InSphero's Akura™ technology platforms team worked with Professor Andreas Hierlemann's Bioengineering Laboratory at ETH Zürich to develop a microfluidic hanging drop system for embryotoxicity testing. What can we help you create?
The Embryonic Stem Cell Test (EST) is used to assess the embryotoxicity of compounds by employing the spontaneous differentiation of murine embryonic stem cells (mESCs) as a model for early embryonic development. In this test, 3D spherical embryoid bodies (EBs) are formed from pluripotent mESCs in the presence of a compound and serve as complex cell models to study spontaneous cell differentiation. The readout is based on the potential of the mESCs to differentiate into contracting cardiomyocytes, which is morphologically assessed by microscopy analysis. The EST is the only ECVAM-validated in vitro system that relies entirely on the use of a cell line.
InSphero has considerable experience with the EST, and we were able to optimize and substantially simplify the workflow using our hanging-drop technology. The original EST protocol called for assays to begin with the formation of embryoid bodies from a suspension of mESCs at the liquid–air interface of a “classical” hanging drop suspended below the lid of a petri dish (Validated EST). After three days of tissue formation, EBs are manually picked and transferred into a suspension culture to replenish the medium, then grown in suspension for two more days. On day 5, individual EBs are manually transferred to a third substrate, a well plate with a bottom surface that allows for tissue adhesion and spreading. This last step is necessary to get optical access to contracting areas within the EBs, the activity of which is used as a readout on day 10 to evaluate the embryotoxic effects of the compounds under test.
With ETH, we wanted to go a step further and investigate whether we could interconnect the target tissue – the EB – with the liver that, ultimately, will enable us to test not only the substance itself but its metabolites generated by the liver in situ. We thus wanted to include metabolic competence in the EST and thereby introduce the “metaEST”.
The platform we developed was based on a “hanging-drop network” approach that fluidically interconnects hanging drops and allows neighboring microtissues to “communicate” with each other. With our 3D InSight™ Human Liver Microtissues (hLiMT), we had a perfect match of technologies to perform the entire metaEST on a single multi-tissue microfluidic platform (MetaEST). During initial EB formation, preformed hLiMTs were cultured in disconnected hanging drops under static conditions. After only 24 hours, fluidic interconnection was established between the drops hosting the hLiMTs, connecting hLiMTs and EBs within the same network.
From that point on, liquid exchange between the compartments enabled direct inter-tissue communication, continuous medium turnover, and, most importantly, the dosage of target compounds and the exchange of metabolites. Medium exchanges throughout the assay were executed directly on‐chip. On day 5, the microfluidic chip was simply flipped upside‐down to attain a standing‐drop configuration. The hLiMTs settled to the substrate of the chip, and the EBs started to adhere and spread on the adhesive surface. The surface below the hLiMTs was nonadhesive, so the hLiMTs preserved their morphology and function. On day 10 of the assay, we optically assessed the EBs for cardiomyocyte differentiation directly on the microfluidic platform by searching for contracting areas in the spread‐out tissues.
The final metaEST design consisted of 9 interconnected drops in a row (4 conditions per chip). The central 3 drops hosted the EBs, the lateral 6 drops the hLiMTs. As a proof of concept, we used the well-known embryotoxicity prodrug cyclophosphamide (CP), an alkylating agent that interferes in mesodermal differentiation and causes severe congenital deformities or even death. The effect of CP depends on biotransformation through the liver to form both active and inactive metabolites. Whilst the ID50 value in the conventional EST was determined to be 1.2 ± 0.3 mM, the metaEST yielded a fourfold lower value of 0.3 ± 0.03 mM. The decreased ID50 concentration in the multi-tissue configuration can, in our opinion, be attributed to the presence of unstable embryotoxic metabolites, such as PM, generated by liver‐mediated biotransformation.
The present study demonstrates how we can forge new frontiers in 3D cell-based assays using organ-on-chip technologies. The implementation of the initial idea posed multiple challenges. We were able to address the requirements of robustness and repeatability by substantially reducing the number of handling steps and integrating everything into a single platform that can be simply flipped around. One of the biggest hurdles was to determine how to get all the required individual elements – including stem cells, functional chips, medium, instruments, etc. – ready for the assay and set up on day 0. An important factor in the success of this study was that we could rely on the availability of InSphero's standardized, assay-ready liver microtissues delivered on time a few days before the planned start of the embryotoxicity assay.
This metaEST platform is still an early-stage academic prototype, similar to where our Akura™ Flow technology was ~3-4 years ago. The next steps will be determined based on the response to this paper, which we anticipate will help us determine to what extent the metaEST assay could complement the validated EST assay in tiered-testing strategies as a second, advanced screening tool to eliminate false negative drug candidates identified by a first EST screen.
We work with a variety of academic and industry development partners to explore the potential of microfluidic and organ-on-a-chip technology in designing advanced in vitro assays. In addition to this metaEST study, we collaborated with FHNW University on a low-clearance drug study that has helped shape a 3D InSight™ DMPK platform using Akura™ Flow technology. If you have an idea for a potential application for Akura™ Flow that you'd like to develop with us, I'd be happy to discuss it with you!
All images in this blog are from this Boos et al. 2019 paper published in Advanced Science: