The paper in Nature Biomedical Engineering is here: https://go.nature.com/2qVA789
When I joined Wilbur Lam, MD, PhD’s lab in 2012, they had just published a seminal description of their microvasculature-on-chip technology via an “endothelialized” PDMS-based microfluidic approach in the Journal of Clinical Investigation1. Because of my background on biomaterials and vascular tissue engineering, I was challenged by Dr. Lam to develop a microvasculature-on-a-chip 2.0, in which he wanted to replace the PDMS with a more physiologically-relevant biomaterial while also maintaining the branching geometry and size-scale of the vascular space (<30 μm). Moreover, the system had to be perfusable and enable the study of pathophysiological cellular interactions that occur in blood diseases, given that Dr. Lam is a clinical pediatric hematologist as well as a biomedical engineer.
I was very excited about this project, and as I surveyed the work from other research groups, I learned that there was a technical challenge in integrating constant controllable flow in hydrogel-based microchannels at a size-scale under 30 μm2. I realized that it would be very difficult to find a single hydrogel material possessing all the required properties, including tunable mechanical strength, cell attachment, optical transparency, and compatibility with conventional microfluidic techniques, to accomplish the task at hand. However, I then hypothesized that mixing two hydrogels together into an interpenetrating network (IPN) might yield a new material endowed with specific properties of the two constituent hydrogels. Indeed, as it turned out, agarose-gelatin IPN hydrogel fulfilled these requirements.
We soon proved that this IPN hydrogel was exactly what we were looking for. After we tuned the IPN to the physiologically-relevant stiffness of the sub-endothelium, we discovered that endothelial cells cultured in the microdevice self-formed a tubular monolayer, self-deposited the appropriate basement membrane matrix, appropriately developed a pro-inflammatory phenotype in response to the cytokine TNF-α and, importantly, exhibited physiological barrier function for well over a month, a feature never reported before for other endothelialized microfluidic devices.
We were very excited about this achievement, as these properties ensure the investigation of the long-term effects on endothelial barrier function that occur in vivo and that are difficult to study using other models, in vitro or in vivo. The first disease we studied was sickle-cell disease (SCD), in which a genetic mutation alters the shape and mechanical properties of red blood cells (RBCs) leading to hemolysis and release of heme in circulation. This platform enabled us to explore how heme and sickle RBC interactions with endothelial cells in and of themselves can induce endothelial barrier dysfunction and microvascular occlusion in SCD, and what the role of the mechanical alteration of RBCs is in this context—an unsolved question in the SCD field (Figure 1).
Figure 1: Representative 3D confocal microscope images of endothelialized microchannels occluded by sickle RBCs (pre-stained with R18; pointed by arrows, left) and the resultant, co-localized increased endothelial permeability and leakage of BSA-AF488 in situ (right). The figure is reproduced from Fig. 4 of our paper.
While conducting these studies we met with a local expert in malaria, Dr. Tracey Lamb, who became very interested in our device. According to her, a long-standing debate exists in the malaria field: it was unclear whether interactions between the endothelium and malaria-infected RBCs alone can cause endothelial barrier dysfunction and vascular occlusion, as other groups have argued that the presence of leukocytes is required. Once again, our device proved to be extremely effective at addressing this longstanding issue and showed striking results (Movie 1).
Looking back at this work, I realized that the journey has been extremely worthwhile and reiterated the importance of developing in vitro microvasculature-on-chip models to be as physiologically-relevant as possible. Only then will we be able to better mimic the in vivo microenvironment and discover the mysterious secrets hiding behind disease. Looking forward, I am very confident that our microvascular-on-a-chip model will broaden our knowledge of pathophysiology of many diseases that are associated with vascular dysfunction and hematology disorders.
Movie 1: Perfusion of malaria-infected RBCs into the engineered microvasculature.
Our paper: Qiu, Y. et al. Microvasculature-on-a-chip for the long-term study of endothelial barrier dysfunction and microvascular obstruction in disease. Nat. Biomed. Eng. doi:10.1038/s41551-018-0224-z (2018).
1. Tsai, M. et al. In vitro modeling of the microvascular occlusion and thrombosis that occur in hematologic diseases using microfluidic technology. J. Clin. Invest. 122, 408-418 (2012).
2. Wong, K.H.K., Chan, J.M., Kamm, R.D. & Tien, J. Microfluidic models of vascular functions. Annu. Rev. Biomed. Eng. 14, 205-230 (2012).