Abstracts for all the speakers are below. To go directly to a particular abstract click on the Title & Name below.
In vitro 3D all-human blood-brain barrier model for the investigation of drug delivery and cancer metastasis studies - Pilkington, G.J., Maherally, Z., Willcock, A., Murray, S.A., Higgins, S., Lalatsa, K., Barbu, E., and Fillmore, H.
A multimodal neuroimaging study of psychosis - Robson, S.E., Hall, E.L., Palaniyappan, L., Kumar, J., Skelton, M., Christodoulou, N.G., Qureshi, A., Fiesal, J., Katshu, M.Z., Liddle, E.B., Liddle, P.F., Brookes, M.J., Morris, P.G.
Functional studies with human gastrointestinal tissues: Raising the bar and challenging the dominance of rodents
Knowles, C, H.1 and Sanger, G, J.1
1 Blizard Institute, Queen Mary University of London
Human tissue functional studies determine the significance of mRNA/
protein expression, validate functional data using recombinant receptors and prove or dismiss animal models. However,
- tissues are removed from patients, so what is ‘normal’?
- variations due to surgical techniques, medications and differences in patient’s age, sex, genetics and disease need to be addressed
Human gastrointestinal neuromuscular functions are studied in response to electrical stimulation of intrinsic enteric neurones. Extrinsic afferent nerve discharge is recorded from mesenteric nerves. Tissues are obtained after informed consent by teamwork (scientists, surgeons, theatre staff, pathologists). Functional viability depends on collection/re-oxygenation times and surgical techniques (open-modified surgery provides shorter ischemia time than laparoscopy; Sanger et al, Br J Pharmacol 2013;168:28).
‘Normal’ is macroscopically-normal tissue ≥10cm from a gastrointestinal tumour in patients without obstruction/chemotherapy. Patient information (sex, age, ethnicity, medical history, medication) identifies variations. Translational value is established using drugs evoking relevant responses at therapeutically-relevant concentrations (eg. increased cholinergic activity by motilin/ 5-HT4 agonists which promote GI motility). By contrast, laboratory mouse colonies are strictly controlled but different strains (or even the same strain from different suppliers/countries) do not necessarily generate similar data (Keane et al, Nature 2011;477:289).
Experiments with human tissues reduce animal use directly and indirectly by establishing how the latter need not reflect human physiology. Thus, significant functional, structural and genetic differences between humans and rodents are increasingly being identified (Sanger et al, 2013;Ibid).
In vitro 3D all-human blood-brain barrier model for the investigation of drug delivery and cancer metastasis studies
Pilkington, G.J.1, Maherally, Z.1, Willcock, A.1, Murray, S.A.1, Higgins, S.1, Lalatsa, K.1, Barbu, E.1, and Fillmore, H.1
1 University of Portsmouth, Portsmouth, UK
The greatest challenge in tissue culture is, perhaps, where multiple cell types are cultured in combination. Such cells may have different growth requirements and population doubling times, therefore compartmentalized culture media conditions and appropriate seeding densities must be configured. In this context, where astrocytes, brain vascular endothelial cells, and pericytes are involved, modelling the blood brain barrier (BBB) has proved to be particularly challenging.
We have recently established all human (2D and 3D) models of the BBB based upon the combination of human cells cultured under human serum supplementation conditions with hitherto un-reportedly high trans-endothelial resistance (TEER) levels and used them in the study of nanoparticulate delivery and lung cancer cell metastasis.
We are now testing a range of different nanoparticles in crossing the BBB model using both ECISTM and cellZscopeTM as well as studying CD15 interference in the context of non small cell lung cancer metastasis. In order to assess the role of different vascular blood flow rates on TEER and on cancer cell adhesion to blood vessel endothelium we have used the Vena8 Endothelial+™ biochip (Cellix Ltd) which is used for cell-cell rolling and adhesion assays on cell monolayers and cell culture under shear flow conditions to interrogate CAMs in brain metastasis. This microfluidics system permits live cell imaging to accompany data on adhesion under different flow conditions.
Humanising drug safety testing – a pragmatic validation study
Coleman, R.A.1, Tsaioun, K.1 and Archibald, K.1
1 Safer Medicines Trust, Kingsbridge, UK
Despite ever-increasing efforts, the pharmaceutical industry is failing to reduce the number of new medicines ultimately proving ineffective or toxic in human subjects. While greater use of human in vitro test methods may serve to reduce this problem, the formal validation process applied to such tests represents a major hurdle to their adoption. We contend that what is really needed to justify the adoption of any new test is not a demonstration of an ability to identify all potential safety issues, but a clear demonstration that it adds value or is superior to whatever is currently in use – ‘pragmatic validation’. A study based on such pragmatic validation, comparing the value of a range of human-based in vitro test methods with established regulatory tests, is currently underway. Importantly, all the in vitro tests have previously undergone significant market testing, they have all passed 2 phases of the rigorous EPA ToxCast Program, and are already used internally to various degrees by many large pharmaceutical companies. The pragmatic validation approach we are testing now is designed to identify where human in vitro-based models can identify toxicities missed by the largely animal-based tests on which the pharmaceutical industry and regulators currently rely. The study employs a range of marketed drugs that passed regulatory safety testing but were subsequently withdrawn, having caused serious toxicity in human subjects. Each of these drugs is paired with a negative control, i.e. a structurally and/or functionally similar marketed drug that does not exhibit such toxicity. This study is now being conducted as a distinct part of US EPA’s Phase 3 ToxCast in vitro profiling program. Data are to be made publicly available when testing is completed, at such time they will be submitted for detailed analysis to compare the performance of the in vitro tests with the regulatory regime that secured the original marketing approval. On completion, the outcome of this unique study will be presented to the regulatory authorities with the aim of developing appropriate documents for use by the industry and published in peer-reviewed media. Preliminary data will be presented.
Murine models of human disease: why we must think outside the mouse
1 Physicians Committee for Responsible Medicine, Washington D.C., USA
Mus musculus – considered to be the epitome of human mimetic species – forms the cornerstone of biomedical research today. However, decades-long research efforts have clearly shown that murine models often fail to provide reliable, reproducible, and translatable insights into human disease pathogenesis and therapeutics. Incompatible findings have raised translational concerns in various fields such as type 1 and type 2 diabetes, cardiovascular disease, cancer, inflammatory disorders, ageing, and neurological disorders. These translational discrepancies are primarily due to immutable species differences that occur at every juncture from nucleic acid to whole organism level, further confounded by biological variability and experimental limitations. We review here the most noteworthy murine-human species differences that significantly impair translation and propose strategies by which researchers can begin to ‘demurinize’ and, conversely, humanize the study of human disease pathogenesis and therapeutics. The time has come for researchers to accept that no amount of molecular or physiological tinkering can lead to accurate recapitulation of human disease in a mouse, even in the presence of conserved genomes and molecular pathways and adherence to rigorous research methods. We are in dire need of a paradigm shift in which we transition from the quest to generate ‘better’ mouse models to the prioritized use of human-based in vitro, ex-vivo, in vivo, and in silico methods more readily transferable to humans in order to advance the understanding, prevention, and treatment of human disease.
Human Thiel embalmed cadavers – developing the optimum training and device testing model
McLeod, H.M.1, Cox, B.F.2, Eisma, R.3, Melzer, A.2, Houston, J.G.1
1 Cardiovascular Imaging Research Unit, University of Dundee, Dundee, UK
2 Institute for Medical Science and Technology, University of Dundee, Dundee, UK
3 Centre for Anatomy and Human Identification, University of Dundee, Dundee, UK
Cadaveric training and device testing models offer superiority in comparison to animal models in terms of anatomical accuracy, variation and potential emulation of the clinical experience. Currently, standard cadaveric models are limited by their lack of function, e.g. respiratory motion and blood flow, in comparison to the dynamic “living” nature of animal models, resulting in animal models remaining the preferred model of choice. The objective of this project is to address this problem by developing function within the cadaveric model by exploiting the unique properties of Thiel embalming. A number of circuits within the vascular tree of Thiel embalmed cadavers are perfused by extracorporeal circulation via a heart lung bypass machine. Successfully established and evaluated circuits, include the common femoral model, venous model, central circulation and hepato-portal model.
Various imaging modality protocols have been adapted (X-ray, Ultrasound and MRI), to verifying flow conditions within the vessels of Thiel cadavers, demonstrating the superiority of the model. Further applications have also been explored including the evaluation of the hepatic model during an advanced interventional radiology training course for junior doctors. Device development was demonstrated using the venous model to test the deployment and imaging of a prototype inferior vena caval filter. The introduction of flow to the vascular system of cadavers, increases the fidelity and potential applications of the model, whilst providing a scientifically valid alternative to using animal counterparts.
New developments in polarised epithelial layers for drug and nanoparticle uptake studies
Garnett, M.C.1, Vllasaliu, D.2, Argent R.H.1, Falcone, F.H.1 and Stolnik, S.1
1 University of Nottingham, Nottingham, UK
2 University of Lincoln, Lincoln, UK
The Caco-2 gut epithelial model is commonly used to study drug uptake from oral formulations. There is now also a considerable interest in studying the transport of biomolecules and nanoparticles across gut epithelial cells for both drug delivery and nanotoxicology. Due to its structure, basement lamina should form a potential barrier to the uptake of macromolecules and nanoparticles, so a basal lamina needs to be present in these models to provide realistic data. Caco-2 cells grown on transwell membranes showed no basal lamina components, but these were present in polarised Calu-3 lung epithelial cell cultures. We therefore decellularised polarised Calu-3 cultures to provide a basal lamina. Characterisation demonstrated that the remaining basal layer stained for basal lamina components and acted as a barrier to diffusion of macromolecules in a size dependent manner. When Caco-2 cells were grown on these basal lamina coated transwells, the Caco-2 cells assumed a more normal phenotype with taller cells, more microvilli and less tortuous tight junctions. The cells supported on basal lamina also produce polarised layers when grown in serum free medium, which does not occur with Caco-2 cells grown directly onto transwells. Growing Caco-2 cells in the absence of serum will be more appropriate for drug uptake studies to eliminate interference from drug binding. We will also be investigating whether our new model gives a better correlation with reported human drug uptake.
A multimodal neuroimaging study of psychosis
Robson, S.E.1, Hall, E.L.1, Palaniyappan, L.2, Kumar, J.2, Skelton, M.2, Christodoulou, N.G.2, Qureshi, A.3, Fiesal, J.4, Katshu, M.Z.2, Liddle, E.B.2, Liddle, P.F.2, Brookes, M.J.1, Morris, P.G.1
1 School of Physics and Astronomy, University of Nottingham, Nottingham, UK
2 Faculty of Medicine and Health Sciences, University of Nottingham, Nottingham, UK
3 Kevin White Unit, Liverpool, UK
4 Herschel Prins Centre, Leicester, UK
Psychosis is a disturbance in the perception and interpretation of stimuli in the environment and in the mind, which causes symptoms such as hallucinations and delusions. It is a feature of various psychiatric disorders, including schizophrenia and bipolar disorder. Despite its prevalence, the underlying neuronal basis of psychosis is not yet known. Moreover, animal models are limited in their scope for modelling the complexity of symptoms experienced in humans.
This project aims to use non-invasive neuroimaging techniques to investigate differences in brain structure, function and chemical composition between individuals with psychosis and healthy controls. We have tested 50 patients and 46 controls, obtaining magnetoencephalography (MEG) data while the participants completed three tasks; magnetic resonance imaging measures of brain structure and function at rest and during a task; magnetic resonance spectroscopy measures of metabolite concentrations in particular brain regions; as well as a range of clinical and behavioural data.
To date, we have observed statistically significant differences between patients and controls in MEG measures of inhibitory neuronal activity and in the concentration of glutathione, an antioxidant previously shown to be involved in the pathophysiology of schizophrenia. In the next stage of analysis, we will investigate whether neuroimaging data relates to measures of clinical symptom severity. These findings could ultimately lead to identification of measurable biomarkers, which can be targeted with behavioural and pharmacological therapies.
What is ‘Responsible Research and Innovation’ and what might it offer to animal replacement science?
1 University of Nottingham, Sutton Bonington, UK
Using animals in research is an intrinsic feature of a wide variety of innovation processes across techno-industrial societies; however, animal use is also highly contested. Increasingly, arguments based on ‘good science’, as much as ethical concerns, are behind this contestation. The 3Rs framework has become conventionalised within animal research policy and practice over the past five decades, as well as being incorporated into legislation internationally. Applying the 3Rs is often understood as a demonstration of scientific rigour, along with responsible and ethical research practices. Recently, a new science-society policy discourse: ‘Responsible Research and Innovation’ (RRI), has become influential in the EU and UK policy context. Broadly, RRI emphasises the importance of reflexivity and broader societal inclusion throughout the lifecycle of an innovation process. The influence of RRI appears to be growing within science funding bodies, who are beginning to embed dimensions of responsible innovation into the funding and management of research.
Based on documentary research and interviews with key actors working in relevant policy areas, this paper will explore what RRI might offer to researchers working in the area of animal replacement science. The paper will suggest that while the 3Rs framework already includes some elements of RRI, other dimensions are currently lacking. In particular, ensuring that there are ‘deliberative’ opportunities for the general public and stakeholders may be an important aspect for animal replacement scientists to consider.
Replacement of animal derived products for prevention of rabies in under-developed countries
Szeto, T.1 and Ma, J.1
1 St. George’s Hospital Medical School, London, UK
We are developing a production platform, using plant biotechnology for protein drugs called monoclonal antibodies. These drugs have many uses, including the treatment and/or prevention of various infectious diseases. In this project we focus on rabies, a disease that predominantly afflicts developing countries (e.g., Africa and Asia). In these under-developed areas, current treatment of suspected rabies infection involves early administration of rabies antibodies (RIG), supplies of which are expensive and in short supply. RIG is currently derived from immunised humans or more commonly in underdeveloped regions, horses, but for some time, a number of rabies groups around the world, have believed that RIG could be replaced by monoclonal antibodies.
To provide a low-tech (low cost) production platform amenable to developing countries, our research has focused on producing monoclonal antibodies secreted from transgenic plants cultivated hydroponically. Focusing on a candidate monoclonal antibody with very high anti-rabies activity, we have shown that it is possible to manufacture relatively high levels of protein and that the purification of the monoclonal antibody is extremely simple. We are working to optimise cultivation conditions to maximise the quantity of monoclonal antibody produced by investigating the underlying biology of plant protein secretion, testing several important growth parameters and developing a simple, semi-automated manufacturing process. Our long-term aim is to establish a simple plant manufacturing system for rabies antibodies that could be adopted locally in rabies endemic areas.
Clinical neuroimaging: Non-invasive measures of human brain function and their animal replacement potential
1 Aston University, Birmingham, UK
Many argue that understanding how the brain works cannot be rapidly progressed without invasive animal experiments, as other techniques lack the sensitivity or specificity to accurately evaluate brain cell function in health and disease.
Our research has focussed on using non-invasive brain imaging techniques in humans to create what we refer to as ‘virtual electrodes’ – these are computationally derived brain signals achieved by measuring the magnetic fields of the brain while we perform various tasks. These ‘virtual electrode’ signals provide us with an output that closely approximate to implanted electrodes. We have been validating such techniques for a number of years, including comparing with measures made invasively in humans and in animals. By combining this technique with other non-invasive neuroimaging techniques, we are building up a compelling portfolio of non-invasive alternatives to animal experiments.
Brain assemblies, not individual brain cells, characterise our cognitive abilities, and techniques studying ‘functional connectivity’ of the brain are beginning to provide vital insights into the way different brain regions work in concert, producing our perceptions and understanding of the world around us.
By working strategically with the larger science community, the pharmaceutical industry and politicians, and thus providing resource and training opportunities for young scientists, replacement in some research fields is a genuine prospect.
Developing an amoeba as an animal replacement model for Alzheimer’s disease research
Otto, G.P.1, Sharma, D.1, Hahn, C.1, Ludtmann, M.H.R.1, Killick, R.2, Williams, R.S.B.1
1 Royal Holloway University of London, Egham, UK
2 King’s College, London, UK
Alzheimer’s disease is a devastating disease that urgently requires the development of new treatments. Research into the disease is based upon animal models, thus alternative non-animal models must be developed. Alzheimer’s disease progress is controlled by a group of proteins called the γ-secretase complex that includes one of two Presenilin proteins (Psen1/Psen2). The complex acts by cleaving target proteins and in a poorly understood ‘scaffold’ role to bring together other proteins. We have developed Dictyostelium as a non-animal model to investigate the γ-secretase complex. Dictyostelium contains all of the core components of the complex, including two presenilin-related proteins (PsenA/PsenB). We show that deletion of both presenilin genes in Dictyostelium leads to a severe block in development, similar to that shown in animals, and that this effect is rescued by the human Psen1 protein. We show that this rescue does not require the two catalytic residues required for cleavage activity – suggesting that the human protein acts as a scaffold in Dictyostelium development. We also show that both human andDictyostelium presenilin proteins (and other complex proteins), localise to the endoplasmic reticulum (ER), similar to that shown in animals, suggesting that the γ-secretase complex forms in Dictyostelium. This research shows that Dictyostelium can be used as a useful non-animal model to understand the role of human γ-secretase complex, without the loss of animal life, to aid in our understanding of Alzheimer’s disease.