Gunnar Cedersund, Linköping University, 250 000 SEK
Project: Knowledge-driven drug development without animal experiments
In the last few years, mathematical modelling has displayed an unprecedented impact on drug development. For instance, recent specific approvals, and even more recent general guidelines, imply that computer simulations now can be used for regulatory approvals. Since model simulations are cheaper and faster, such approvals often end the use of test animals for that application.

We have modified such an approved type 1 diabetes model to also describe type 2 diabetes, a much more widespread and rapidly growing disease. Our award-winning model has been developed through numerous experimental/modelling iterations, and is already at use in several major pharmaceutical companies. However, this usage is still mostly done in traditional drug development pipelines, involving a series of cell and animal systems, which often display highly different outcomes.

To allow us to move to a radically new, knowledge-driven, more efficient, and increasingly human-centered pipeline, I will use uniquely informative data to do the four most important additions still needed in the model: i) storage and release of fatty acids, ii) specific drug-targets in heart and adipose tissue, iii) multi-level and long-term translations, iv) mapping from other animal-free systems.

With this grant, I could demonstrate in real ongoing drug development projects, done together with AstraZeneca, how our new knowledge-driven and increasingly animal-free drug development pipeline can work in a way that is economically beneficial for them, and which also helps to save both human and animal lives.

 

Pernilla Eliasson, Linköping University, 200 000 SEK
Project:  How do tendons heal? – and what happens when they don´t heal?
The general aim is to elucidate molecular mechanisms in tendon healing, with emphasis on the role of loading and new tendon tissue formation. Tendon ruptures are common and represent a clinical challenge.

The understanding of molecular processes during tendon repair is poor. Mechanical loading improves healing, but it is unclear how. Most likely, loading can act through two mechanisms to stimulate healing, microdamage and mechanotransduction. Mechanotransduction is the way a cell converts mechanical stimulus into a biochemical response. Microdamage is associated with influx of inflammatory cells, whereas little about mechanotransduction in healing tendons. It is difficult to separate between these two mechanisms in animals. Instead, I´m using a model for 3D cell culture studies and load these artificial tendons with fluid flow to study the effect the mechanotransduction component during loading.

Moreover, I have previously observed that tendon fibroblasts from some patients lack the ability to form tendon tissue, in vitro. We will study the differences between cells that can form tendon tissue with those that can´t. This might help to identify factors that are vital for tendon formation.

Furthermore, constructs will be cultured with serum containing high levels or normal levels of cholesterol to understand how this influences tendon tissue. This 3D model also makes it possible to study effects on mechanical strength of the tissue after different manipulations, e.g. drug treatment or changes in loading situations, in contrast to 2D cell culturing. Studies on mechanical strength is an important tool in orthopedic research and this has previously only been done in animal models. The goal with these studies is to increase the understanding on tendon repair, so treatment and rehabilitation protocols can be improved.

Anna Forsby, Stockholm University, 250 000 SEK
Project: Validation of mRNA markers for prediction of developmental neurotoxicity
Developmental neurotoxicity testing according to regulatory test guidelines requires a vast number of laboratory animals. The objectives of this study are to map transcription of the whole genome of the human neuroblastoma SH-SY5Y cell line and to validate neuronal mRNA markers that are identified to be significantly expressed during retinoic acid-induced differentiation for prediction of developmental neurotoxicity.

The SH-SY5Y cells are exposed during differentiation to chemicals that are known to induce developmental neurotoxicity and to chemicals that are considered to be non-neurotoxic. mRNA is isolated after 5-9 days of differentiation when differentiation markers are significantly expressed.

Our results show that acrylamide impair neuronal differentiation by interfering at least two signaling pathways, i.e. the retinoic acid and CREB pathways. Furthermore, acrylamide attenuates gene expression of brain derived neurotrophic factor BDNF. Data analysis of the effects of 32 other compounds on the expression of over 3300 genes is in progress.

The focus in this project will be on genes that are involved in neuronal function and development. Functional endpoints such as viability, calcium signaling, mitochondrial potential and neurite length can be associated with transcriptomic markers. The identified genes of interest as biomarkers for developmental neurotoxicity will be validated by RNA sequencing after exposure with positive and negative model compounds.

 

Robert Fredriksson, Uppsala University, 200 000 SEK
Project: Development of a novel method for an animal free method for Botulinum toxin testing
Botulinum toxin type A (BoNT/A) is one of the most potent neurotoxins known and its mechanism of action is to cleave SNAP-25 in neurons to make them unable to signal. BoNT/A is used in the beauty industry to reduce facial wrinkles, but also in medicine to treat various conditions such as chronic pain and excessive sweating.

Accurate potency determination is vital for its safe and effective use as a commercial substance and potency measurements must be assessed for each batch produced. Today potency measurements are done to a large extent using LD50 assays in mice.

There are a few patented protocols for animal free BoNT/A potency measurements, but these seem to be used to a limited extent in practice. The challenges have been to identify a cell-line with high enough sensitivity towards the BoNT/A and a strong readout of the BoNT/A effect, to enable development of a validated method for release of drugs.

We have chosen to use the most direct readout, to measure amount of cleaved SNAP-25. We have developed a protocol which enables us to from human embryonic stem cells differentiate cholinergic motor neurons with very high sensitivity towards BoNT/A. We have setup a provincial method based on antibodies which can give us a dose response, and we are now trying to replace the antibodies with synthetic antibodies (affibodies and aptameres).

Within this project we will develop a method to detect SNAP-25 using synthetic antibodies and refine the cell-culturing methods, to shorten down the culturing time and introduce points where cells can be frozen and stored. The main outcome we are aiming for is to have a Proof of Concept protocol including cell-culture, BoNT/A treatment and detection of cleaved SNAP-25 including preliminary determination of internal assay parameters such as sensitivity, variability and reproducability. This can be used to show that there is a possibility to establish a cell based assay for detection of BoNT/A potency determination.

 

Marianne Jansson, Lund University, 200 000 SEK
Project: Djurfri lymfoid infektionsmodell; svaret på varför HIV-2 är mindre aggressivt och hämmar HIV-1?
The HIV-1 epidemic is still spreading. Thus, the need for a vaccine or cure is urgent.

To identify virus-host mechanisms that could be targeted for cure strategies a fraction of HIV-1 infected individuals with spontaneous virus control, so called elite controllers, have been studied. However, only 0.15% of HIV-1 infected individuals are elite controllers. In contrast, the proportion of elite controllers among those infected with HIV-2, a benign form of HIV, has been estimated to 37%. Therefore, HIV-2 infection may represent a better model for finding cure strategies.

In an epidemic study we showed that HIV-2 also delays HIV-1 AIDS development with 90% in HIV-1/HIV-2 dually infected. Furthermore, the blunting effect that HIV-2 has on HIV-1 progression was observed to be an early HIV-1 infection event.

We now want to identify mechanisms behind HIV-1 and HIV-2 differences, and how HIV-2 suppresses HIV-1. Since HIV primarily replicates in lymphoid tissue, infection experiments using bloodderived lymphoid cells do not provide a clinically relevant model. Humanized mouse and non-human primate models are therefore increasingly used to study early infection events taking place in lymphoid tissue.

To avoid use of animal models, we here aim to establish a novel model where infection of lymphoid tissue, from human tonsils, is used to dissect early events of HIV-1, HIV-2 and dual infections. The tonsil tissue will be infected with HIV-1 and HIV-2, individually and dually, and gene specific chimeric viruses will be constructed for deciphering virus-specific determinants of replicative capacity and host interactions.

The replication of the viruses, as well as the expression of genes and proteins in different cell populations, infected and bystander, will be analyzed using advanced bioinformatics including network analysis and modeling. By this we hope to contribute to the development of animal free infection models and strategies to prevent AIDS development in HIV-1 infection.

 

Maria Karlgren, Uppsala University, 100 000 SEK
Project: Human cell based models for accurate assessments of brain drug uptake
Poor predictions of CNS drug exposure is a major problem in CNS drug development and is primarily the result of relying on animal-based models although major species differences are well-known for the blood-brain barrier (BBB) and the expression of drug transporters.

Here, we will solve this problem by developing predictive models taking the drug transporters expressed in human BBB into account and thereby accurately quantify drug delivery to human brain. This will be done by combining in vitro kinetics from mechanistic transporter models and human BBB in vitro models with quantitative proteomics and physiological parameters to predict the human in vivo situation.

The project will be performed at Uppsala University in close collaboration with UDOPP (part of DDDp/SciLife Lab) giving a unique opportunity for validation in real-life drug development projects. With our model drug candidates with predictable and favorable human BBB properties will be identified and selected early in drug discovery/development. As a result an initial reduction of animal PK experiments of 50-60%, with further reduction potential, is estimated. Thereby providing the refined tools needed to substantially reduce and, in a longer perspective, replace animal BBB studies throughout the drug discovery/development process.

 

Pekka Kohonen, Karolinska Institute, 200 000 SEK
Project: Deepened mechanistic validation of toxicity pathway functionality in a patented analysis tool
The Predictive Toxicogenomics Space (PTGS) tool [1], has been developed for application to dose-response analysis with benchmark dosing technology. Importantly, the work from last year shows that the PTGS-based analysis provides 10-100 times higher sensitivity that common assays for diverse genomic and cellular toxicity endpoints.

The proposed continuation incorporates independent validation and completion of the project into a novel, sensitive and mechanism-driven benchmark dosing tools. Specifically, project funding is now applied for validating the workflow outlined in the application with a TempO-Seq high-throughput transcriptomics study utilizing 81 test compounds on in vitro data generated with the US EPA S1500+ platform that has been optimized and developed for high-throughput transcriptomics prediction of drug-induced liver injury.

The dataset has been generated by our collaborators in the area of pharmaceutical industry. Data was generated using HepG2 human liver cells in a traditional cell culture setting. Another test set includes the use of cardiomyocytes which tests the potential of PTGS to predict toxicity of organs other than liver, thus potentially widening its application domain.

Overall, the project is set to generate needed proof-of-concept for a widely applicable solution towards toxicity prediction of environmental agents. This goal covers creating standard operating procedures for a real product-based applicable testing tool. Agreeing fully with the currently Forska Utan Djurförsök-supported work, the ultimate goal is to act in a translational manner relative societal needs of improving environmental health and safety with precise and costeffective in vitro and in silico assays that fully replaces animal testing. The proposed analyses in this application build fully on the previously funded proposal and are aimed to complete the tool development for publication in a high impact journal.

 

Per Malmberg, Chalmers, 200 000 SEK
Project: Chemical imaging for skin permeation studies – an analytical approach to replace animal experiments
Every day our skin is exposed to various metals and chemicals when it meets everything from tools and electronics to cosmetic products. Exposure to metals and chemicals can cause contact allergy and can also cause long-term health effects such as increased cancer risk. In occupational health, healthcare and legislation, it is therefore of great interest to be able to follow where nanoparticles, metals, pharmaceuticals and chemicals end up in the skin, how large the absorption is and how the cells in the skin are affected.

Today the use of experimental animals or isolated animal skin to study the absorption of molecules in the skin is widespread. Only within the REACH program more than 2 million animals have been sacrificed so far. Therefore, there is a critical need to develop scientifically valid alternatives for safety testing of pharmaceuticals and chemicals. In this interdisciplinary project we want to develop an in vivo method for studying skin uptake using chemical imaging. The method uses donated human skin where we follow the exposure of chemicals and allergens in a diffusion chamber. Because our method is independent of antibodies or probes, it can be used to study the penetration of almost all compounds and used to establish hygienic limits in combination exposure. It can also provide information on how body-like molecules, such as peptides and lipids, are affected by the exposure.

Our goal is to offer a more flexible and sensitive analysis method that can replace and/or supplement existing methods in the OECD guidelines for skin adsorption (no. 427 and no. 428). During the project period, we will validate and establish this approach and demonstrate the benefits compared to current in vivo and in vitro methods. The method will provide a unique way to look at the skin and provide a cost-effective animal-free alternative that can be used for occupational studies, basic research and product testing.

Mats Nilsson, Stockholm University, 200 000 SEK
Project: The Design of a Glioblastoma Organoid-on-a-Chip for Drug Testing
Glioblastoma multiforme (GBM), in addition to being the most common type of malignant primary brain tumor, is one of the tumors that has seen the least improvement in treatment options in the last half-century. The existing tools we have (mouse models, cell culture models) for testing efficacy are insufficient and inefficient. New tools are needed. Specifically, a tool that tests a drugs ability to pass the BBB and reach the tumor, that recapitulates the GBM tumoral environment and complexity, while allowing efficacy to be assessed in the context of biomarkers.

Toward that end, the project proposal is to develop an in vitro glioblastoma organoid on-a-chip that can be integrated into the drug discovery systems of tomorrow, and used to assess preclinical effectiveness of drugs on glioblastomas.

A wide range of expertise is required to succeed. Toward that goal, we have gathered leading researchers in Uppsala and Stockholm spanning the fields of cell biology, chemistry, engineering and biotechnology to develop the key components of the project.

Stina Oredsson, Lund University, 200 000 SEK
Project: Novel 3D Cell Culturing Methods in Cancer Research
The main final goal of the present project is to create a human 3D tumour outside of the body that is totally animal product free and that can be used at a reasonable cost. The purpose is to use this model in pre-clinical testing of novel potential chemotherapeutic drugs that not only target the cancer cells but also target normal cells as well as secreted molecules in the tumour environment that provide protection to toxic insult to the cancer cells.

To enable this goal, there are two major issues related to animal-derived products to be solved: the extracellular matrix and the medium. As an artificial 3D collagen network, we use electrospun polycaprolactone (3D) fibres. We have shown that this network is biocompatible and that normal cells and cancer cells thrive in the 3D fiber mesh. The growth patterna of Co-cultures of cancer cells and fibroblasts are treated with experimental compounds and chemotherapeutic drugs.

The preliminary data show that fibroblasts protect the cancer cells from toxic insult which is in corroboration with present knowledge and thus contributes to validation of our model. Preliminary data of adding NK cells and dendritic cells demonstrate a role for immune cells in the tumour environment. We will further validate the model e.g. treating with immune checkpoint inhibitors.

So far, we have used donor heard horse serumsupplemented medium as a means to eliminate fetal bovine serum but we are now taking this to the next level by developing a totally defined medium. Thus, a number of cell lines are being adapted to totally defined medium and in preliminary experiments we have found that these cells also thrive in the 3D PCL fiber mesh. Thus, we are not far from our goal of human 3D tumour outside of the body that is totally animal product free but much validation work is needed.

 

Lena Palmberg, IMM, Karolinska Institute, 200 000 SEK
Project: Development of lung models with multiple cell types for treatment
strategies of chronic lung diseases
Air pollutions caused 9 million premature deaths in 2015. Chronic obstructive pulmonary disease (COPD) can be caused by air pollution and is the third leading cause of death worldwide. By using multi-cellular lung mucosa models with human primary bronchial epithelial cells, combined with immune effector cells (macrophages), air pollution related health hazards will be assessed. In the research team pre-clinical and clinical experts as well as scientist from the pharma industry warrant a comprehensive assessment of scientific methodology and data interpretation. The goal including evaluation of different treatment strategies.

The studies explore the interaction of exposure, therapeutic effects, innate immunity, protease/antiprotease balance and oxidative stress as well as the interactions of various cell types. This knowledge can be utilized to develop preventive measures and treatment strategies for patients with COPD and chronic bronchitis. Validation of the models enables us to develop a systematic in vitro-testing strategy during drug development in order to reduce the requirement for animal inhalation studies.

Olena Prykhodko, Lund University, 200 000 SEK
Project: 3D-tissue in vitro model of human intestine
Intestinal health and integrity is a new target for disease prevention and therapy. A novel, state-of–the–art, in vitro reconstructed 3D human intestine is engaged as a model for testing of food, food-contact materials and microbial components. e.g. new proteins, food additives, new probiotics or packaging materials, yet taking in account differences in processing and formulation. This includes evaluation of tolerable amounts, direct uptake to the blood circulation and estimation of inflammatory response to the test-compounds. The results obtained systematically throughout assessments such as risk identification, dose-response assessment, and uptake assessment, can be further used for machine-learning calculations of probability characterizations, for instance categorization of foodassociated risks on human health, approximating exposure periods and dose-responses.

While there are other in vitro technologies in Sweden that are used as alternatives to animal models to study absorption functions, such as human cancer cell as Caco-2 which is a monolayer and presented just by one type of cells, the new Epi-Intestinal Model, is far superior due to the ability to re-construct human intestinal cells from healthy donors in vitro as a 3D structure, containing villi with brush borders and also mucous secretory cells.

Ulf Yrlid, Göteborg University, 200 000 SEK
Project: Human intestinal organoids for assessment of fucose-based blockers of enterotoxin-mediated disease
Diarrheal diseases kill 2.5 million children (under the age of five) annually in underdeveloped countries. Two of the main disease-causing pathogens are enterotoxigenic E. coli (ETEC) and Vibrio cholerae (V. cholerae). There is no effective vaccine against ETEC and the vaccine against V. cholerae has 65% efficacy in endemic areas. In order to cause disease these bacteria colonize the gut and secrete toxins that after binding to the epithelial cells are internalized resulting in fluid secretion that if untreated can be lethal.

Recently we published a finding that changes the dogma for how Cholera toxin (CT) binds to human intestinal cells. Whereas the glycolipid GM1 was the only previously known functional receptor for CT, we identified that CT, like ETEC LT toxin, can also bind to fucosylated glycoproteins. Our follow-up experiments now show that blocking the fucosylated structures, that GM1 does not carry, efficiently inhibits binding of toxins to epithelial cells.

The aim of this project is to assess if new blocking reagents carrying fucoses can bind to the toxins, stop internalization by intestinal epithelial cells and thereby block the fluid secretion. Intoxication has for many years been studied using the infant mouse model. This is because adult mice are not colonized with ETEC and V. cholera. We wish instead to use organoids derived from human small intestinal biopsies and thereby completely replace experimental animals. The organoids will after expansion be seeded into wells for differentiation into epithelial cell monolayers. This allows measurement of resistance/ion transport a direct correlate of fluid secretion following apical exposure to toxins pretreated or not with fucosecontaining blocking reagents.

The extent of fucosylation differs between species and also among individuals. Hence establishing organoids from multiple donors will not only ensure that blocking of the relevant glycans is measured but that the results represent the human population.