(G-251) Self-differentiable in vitro coronary arteriole model based on the microfluidic chip

Cardiovascular disease is a major global cause of death, with Coronary Microvascular Dysfunction (CMD) being a severe and yet not fully understood condition. CMD refers to structural and functional defects in the coronary microvasculature and has recently been recognized as a distinct disease entity. However, due to the challenges of in vivo manipulation and observation, particularly in humans, the pathogenesis of CMD remains elusive.

As an alternative for in vivo CMD studies, several in vitro coronary vasculature models have been proposed. These models enable the application of various physical and biochemical stimuli and the analysis of resulting cellular responses. Unfortunately, most previous models oversimplify the tunica media of the vessel, a critical player in tone regulation, to a Smooth Muscle Cell (SMC) laden extracellular matrix (ECM) without proper SMC differentiation. Notably, some researchers have proposed an advanced tissue-engineered blood vessel with a functional tunica media, achieved through a condensed SMC layer. However, building this model necessitates manual ECM dehydration using tissue and tying the interconnections by hand to prevent leakage.

Here, we present an arteriole-sized in vitro coronary microvascular chip that tackles these aforementioned challenges. Our model features a properly condensed and differentiated tunica media layer, constructed through the self-contraction of the SMC layer. Additionally, employing two types of surface coatings, polydopamine (PDA) and F-127, facilitates flow induction in the in vitro vessel, ensuring no leakage without any additional experimental steps after seeding. We believe our model enhances efficacy, reduces contamination, and can be applied to various CMD studies.

To construct the Tunica intima and Tunica media, RFP-labeled HUVECs (Angioproteome) and GFP-labeled AoSMCs (Angioproteome) from pasages 4-7 are respectively used. These cells between passage 4-7 were cultured in EGM-2mv and Smgm(Lonza)).

A gel channel with precise dimensions(1600µm(w),1500µm) was created by curing PDMS on a 3D-printed mold and bonding it with glass using plasma treatment. Notably, to create a void space that will serve as a guide for the stainless tube to be inserted for tubing, curing is done while inserting 25G needles from both sides. Following the pre-made void hole, a suture coated with 4% F-127 is inserted and bonded. Subsequently, the entire gel channel is recoated with a 4% F-127 solution for 2 hours. This step increases hydrophobicity within the gel channel, promoting SMC contraction. Afterward, all F-127 is removed, and the stainless tube, coated with a 4% PDA solution, is inserted along the guide, completing the chip fabrication.

SMC-laden Col-Fib co-gel (SMC : 3e6/ml, COL: 0.175mg/ml, Fib: 7.5mg/ml) was inserted into the gel channel by traversing the suture. Remarkably, distinct variations in cell differentiation were observed based on these concentrations. After 20 minutes of gelation, the suture was removed, and RFP-labeled HUVECs(4e6/ml) were injected. The HUVECs were allowed to adhere for an hour, completing the model.

With a flow rate of 5μl/min applied for three days, the researchers observed self-contraction of SMCs and the formation of a dense tunica media. They also made   diverse cellular responses to the presence of flow and interluminal pressure.

Using the fabrication method described above, we successfully created an in vitro coronary microvascular model with a diameter ranging from 50 to 150 μm. The most noteworthy aspect is that we were able to mimic the dense tunica media found in vivo through Self-contraction, without the need for extra handwork. The model remained viable for up to 6 days and allowed for easy addition of biochemical drugs or cytokines through the space created by Contraction of the gel channel. Furthermore, we confirmed the possibility of immunostaining.

During the optimization process of the model, we encountered issues such as lack of SMC contraction or in vitro vessel detachment from the stainless tube when the chip and stainless tube were not coated respectively. Additionally, low SMC density or excessively high Fibrinogen concentration resulted in inadequate contraction, while low Fibrinogen concentration led to gel collapse during Flow induction. Through the optimization process, we established a robust fabrication process and improved the yield.

Notably, the diameter and SMC differentiation within the channel were influenced by intraluminal pressure and the presence of flow, confirming that the physical stimuli play a crucial role. We believe that our model mimics the in vivo arteriogenesis process.

This work was supported by the National Research Foundation of Korea(NRF) grant funded by the Korea government(MSIT) (No. 2020R1A5A8018367).

[1] Zhang, X., Bishawi, M., Zhang, G. et al., Nat. Commun. 11, 5426 (2020).

[2] Gao, G., Kim, H., Kim, B., Kong, J., Lee, J. Y., Park, B. W., Chae, S., Kim, J., Ban, K., Jang, J., Park, H.J., Appl. Phys. Rev., 6, 041402(2019)

[1] Zhang, X., Bishawi, M., Zhang, G. et al., Nat. Commun. 11, 5426 (2020).

[2] Gao, G., Kim, H., Kim, B., Kong, J., Lee, J. Y., Park, B. W., Chae, S., Kim, J., Ban, K., Jang, J., Park, H.J., Appl. Phys. Rev., 6, 041402(2019)