Speaker Abstracts

Characterising an animal origin-free in vitro model of the human intestinal epithelium for drug permeability studies

Argent, R.H. (1), Falcone, F.H. (1), Stolinik, S. (1), and Garnett, M.C. (1)

(1) School of Pharmacy, University of Nottingham, Nottingham

The permeability of gut epithelium to drugs and nanoparticles is critical to the pharmaceutical industry, yet pre-clinical testing is performed using inaccurate models (in vivo, ex vivo, in vitro). The standard Caco-2 in vitro cell model of intestinal uptake is conducted in the presence of animal-derived serum and lacks key features of human intestinal epithelium. By making the model more realistic we aim to improve drug permeability measurements to more closely reflect human clinical data and ultimately reduce the number of animals used by the pharmaceutical industry for this purpose.

A basal lamina, deposited by Calu-3 cells cultured in serum-free conditions before decellularisation, was required for the culture of Caco-2 cells in serum-free conditions in transwells. Proteins known to be incorporated into native basal lamina, including laminin, collagen and nidogen, were demonstrated to be present. Additionally, atomic force microscopy confirmed the presence of a structured proteinaceous layer and showed the presence of pore-like features.
Caco-2 cells cultured on the basal lamina became polarised with low trans-epithelial resistance, akin to that of the human intestine, formed better organised ZO1-positive tight junctions, and possessed more prominent microvilli at their apical surface. The permeability of polarised Caco-2 cells to FITC-dextrans was also decreased, demonstrating that an important additional barrier to transport normally present in gut epithelium was functionally restored.
In summary, we have established a more phenotypically realistic model of the human intestine and are working on genotypic and functional characterisation of permeability to demonstrate greater relevance for permeability testing. 

An in vitro culture system to study the development and maintenance of human IgE-secreting plasma cells.

Ramadani, F. and Gould, H.

Asthma and Allergy, Randall Division of Cell & Molecular Biophysics, King's College London

The UK has an extremely high prevalence of allergic disease, affecting up to 30% of adults and 40% of children. Asthma care alone costs the NHS £1 billion p.a.
Allergen-specific IgE antibodies, secreted by plasma cells (PCs), play a fundamental role in the pathogenesis of allergic disease, including asthma. These PCs are maintained in specialised survival niches and resist conventional immunosuppressive treatments. Gaining new understanding of how the IgE+PCs are developed and maintained may lead to a major breakthrough in the treatment of allergic diseases.
Most recent work has relied on mouse studies. However, there are fundamental differences between human and mouse IgE biology; so it is critical to investigate IgE-related biology using human samples. One important difference is that humans express a short and a long form of membrane IgE. The long form is unique to human IgE and - significantly - has been previously associated with cell survival.
Hitherto, study of the IgE+PCs in humans has been hampered by low yields of human IgE+ cells in in vitro models and their rarity in vivo. Recently we have established an in vitro tonsil B cell culture system that generates high yields of human IgE+ cells, which enabled us, for the first time, to characterise their differentiation pathway into PCs.
This in vitro culture system provides a unique platform for gaining new understanding of what makes IgE+PCs develop and survive. This will make possible new strategies for suppressing or controlling these cells, which may provide effective treatments for allergic diseases. 

Utilising live human peritumoural brain tissue to understand tumour associated seizures.

Cunningham, M.O.

Institute of Neuroscience, The Medical School, Framlington Place, Newcastle upon Tyne, NE2 4HH, UK

Brain tumours present with seizures as a major symptom in 30% of all cases. Seizures complicate the management of patients with glioma and contribute to significant morbidity. Tumour associated seizures also demonstrate significant resistance to treatment with anti-epileptic drugs. Animals models have been developed to address questions surrounding epileptogenesis due to gliomas. There are issues surrounding the validity of these animal models and how adequately they recapitulate the human condition. For example, current models match the clinical course of high grade gliomas but not that of low grade gliomas, which clinically are more likely to be epileptogenic at presentation. Moreover, studying seizures towards the end stage of the animal life is complicated by considerable welfare concerns about animals with large intracranial brain tumours. In addition to these concerns, a scientific limitation is the lack of availability of animal models of low grade glioma. To overcome this, we access human brain tissue obtained from patients with seizures and low grade gliomas undergoing neurosurgery. We examine tissue around the tumour, in the so-called ‘peritumoural’ region. This region is of great interest as it contains the interface between ‘normal’ neurons and invading glioma cells. Increasing evidence points to this region as the area from which seizures arise. As such, understanding the processes that occur in this area will provide a better insight into the mechanisms that underlie tumour associated epilepsy. I will present data that will highlight the validity of using live human peritumoural tissue for scientific studies of tumour associated epilepsy.

Development of an ex-vivo human skin model for impaired wound healing in diabetes: initial studies including chemokine expression

Bevan D (1), Keppie S (1), Moncrieff M (2) and Gavrilovic J (1)

(1) School of Biological Sciences, University of East Anglia, (2) Norfolk and Norwich University Hospital

Diabetic foot ulcers (DFUs) remain a major complication of diabetes with 135 amputations a week performed in England. Understanding of mechanisms underpinning poor healing of DFUs is urgently required. Here a human ex-vivo human skin wound healing assay has been employed to explore wound healing under high glucose conditions, mimicking those levels found in people with diabetes. Skin wounds were made in biopsies from donated surplus abdominal skin tissue from reconstructive surgery. Wounded and parallel un-wounded explants were cultured at the air-liquid interface (mimicking the in vivo situation) and exposed normal or high levels of glucose in the presence or absence of recombinant human epidermal growth factor, known to promote aspects of skin wound healing. Samples were then analysed histologically for initiation of re-epithelialisation of wounds and for expression of candidate genes. A trend towards a reduction in wound healing in high glucose conditions and a clear effect on chemokine repression were observed, possibly connected to the inadequate recruitment of macrophages seen in early stage diabetic wounds. Skin wound healing in relation to diabetes has generally been performed in mouse or rat models with skin wounds made on the dorsal skin of animals under anaesthetic, a moderately severe procedure. In addition ex-vivo skin wounds have been made from porcine skin biopsies in which keratinocytes or fibroblasts from people with diabetes have been transplanted. Here, this novel ex-vivo human skin wound model under high glucose conditions paves the way for further refinement incorporating the inflammatory aspect of diabetes through cytokine inclusion.

Identification of suitable culture medium for the development of an animal-free oral mucosa model

White, C. L, Addison, O (1,2), Landini, G, Cooper, P. R and Shelton, R. M

(1) School of Dentistry, University of Birmingham; (2) Oral Health Clinical & Translational Research Centre, University of Alberta

Tissue engineering recreates complex 3D tissue within an in vitro environment. Animal-derived products such as foetal calf/bovine serum are commonly used to support the engineered tissue. However, their use has many disadvantages, including potential disease transmission, contamination, inter-batch variability and ethical concerns. This study aimed to identify a suitable animal-free culture medium that can be used to develop long-term 3D cultures of human oral mucosa. This model will then be used to analyse mucosal soft tissue interactions with dental implant surfaces.
Primary human gingival fibroblasts (pHGFs) and H400 cells (human keratinocyte cell lines isolated from an oral squamous cell carcinoma) were maintained in three commercially available keratinocyte chemically defined culture media; Epilife® supplemented with S7, CnT-04 and KGM®-CD and DMEM supplemented with 10% human serum (used as a comparator). Doubling-time and viability of cultures were analyzed for 21 days from growth curves established using trypan blue dye exclusion and Alamar Blue assays. Gene expression analyses of cytokeratins-1, -5, -10, -14, and -19, vimentin, and E-cadherin were performed using semi-quantitative RT-PCR.
KGM®-CD exhibited highest growth rates for both cell types during the experimental period. There was minimal difference between cytokeratin expression levels of cultures grown in KGM®-CD and the control, and these cultures expressed similar levels of vimentin and E-cadherin.
The animal-free culture medium KGM®-CD has the potential to support development of a 3D model of oral mucosa. The ability to construct oral mucosa without using animal products would enhance the relevance of tissue engineering research and reduce animal use.

Multi-modal integration of brain function measurements: Compelling evidence for alternatives to animal experimentation

Furlong, PL., Witton, C

Aston Brain Centre, Aston University, Birmingham, United Kingdom

A consistent argument used to justify animal brain research in-vivo, particularly in primates, is that alternative non-invasive techniques lack the ability to study activity at the individual neuronal level, which only invasive techniques can provide.
A counter argument is that in the study of behaviour, and especially cognition, that measurements that sample at the individual neuron, or even multi-unit activity, do not accurately reflect the neuronal networks involved sufficiently to understand the brain processes that underlie typical or atypical human brain development.
Over the last ten years, our research has demonstrated the ability of Magnetoencepholography (MEG) to accurately reflect Local Field Potentials in the human brain, and that only by studying whole brain function can we unravel its complexity. We have validated our methods through direct comparison with animal data evidence and through our many studies in patients undergoing neurosurgery.
Through our clinical activity we are able to obtain living human brain tissue from patients who we have studied pre-surgically, and this is yielding novel and unique insights into pathological brain processing.
We now combine a number of non-invasive brain measurement techniques including MEG, high field MRI and Trans-Cranial Magnetic Stimulation, and use novel developments in functional connectivity analysis, to characterise brain function. Multi-modal techniques are making important impacts on fundamental neuroscience, and provide compelling evidence for the value of these techniques as alternatives to inferior animal models.

The future of Thiel embalmed human cadavers as an alternative to training and testing using animals

H Mcleod (1), R Duncan (1), T Wilkinson (2), G Mcleod (3) G Corner (4) S Matthew (1), A Melter (5) G Houston(1)

(1) Cardiovascular Diseases and Diabetes Research Division, School of Medicine, University of Dundee; (2) Centre for Human Anatomy and Identification, School of Science and Engineering, University of Dundee; (3) Division of Neuroscience, School of Medicine, University of Dundee; (4) School of Science and Engineering, University of Dundee; (5) Institute for Medical Science and Technology, School of Medicine, University of Dundee.

Thiel embalmed human cadavers have the physical properties, accurate anatomical features and preservation to make them an ideal model for teaching in medical sciences.  From the initial pilot work initiated in 2009 by CAHID (Centre for Human Anatomy and Identification, University of Dundee), the many benefits of Thiel embalming have been investigated.

Vascular patency was demonstrated by Andreas Melzer, IMSaT UOD(2010; MRC and FP7 funding), using extracorporeal flow to deliver contrast throughout the vascular tree, and visualising this with X-ray. In 2012, Dr Hadwen Trust funded a novel project developing methods to utilise Thiel embalmed human cadavers as an alternative model to animals for preclinical device testing, and for training medical professionals. Subsequent multimodal imaging technique development, together with advanced perfusion methods in human cadavers, has opened up further potential fields and applications. The success of the Hadwen project was ultimately measurable by academic, clinical and industrial demonstration of Thiel embalmed cadavers as a potentially superior alternative to the large mammals commonly used in this field.

In the new phase of our DHT funded project, our understanding, characterisation and development of applications using Thiel continues to grow. The need to consolidate with other research groups across the UOD to develop a network to raise the clinical, academic, industrial and public profile of Thiel research is recognised.

This aim is realised by the formation of the Thiel Forum, a multidisciplinary group, formed to develop, expand and disseminate collaboratively the scientific merit of using Thiel embalmed human cadavers in this novel context.

Multimodality imaging of a perfused human renal cadaveric model as an alternative to animal testing for pre-clinical device testing.

Duncan, R. (1), McLeod, H. (1), Xiao, X. (1), Jagpal, B. (3), Wilkinson, T. (2), Melzer, A. (1), Houston, G. (1)

(1) Institute for Medical Science and Technology, University of Dundee; (2) Centre for Anatomy and Human Identification, University of Dundee; (3) Clinical Research Imaging Facility, Ninewells Hospital, Dundee.

Thiel cadavers offer a viable alternative to replace animals in clinical research. The aim of this study is to provide proof of concept that a perfused cadaveric renal model can be successfully imaged using four separate modalities for pre-clinical device testing. Utilising a Thiel-embalmed human cadaver with extracorporeal flow, this study imaged the perfused renal model with ultrasound, x-ray, CT and MR.

Contrast enhanced ultrasound scanning using Acuson S2000 (Siemens) identified the left kidney in B-mode and with Colour Doppler indicating the presence and direction of flow. The left renal artery and kidney capillary bed were imaged with contrast using OEC 9900 Elite C-Arm (GE, USA). A contrast enhanced abdominal MRI using 1.5T GE Scanner imaged the left kidney with a 3DFIESTA pulse sequence. Scanning with a PET/CT Scanner (Biograph mCT-128, Siemens) produced contrast enhanced CT data of the abdominal aorta showing the left renal, suprarenal and inferior mesenteric arteries.
Perfused Thiel cadaver models offer a valid model that can be imaged successfully in four clinical imaging modalities commonly used for medical device development and testing. This novel application is pivotal for the pre-clinical testing of EU-funded focused ultrasound therapy development presenting a viable alternative to animal testing for clinical research.

A more physiologically relevant set of culture conditions for EndoC βh1 human beta cells?

Nicola Jeffery (1), Sarah Richardson (2) and Lorna W. Harries (1)

(1) RNA-mediated mechanisms of Disease, Institute of Biomedical and Clinical Sciences, University of Exeter Medical School, University of Exeter, Devon, UK.; (2) Islet Biology group, Institute of Biomedical and Clinical Sciences, University of Exeter Medical School, University of Exeter, Devon, UK.

Recent studies in mice have found apoptosis alone does not account for loss of beta cell mass in type 2 diabetes, and suggest a role for changes in their differentiation status as an alternative. However, anatomical and physiological differences between rodents and man mean these findings require investigation in humans.

Here, we acclimatised the human EndoC βH1 beta cell line to a normal human cellular microenvironment, in vitro, and measured effects on the transcriptional and immunofluorescence profile of genes involved with beta cell fate.

Characterisation by immunofluorescent microscopy demonstrated the cells contain and secrete insulin and express mature beta cell markers NEUROD1 and PDX1. We found differences in expression of 5/30 genes when cultured in a human cellular microenvironment. Increases in expression of GLUT2, PAX4 and MYC and a decrease in the alpha cell marker ARX were noted, but no differences in expression of mature beta cell markers.

This study presents a novel culture methodology for the EndoC βH1 cell line and describes the most comprehensive gene expression profile to date of genes involved in beta cell function and cell fate in this cell line. 

Phosphoproteomic analysis identifies STAT1 as a novel target in meningioma

Sara Ferluga (1), Jemma Dunn (1), Daniele Baiz (1), David Hilton(2), Kayleigh Bassiri (1), Vikram Sharma(3), Claire Adams (1), Edwin Lasonder (3) and Oliver Hanemann (1)

(1) Institute of translational and stratified medicine, Plymouth University Peninsula Schools of Medicine and Dentistry; (2) Clinical Neurobiology, Peninsula Medical School, Plymouth University; (3) School of Biomedical and Healthcare Sciences, Plymouth University

Meningiomas are the most common primary tumours of the central nervous system in adults and despite the majority of them displaying benign features they can cause mild to severe morbidity. The current therapeutic approach is complete resection commonly with adjunct radiotherapy. No pharmaceutical treatment is available and surgical resection is not always possible, so new therapeutic options are needed. In the context of molecular targeted therapy, we aimed to identify novel targets/biomarkers by analysing the proteome of different grade meningiomas using an in vitro approach.
Frozen tumour specimens as well as meningioma primary cells were analysed by mass spectrometry to decipher the global- and the phospho-proteome. Comparative studies were performed to identify aberrantly overexpressed proteins and dysregulated pathways compared to human meningeal tissue and human meningeal cells respectively. Among the phosphoproteins found significantly upregulated in meningioma vs. normal controls we identified STAT1. Validation studies confirmed that the total amount of the protein was overexpressed about four times compared to normal across all the meningiomas analysed. Additionally, both phosphorylation sites (Y701 and S727) on STAT1 were aberrantly activated only in meningiomas but not in normal controls. Immunohistochemical studies showed a marked upregulation of phosphorylated STAT1 on grade II and III meningiomas, in agreement with expression studies, suggesting a direct correlation with tumour progression. STAT1 gene silencing resulted in a significant reduction of cellular proliferation showed as a decrease in Ki67-positive cells.
In conclusion, by a proteomic approach we identified STAT1 as a novel potential biomarker and/or therapeutic target in meningiomas.

'All human’ in vitro blood brain barrier (BBB) modelling: implications for CNS nanoparticle drug delivery studies

Maherally, Z (1)., Kumar, S.A (2)., Philpott R (3), Alder J.E (2)., Barbu E (1), Fillmore, H.L (1)., and Pilkington, G.J (1)

(1) Brain Tumour Research Centre, University of Portsmouth; (2) University of Central Lancashire; (3) University of Southampton.

With increased interest in pre-clinical, high-throughput, in vitro screening of CNS therapeutics, the use of reliable in vitro models for compounds crossing the blood brain barrier (BBB) without damage to its function could provide significant cost saving. We have, therefore, developed an ‘all human’ model better reflecting in vivo conditions to assess nanoparticle (NP) delivery. A 2- dimensional BBB model was developed by incorporating human brain endothelial cells (hCMEC/D3), human cortical astrocytes (SC-1800), human brain vascular pericytes (HBVP) and human basal lamina protein (Perlecan), cultured with human serum as mono-, co- and tri- cultures. Trans-endothelial electrical resistance (TEER) was measured using an Electric Cell-Substrate Impedance Sensing (ECIS) system. In parallel, a 3D BBB model was assembled using the same cell lines and conditions and TEER recorded using a voltohmmeter, EVOMTM. Expression of occludin (tight junction protein) was evaluated by Western blotting (WB) and immunocytochemistry (ICC). Barrier integrity following NP addition was monitored. In both paradigms, co-culturing hCMEC/D3 and SC-1800 resulted in higher resistance compared to endothelial cells alone. WB and ICC correlated with resistance. Highest expression of occludin occurred in co-cultures. Post addition of kytozyme and lipid-based NPs resulted in decreased resistance followed by full recovery of the barrier within 12 hours confirming successful passage through the BBB model. Addition of Evans Blue did not effect TEER suggesting a non-leaky barrier. This human in vitro model better mimics the in vivo human situation, supporting the case for increased pre-clinical testing in vitro and obviating the need for pre-clinical animal testing. 

Chronic Myeloid Leukaemia tyrosine kinase inhibitor response classification by quantitative RT-PCR

Helen Wheadon (1), Heather Morrison (1), Hothri Moka (1), Parto Toofan (1), Stephen OBrien (2), Mhairi Copland (1)

(1) Paul OGorman Leukaemia Research Centre, Institute of Cancer Sciences, University of Glasgow, Glasgow, UK; (2) Northern Institute of Cancer Research, University of Newcastle, Newcastle, UK

Predicting response to tyrosine kinase inhibitor (TKI) therapy is one of the unmet clinical challenges in chronic myeloid leukaemia (CML).  This study aimed to identify a panel of genes as valid predictive biomarkers to TKI therapy response. Using extensive microarray data from CML patients, 270 genes deregulated either in the stem cell population or during disease progression were selected and stratified. These genes were then validated against 60 samples from the SPIRIT 2 clinical trial, with 18 months follow up data regarding molecular response to TKI treatment. SimpleLogistic classifier analysis was applied to stratify patients into good/intermediate/poor response to TKI and to identify the gene signature with significant predictive value. These data identified the DNA repair genes as having significant predictive value, in particular the minichromosome maintenance protein (MCM) family of genes. We are currently investigating the efficacy of the antibiotic heloquinomycin, a potent therapeutic inhibitor of MCM as a future therapy using CML patient samples and patient derived induced pluripotent stem cells as an alternative to animal (mouse) testing. Future development of a reliable test to quickly identify patients who are likely to be unresponsive, will also enable administration of the appropriate TKI/combination therapy following diagnosis thus substantially improving patient outcome. System biology approaches have great potential to identify novel pathways involved in diseases, and to use drugs already clinically approved which target these pathways, mitigating the need for additional animal testing. This approach could substantially reduce the reliance on animal experimentation in the leukaemia field in the future.

Developing in vitro model systems with demonstrable relevance to human tumour biology and therapeutic development.

Biddle, A. (1), Gammon, L. (1), Mackenzie, I.C. (1).

(1) Blizard Institute, Barts and The London School of Medicine and Dentistry, Queen Mary University of London, UK

The aim of our research is to develop human cell culture models that have demonstrable relevance to human cancer and can replace mouse models. In our recent work, we have identified heterogeneous cancer stem cell sub-populations in human oral cancer cell lines growing in 2D culture, fresh tumour specimens transplanted into 2D culture, and in direct analysis of fresh tumour specimens. The cancer stem cell sub-populations differ in both their invasive ability and their resistance to chemotherapeutic drugs. The demonstration that cancer stem cell sub-populations can be modelled using cell lines in vitro provides an opportunity for the development of novel therapeutic strategies targeting cancer stem cell sub-populations that may otherwise survive therapeutic intervention and drive tumour recurrence. However, there is currently some uncertainty over the relevance of cell culture models to human cancer, and this has driven an increase in the use of mouse models. In the last three years there have been over 1,000 publications reporting the use of mouse xenograft models to study the properties of cancer stem cells, with the use of many thousands of animals. These mouse models are slow, expensive, and often provide a paucity of useful data. New models that combine physiological relevance with the experimental tractability of in vitro models are therefore required. Our current aim is to develop an in vitro 3D model system that can incorporate both human cancer cell lines and explanted fresh tumour specimens, and can be shown to accurately predict human tumour biology and therapeutic response.