Interview U.S. Pharmaceutical Industry Newsletter
Building High-Throughput Organs-on-Chips
Jos Joore, Ph.D., Co-founder and Chief Business Officer, Mimetas B.V.
An interview by Chris Vickrey and Asuka Ishikawa for U.S. Pharmaceutical Industry Newsletter, published in Japanese
Based in the Dutch City of Leiden, Mimetas is an early leader in the emerging field of organs-on-chips. We spoke with Dr. Jos Joore, the company’s co-founder and chief business officer, about the advantages of organs-on-chips, the company’s OrganoPlate platform, and the ways in which Mimetas collaborates with pharma companies.
Please describe the background to the establishment of Mimetas.
Joore: About four years ago, when Paul Vulto, who invented the phaseguide microfluidics technology underlying our products, returned from a postdoc post, we got together and started to think of applications for the technology. We knew that 3D cell culture was an area that was rapidly developing, and there was a great need for high throughput and ease of use, without sacrificing the complexity or physiological value of 3D cell culture models. We decided to develop the microfluidic technology that Paul invented into an organ-on-a-chip, and that was the start of Mimetas.
What are some of the advantages of organ on a chip technology over animal models?
Joore: The first is that we can work with human cells, either primary cells or iPS-derived cells, in an environment that is fully physiologically relevant. That means that we can actually create models that are defined in their function. For animal experiments, you have to take into consideration the entire animal, which is a complex system, whereas we can do a very defined experiment using human cells.
For toxicology, for example, we know that rat models are not very predictive of how a compound will function in humans. With organs-on-chips, it is fairly easy to create a full panel of human liver cells, for example, that cover every racial background in the world. So the technology opens up possibilities that go beyond what can be accomplished with animal models. On the other hand, whole animal models can detect systemic toxicity, where one organ produces something that targets another organ, and there could be side effects down the line. So we are not about to fully replace animal models. It is still an emerging technology, but one with great promise.
Please explain how your OrganoPlates work, and how you typically work with pharma companies.
Joore: Our OrganoPlates use a fully compatible, 384 well plate to access microfluidics, and that is unique. The 384 well plate is actually the user interface to do 3D cell cultures, so the product is fully compatible with any liquid handling imaging equipment that pharma companies or academic labs already have. They do not need any specialized equipment to run hundreds, thousands, or ten thousands of experiments. That is in contrast to many other organ on a chip systems, which have tubes and pumps attached. With the 384 well plate, we can create little lanes of fluid, which are typically 200 microns wide and 100 microns in height.
All of these individual fluid lanes can be layered together. This way we can build tissues, layer by layer, just in the way as they are built up in our bodies. For example, in kidney tissue, there is a blood vessel, which has blood vessel wall. Then, there is intermediary tissue with fibroblasts, and then some immune cells. There is the epithelium of the kidney tubule. These different layers are adjacent to each other, and together make up the function of the kidney.
This is exactly what we do on the OrganoPlate. We layer tissues next to each other, in a horizontal fashion, and perfuse them with an artificial blood stream using a medium that keeps the cells live and also gets flow, mimicking the flow in normal organs.
In collaborations with pharma companies, we go from milestone to milestone, which are predefined. Typically, for example, we define setting up an initial model or individual cell types, create boundary tissue, then we hit a milestone and move towards an integrated model, and we end up, for example, with establishing a specific assay in that model.
This way, we work through a budget, step by step, in close collaboration with our customer, and, finally, we end up with a supply agreement where we provide them with OrganoPlates, which can be tailored to their needs or to the specific model, and we supply them for a year, for example, to run the model in their own labs.
What types of validation studies have you done to show that your technology is a good predictor of toxicity or efficacy in humans?
Joore: We are involved in a large range of model development projects where we collaborate with pharmaceutical companies to develop a model exactly to their specifications, either for a disease, or for a tissue, mostly in the context of toxicology, or ADME, or PK/PD like work.
At this point, we are very much still in the phase where we build the models, and we validate them against the gold standard among the systems that are currently available. It could be 2D cell cultures or animal models. The next phase we are heading toward is to show that these models have a strong predictive value. No company in the organs-on-chips field is there yet, but that is where we are heading.
Many pharma companies are re-evaluating their screening strategies. They are realizing that, with screenings and assays, it is rubbish in, rubbish out. If you do not put the right cells in, you are never going to measure what you really want to measure. More and more pharma companies want models with primary patient cells. They know it is more difficult and that it will probably take a bit more time, but when they have the model up and running, they have more confidence in what the data tell them.
Lastly, we understand that poly-dimethylsiloxane (PDMS), the typical polymer that is used as a substrate in organs-on-chips, can absorb particles, affecting the results. Could you comment on that?
Joore: Microfluidics has been relying on PDMS, a soft, silicone rubber. The downside is that pharmacological compounds, which are often hydrophobic compounds, strongly adhere to PDMS and are absorbed by it. Hence, doing any experiment where you have to tightly control the concentration of a compound at a low concentration is very difficult to do in PDMS-based devices. We decided to work with alternative materials that are low absorbent. For example, our plates are glass plates. The other polymers that we use in the process are also nonabsorbent, and they have all been selected for that reason.
Profile Jos Joore, Ph.D.
Jos Joore is a biotech entrepreneur with over 16 years of executive level bio-business experience. He held positions within biotech companies Pepscan, Skyline Diagnostics, Kreatech and Westburg. Jos holds a Master’s degree in Biology, a Ph.D. in Developmental Biology and a cum laude Master’s degree in Business Marketing.