arrhythmogenesis is the integration of a GM mouse model with other laboratory experiments. Mice lines with defined and well characterised genetic background such as Scn5a+/- sodium channel haploinsufficiency; deltaKPQ sodium channel gain of function; RyRP2328 ryanodine receptor mutation; PGC receptor mutation, Atf5 deletion are readily available and suit the work of this project. The mouse lines chosen have extensively been validated and have previously been successfully used to recapitulate the human condition thus proving to be a superior translational model. All lines used in our work will thus conform to such rigorous selection ensuring that only the most reliable and translationally relevant modules are utilised.

All the mice will be protected against any potential infection as they will all be maintained in a full-barrier clean environment in our recently opened state-of-art new Biomedical Research Facility. Environmental enrichment will provide good living conditions to the mice and reduce the stress and mice will be checked at least once daily. Enrichment for aged mice will follow best-practice guidelines as recommended by the Experimental Biology Unit. All procedures will be performed by trained and competent persons and following NVS advice. We will perform ECG which are well characterised procedures carried out routinely and very well tolerated in rodents. To minimise stress ECG will be carried out under anaesthesia following NVS advice and animals will be checked after the procedures. We have extensive experience (15 years) in ECG recording and experienced problems in less than 2%.

Animals undergoing surgery may experience transient post-operative pain or discomfort. Anesthesia and pain management regimes will be administered upon advice from our NVS. Adequate analgesia will be provided at the beginning and after the surgery to minimize animal suffering. We will provide particular care for recovery after surgery and will include an animal welfare scoring sheet for each animal for health check. If the levels of pain, suffering or distress exceed severity limits indicated by the protocol, we will take advice from our NVS and, if necessary, sacrifice the animal with a Schedule 1 method.

Our animal technicians and NACWO are experienced in management of animal population health and welfare. We also have extensive experience (6 years) in surgery to implant telemetry and have rarely experienced any adverse effects (less than 5%). Surgery for implantation of telemetry device is performed under aseptic conditions and has a moderate severity limit. In addition, this surgical procedure is a well-characterized technique done routinely by our group and the implant is well tolerated by rodents. Mice will be euthanized by schedule 1 methods or under terminal anaesthesia under protocol 1,2,3,4 and, whenever reaching humane endpoints detailed in the protocols, with a Schedule 1 method.

Animals receiving drugs treatment will be monitored appropriately and whenever possible oral route of administration will be favoured to parenteral routes. Drugs tested, including their dosage and length of administration, will be selected mainly among well characterised drugs and their use will be preceded by literature review and eventual previous in vitro tests. We will use single use needles for all our injections and avoid tail handling our mice.

Animals will not be single-housed and the last mouse in a cage will be used within 1-2 days after its last cage mate is taken. The study design has included comparison between knockout and wildtype littermate controls which will reduce the biological variation and increase the sensitivity, reducing the number of animals needed.

In carrying out our work, we aim to use older mice with florid phenotypes in order to minimise their suffering and the total number of GM and WT mice in use.

No severe protocols will be used.

Finally, we will adhere to the ARRIVE guidelines in regards to the reporting of research involving animals.

How will you stay informed about advances in the 3Rs, and implement these advances effectively, during the project?

We will be checking the literature for 3Rs advancements regularly and implement these accordingly. In addition, we will be working closely with the NACWO, NIO and NVS, as well as engage with the University’s User Forum and latest guidance from the NC3Rs on recent and general advances in the 3Rs.

Project duration

5 years 0 months

Granted 11/04/2022

Project purpose

• (a) Basic research

Keywords

sleep, neuroplasticity, in vivo, rodents, dendrites

Animal types and life stages

What's the aim of this project?

This research project addresses why sleep is important and how it helps our brain function. In particular, the project aims to better understand how sleep influences our ability to process and retain information and how age influences this ability. This project focuses on the basic molecular and cellular mechanisms that are involved and how the different sleep stages (rapid-eye-movement [REM]/dream sleep and non-REM [NREM] sleep) may have a different, but complementary function in this process.

Why is it important to undertake this work?

Sleep is an integral part of human life and makes up 1/3 of our lives. Sleep disorders are a major societal problem and are associated with a wide range of human conditions. In fact, most brain disorders during development and ageing are associated with some form of sleep disturbance (e.g., insomnia, sleep apnoea, sleep fragmentation). Natural ageing is also accompanied with notable changes in sleep pattern, especially in an increase in fragmented sleep. This has led to the idea that poor sleep may be both a symptom and a cause of the progression of many brain disorders. Thus, knowing what sleep does to our brain is essential to understand what goes wrong in human conditions where sleep is impaired. We are addressing these questions in animals where the physiology can be investigated and new insights into basic mechanisms can be obtained.

Our research focuses on synapses, the site where two brain cells communicate. The brain displays a remarkable potential for change in response to new experiences in life and those changes occur specifically at synapses, a mechanism termed synaptic plasticity. Sleep and the ability of the brain to express synaptic plasticity vary considerably across the lifespan, with both evident in high amounts during early life and at lesser levels as we age. This further suggests an important link between sleep and the ability of our brain to adapt to environmental changes. Our study will thus provide important insights on the nature of this link and will establish new starting points for future investigations at both the fundamental and the clinical level.

What outputs do you think you will see at the end of this project?

1. The direct benefit of this project is to increase our fundamental knowledge of the relationship that exists between sleep and brain function. More specifically, the data collected in this research will reveal how different brain areas changes communication during different sleep stages and how this impacts the molecular landscape at synapses. Our research will also highlight how all those changes vary across age, which is essential information to understand the beneficial role of sleep across the lifespan.
2. The indirect, more long-term, benefits of this project relate to the potential for clinical applications. This research combines two types of physiological measures: electrical brain activity and large-scale molecular screenings.

Brain activity is measured using surface (electroencephalography, EEG) or intracranial electrodes. While EEG is a widely used method in humans, intracranial electrodes are also used in some patients with neurological disorders (e.g., epilepsy). Our research will try to identify electrical biomarkers associated with ageing and memory performance. Our molecular screening will be essential in revealing which cellular functions are recruited during sleep to help the brain to adapt to new experiences. This will open new avenues for therapeutic targets and will have notable implications for brain disorders known to be associated with alteration of sleep and synapses (e.g., autism spectrum disorders, Alzheimer’s).

Results from this project will therefore benefit fields in neuroscience and be of interest for anyone who studies the fundamental and clinical aspects of cognitive functions and dysfunctions and how they interact with sleep. We therefore expect to publish our findings in high impact factor journals.

Who or what will benefit from these outputs, and how?

This project will benefit our knowledge at both the fundamental and clinical levels.

In the short-term, the first beneficiaries will be neuroscientists, computational scientists, and clinical researchers in various fields related to this project (e.g., sleep, brain plasticity, memory, development and ageing).

Data from this research will increase our knowledge on the electrical and molecular changes in the sleeping brain related to learning. Importantly, these changes will be compared across ages, an aspect that is often minimised. Our project is also unique as it will broaden our understanding on the relative contribution of different sleep stages in all those processes.

The data generated during the project will also be of benefit to computational researchers. This is because we will deliver new, large-scale, data to help model the link between specific physiological measures (e.g., brain rhythms), learning and sleep. Current computational models of memory formation do not take into account the impact of sleep, a gap in knowledge our project should fill.

In the long-term, and beyond the end of this project, individuals may benefit. This is because data from our research may identify new targets to diagnose and treat certain brain disorders. Our research will provide brain activity (e.g., EEG) and molecular data that can be linked to ageing and specific pathologies.

How will you look to maximise the outputs of this work?

Results and data will be presented at international and national conferences during the lifetime of the PPL to share ideas with the broader scientific community. These include large conferences in neuroscience and sleep and more specialised conferences that are particularly helpful to establish new collaborations. Those conferences are also key to discuss negative results and alternative approaches. Negative results will be included in publications when relevant.

Species and numbers of animals expected to be used

• Mice: 1032
• Rats: 228

Explain why you are using these types of animals and your choice of life stages:

The beneficial role of sleep on brain function involves many different brain structures and physiological systems that can only interact when kept intact in vivo. Furthermore, sleep states and brain activity, similar to those known to be important for brain function in humans, are only seen in mammals and can therefore not be studied in lower organisms or simplified systems (i.e., in vitro). In this context, the combination of a wide range of molecular and genetic tools, as well as available data on the relationship between sleep and brain plasticity, make rodents our model of choice. All methods (physiological measures and behavioural paradigms) we propose to use have been developed and validated in rodent models. This will help ensure the success of our proposed experiments and will also allow us to provide useful information for comparison with already published fundamental and clinical data in this area.

We are proposing to use ‘young’, ‘adult’ and ‘old’ animals in our studies because sleep and the ability of synapses to modify vary greatly with age. This will allow us to understand which physiological mechanisms are preserved or changed at different ages, and how brain cell communication during sleep is altered with ageing.

Typically, what will be done to an animal used in your project?

Some animals will be used for breeding purposes only.

Most animals (not involved in breeding or pilot experiments) will undergo one or two brain surgeries and after a period of recovery, a behavioural assessment.

All animals will undergo one surgery to place wireless devices to measure the electrical activity of brain cells. The procedure is performed under general anaesthesia, typically lasts less than 1.5 hours and animals recover to normal behaviour in 2 to 3 days.

A subset of animals will also have fibreoptic probes implanted in the brain during the same surgery and will not require an additional incision. The fibreoptic probes are used to either assess or manipulate activity in specific groups of brain cells. Addition of these probes extends the duration of surgery to about 2 hours but the full recovery period of the animals remains the same (i.e. 2 to 3 days). Animals that receive the fibreoptic probes need to undergo a separate injection procedure prior to the surgery. The injection involves the delivery of a substance into the brain and is necessary to allow the fibreoptic probe to work. This more minor procedure is performed under anaesthesia, lasts no more than 1 hour and has a period of recovery of 1-2 days.

Finally, animals may be subject to behavioural experiments. These experiments last a maximum of 72 hours during which brain activity is monitored/manipulated in combination with different types of behavioural paradigms such as housing in the enriched environment cage (up to 12 hours), learning a simple task and/or short sleep deprivation (< 6 hours).

A two-week rest period will be allowed between each surgical procedures and the start of behavioural manipulations. Thus, animals that undergo all procedures undertake these across a minimum of 5 weeks, starting with the brain injection and finishing after behavioural manipulations.

Less than half of the animals (38%) will go through all the procedures and the rest will either only receive one surgery for wireless device placement before behavioural experiments (43%) or used for breeding (13%) and pilot experiments (6%)

To minimise stress, animals will be habituated to behavioural and recording set-ups for several days prior to the experimental phase. At the end of the behavioural experiment, animal will be humanely killed (according to the ASPA Code of Practice) for tissue collections.

What are the expected impacts and/or adverse effects for the animals during your project?

The interventions involved in our experiments include breeding of genetically modified mice, monitoring of brain activity using electrodes, imaging/manipulating brain activity using fibre-optic light stimulation, and behavioural experimentations. These are considered as follows:

• Breeding is not expected to generate any adverse effects as these have not been observed previously.

• Surgical interventions for brain injections and electrodes/optical fiber probe implantations are expected to cause temporary (1-2 days) pain and discomfort post-surgery which will be alleviated by analgesia. This may include irritation at the site of surgery or where the implant is positioned (i.e. head or dorsal flank).

• Behavioural manipulations include short term (6 hours) sleep deprivation, simple learning tasks and housing in an enriched environment, none of which is expected to cause any adverse effects. Some stress is expected when animals are first introduced to the behavioural and recording environment, and this will be mitigated by a habituation phase (3-day) prior to the experiments. Sleep deprivation (≤ 6 hrs) has been kept to the minimum required to test our hypothesis. Housing in an enriched environment is known to increase the animal’s well-being.

What are the expected severities and the proportion of animals in each category (per animal type)?

The severity for the mice and rats used in this licence is expected to be mild to moderate. The cumulative severity has been noted as moderate due to the two surgeries and behavioural experiments. This will be experienced by 80% of mice and 95% of rats. The remaining animals will either be breeders (15% of mice) or included in pilot experiments (5% of mice and 5% of rats) which are expected to experience a cumulative severity of mild.

What will happen to animals at the end of this project?

• Killed

Why do you need to use animals to achieve the aim of your project?

Sleep is a complex state, involving changes in the entire organism and many physiological parameters (e.g., temperature, hormone levels, and type of brain activity). The interaction of all those factors is necessary to fully express the beneficial role of sleep for brain function. It is therefore necessary to study the mechanisms underlying this role in the intact organism (i.e., in vivo).

Which non-animal alternatives did you consider for use in this project?

We have considered using in vitro approaches (e.g., brain slices, dissociated neurons) and in silico (e.g., computational models) in our research.

Why were they not suitable?

While questions about sleep function have been in some cases addressed using in vitro or in silico models, those approaches are not appropriate to address the scientific questions in our research that relies on dynamic interactions with environmental stimuli. Simplified models (e.g., in vitro) are also inherently associated with disrupted connections between brain cells and can therefore not adequately reproduce the complete array of biological interactions and mechanisms investigated in this project. To this date, in vitro approaches using dissociated cells in a petri dish are very limited to study sleep functions and poorly recapitulate brain cells activity found during sleep. The other alternative approaches involving mathematical models are also limited to mimic the complex interaction that exists between the brain and our environment, which is a critical process studied under this project. Thus, there is no feasible alternative that would entirely replace the use of a living animal and would allow the project objectives to be met.

How have you estimated the numbers of animals you will use?

The numbers of animals have been estimated according to the methods and experimental design for each Objective, the expected magnitude of changes in our measures (based on existing literature and experience from the principal investigator), and the information provided by the commercial suppliers on the mouse lines for breeding plans. The number of animals and planning of experiments have been established with advice from our in-house statisticians and with the help of the NC3Rs Experimental Design Assistant tool. Of note, all groups will have an equal number of males and females unless specified otherwise. Although we do not expect to see major sex differences in our measures, some studies suggest that sex, especially hormone fluctuations, may influence the ability of brain connections to change. We thus include gender as a variable across ages when we estimated the numbers of animals in each group.

Pilot experiments with small numbers of animal numbers will inform and refine the methods to obtain physiological and behavioural measures.

What steps did you take during the experimental design phase to reduce the number of animals being used in this project?

We used the NC3R's Experimental Design Assistant (EDA) to help with power calculations for group size and strategy for blinding of experiments. Given the number of factors to consider for each experiment, the statistical methods for data analysis were discussed with our in-house statistician, as suggested by the EDA.

We implemented efficient experimental designs to apply statistical analysis that will assess simultaneously the contribution of several factors (e.g., “age”, “experience”, “brain state”) on our output measures without increasing the number of animals for each experiment

To reduce the number of experimental groups, we will use repeated measures to allow within animal comparisons for some outputs measures (e.g., brain activity, memory performance). The use of telemetry approaches allows the recordings of several parameters simultaneously (electrical activity, movement, brain temperature) in the same animal, which also contribute to reduction.

What measures, apart from good experimental design, will you use to optimise the number of animals you plan to use in your project?

Other aspects that contribute to optimising the number of animals include:

1. The development of a breeding strategy using online information from transgenic mouse lines suppliers and in consultation with the NACWOs and NVS. The use of transgenic mouse lines will allow to optimise the specificity of our physiological measures, which contributes to both reduction and refinement.
2. The inclusion of pilot studies to establish a reproducible behavioural assessment of memory performance in mice and rats. Pilot studies will also be used to optimise the protocol for animal sample collection and processing prior to its application to animals undergoing the combined physiological and behavioural measures. Those pilot studies will also provide extra tissues to compare results in animals without wireless implants, which is important to validate our findings.

At the end of the experiments, brain tissue and other organs will be harvested. Extra tissue will be made available to other researchers within and outside the University of Surrey.

Which animal models and methods will you use during this project? Explain why these models and methods cause the least pain, suffering, distress, or lasting harm to the animals:

Sleep states and electrical brain activity, similar to those known to be important for brain function in humans, are only seen in mammals and can therefore not be studied in lower organisms or simplified systems (i.e., in vitro). In this context, the combination of a wide range of molecular and genetic tools, as well as available data on the relation between sleep and synaptic plasticity, make rodents our model of choice to address the scientific questions in this research

All the methods and behavioural paradigms we propose to use have been specifically developed and validated in the rodent model for more than three decades. Of note, in vivo electrophysiological recordings have been applied to rodents for more than 70 years. Over this time a number of refinements designed to reduce suffering have been included such as improved surgical techniques and recovery procedures, reduction of the size of the implants, and non-tethered recordings.

Finally, our research group, including the principal investigator, have extensive experience in rodent behaviour and in vivo brain activity monitoring (electrophysiology > 15 years; fiber optic > 8 years). The wireless telemetry system for EEG measures has been used in-house for several years and procedures have been refined to minimise pain and distress to the animal over the years following the advice from the NVS such as peri-operative analgesics regimens and improvement in the telemetry device and electrodes placements.

Why can’t you use animals that are less sentient?

We are interested in the role of sleep stages and specific brain waves during sleep that have only be characterised in mammals. The questions addressed by our study can therefore not be answered in lower organisms such as flies or nematodes, which are alternative models used to address some questions related to sleep.

We use animals at different life stages because we want to specifically characterise the similarities and differences in the physiology underlying the relationship between sleep and brain function across the lifespan. Results obtained in one life stage do not necessarily translate to others and this distinction is critical to understanding the role of sleep in brain function across ages.

How will you refine the procedures you're using to minimise the welfare costs (harms) for the animals?

Animal purchased from breeders will be acclimatised to their accommodation for at least 7 days prior to the start of any procedures. All animals (purchased or bred in-house) will be familiarised to handling using refined methods such as tunnel or cupping handling.

Animals will be monitored with health checks and score sheets designed for the protocol and ages. The use of anaesthesia and analgesics will be implemented as advised by the NVS to prevent any potential adverse effects, including stress and pain, to the animals.

Old age can be associated with deterioration of general condition and monitoring regimes and health check will be adapted to include signs of age-related diseases so that these subjects can be humanely euthanized. As aged subjects might be more sensitive to temperature fluctuations during anaesthesia, we will consider special recovery heating chambers and parenteral fluid administration or whatever is more appropriate as advised by the NVS for protocols involving surgeries.

To reduce the stress of animals during behavioural manipulations and brain signal recordings, we will habituate the animals to the behavioural and recording cages and devices at least for 3 days prior to the start of the experiments. From the PI’s experience and reports from other labs, rodents habituate well to novel environments, including recording and behavioural arena so we do not anticipate any signs of stress during the habituation period. It has been reported that mice, especially males, can express some aggressive behaviour when put in cage with additional male companions (e.g., enriched environment cage). Any mice that display stress or aggressive behaviour will be excluded from the experiment. Pilot studies will be implemented to reduce the stress of animals during specific behavioural paradigms.

What published best practice guidance will you follow to ensure experiments are conducted in the most refined way?

We will follow guidelines from the NC3Rs and ARRIVE guidelines.

How will you stay informed about advances in the 3Rs, and implement these advances effectively, during the project?

The 3Rs are regularly discussed and best practice shared during the BRF user forum at the University of Surrey, which the principal investigator chairs and organises with help from the BRF manager. The principal investigator is also a AWERB member at Surrey and stays informed about on-line sources of information about the 3Rs through subscription to NC3Rs, RSPCA and Norecopa newsletters. Throughout this PPL, we will review our results regularly and integrate any new knowledge/experience from other publications and collaborative network.