Poster L1 - Collagen I and hyaluronic acid hydrogels for bioengineering perfused brain microvascular networks

Thursday, October 24, 2024

10:00 AM – 11:00 AM EST

Location: Exhibit Hall E, F & G

Presenting Author(s)

Helena Zheng

Bioengineering & Biochemistry Undergraduate Student
Northeastern University, United States

Co-Author(s)

Christina A. Velez (she/her/hers)

PhD Candidate, Bioengineering
Northeastern University
Boston, Massachusetts, United States

Primary Investigator(s)

Guohao Dai, PhD

Professor of Bioengineering
Northeastern University
Boston, Massachusetts, United States

Introduction: The blood-brain barrier (BBB) is a tight network of blood vessels in the central nervous system, uniquely characterized to regulate neural function by its low permeability. The BBB is composed of vascular endothelial cells as well as astrocytes and pericytes, which are both supporting cells that contribute to BBB integrity. The extracellular matrix (ECM) plays key roles in the development and maintenance of the BBB, and is composed of collagen IV, laminin, heparan sulfate proteoglycans, and other proteins. Similar to in vivo, neurovascular models (NVMs) of the BBB rely on its hydrogel microenvironment for biochemical and mechanical cues, highlighting the importance of a biomimetic hydrogel. Current 3D NVMs use proteins, polysaccharides, or synthetic biomaterials. While synthetic materials produce less lot-to-lot variability, they may induce unpredictable cell-cell or cell-matrix interactions that reduce longevity and are not as physiologically relevant as natural biomaterials such as fibrin or collagen[1]. Fibrin is a natural biomaterial that is used for 3D vessel formation; however, it is not native to the brain, greatly reducing NVM longevity when neural progenitor cells (NPCs) are included and is responsible for thrombosis and certain inflammatory pathways in the brain[1]. This study explores the use of collagen I, a mechanically versatile protein-based hydrogel that supports BBB angiogenesis and cell attachment[2], and hyaluronic acid (HA), a structural glycosaminoglycan of the brain ECM vital for BBB development[3]. We hypothesize that optimizing the hydrogel material of NVMs to be more brain-specific improves the longevity of self-organized microvasculature and NPCs to create long-lasting, functional NVMs.

Materials and

Methods: ColHA hydrogels were made with final concentrations of 2, 2.5, and 3mg/mL collagen while HA was kept consistent at 10mg/mL. Lower working concentrations of the hydrogels are more thoroughly explored to represent softer brain tissue. Collagen is diluted with HCl, then neutralized with 10x PBS containing ions and NaOH. We opted for glycidyl methacrylate hyaluronic acid (GMHA), a methacrylated HA that has a higher molecular weight and forms looser scaffolds compared to other methacrylated HA’s, which results in a softer hydrogel to further mimic brain tissue. GMHA is obtained as a lyophilized solid that is solubilized in media, then combined with LAP photoinitiator to allow for free radical polymerization when exposed to UV light. Neutralized collagen solution and solubilized HA are then mixed together to create the ColHA hydrogel. Fibrin is prepared by mixing fibrinogen and thrombin to a final concentration of 6mg/mL and 3U/mL, respectively, and was used as a positive control for all experiments.
Endothelial cells expressing tdTomato (tdT-ECs), astrocytes, and pericytes are collected and resuspended in ColHA, then plated and placed at 37℃ for 30 minutes to polymerize collagen. The plate is then exposed to 405 nm UV light for 5 minutes to crosslink the HA. The cells are fed everyday with media and imaged to track self-organization of BBB microvascular networks over time using tdTomato fluorescence. Astrocytes and pericytes are visualized in fixed samples through antibody staining for glial fibrillary acidic protein (GFAP) and neural-glial antigen 2 (NG2), respectively.

Results, Conclusions, and Discussions: Fibrin control groups exhibited self-organized microvasculature formation by day 5, as expected from previous experiments and literature[4]. All collagen and HA only controls were incapable of self-organizing lasting microvascular networks. This is potentially due to a low collagen concentration which can result in loose, uncrosslinked fibrils dissolving in the media without additional support and lack of cell attachment sites in HA. ColHA hydrogels last a similar duration as fibrin of 21+ days, supporting the idea that HA provides structural integrity to the gel that collagen alone lacks.
ColHA with 3mg/mL collagen was the most comparable to fibrin, presenting with denser microvascular networks than lower ColHA collagen concentrations. The microvascular networks took longer to self-organize in ColHA compared to fibrin. The hollow lumen of the microvascular networks are more apparent in fibrin conditions compared to 3mg/mL ColHA, though this has only been investigated in a static culture and must be explored with the influence of interstitial flow which is known to improve the perfusion of bioengineered BBB microvascular networks[4].
These results so far suggest that while it is possible for BBB microvasculature to self-organize in a ColHA hybrid hydrogel, there is more optimization needed to obtain microvascular networks that are as dense and connected as those formed in fibrin, such as validating collagen and HA influences and stability in the hydrogel or adding other ECM proteins. The optimized ColHA will then be used in a microfluidic device to explore the influence of interstitial flow on the formation, longevity, and functionality of the brain microvascular networks. Finally, NPCs will be introduced within a microfluidic device to create a true NVM to assess its longevity and translatability.
Despite its early stages, a ColHA hydrogel shows promise in supporting NVM longevity, mitigating issues presented by other hydrogels while creating a brain-specific environment optimized for microvasculature self-organization. This contributes to the goal of making 3D models of the BBB a competitive alternative to animal models, which often fail to translate to humans and hinder drug development for neurological disorders.

Acknowledgements (Optional): I extend a special thank you to my graduate student mentor, Christina Velez, for her unwavering guidance and encouragement and Dr. Guohao Dai for his continued support!

APA References
[1] Adams, R. A., Passino, M., Sachs, B. D., Nuriel, T., & Akassoglou, K. (2004). Fibrin mechanisms and functions in nervous system pathology. Molecular interventions, 4(3), 163.
[2] Potjewyd, G., Kellett, K. A., & Hooper, N. M. (2021). 3D hydrogel models of the neurovascular unit to investigate blood–brain barrier dysfunction. Neuronal Signaling, 5(4), NS20210027.
[3] Bignami, A., Hosley, M., & Dahl, D. (1993). Hyaluronic acid and hyaluronic acid-binding proteins in brain extracellular matrix. Anatomy and embryology, 188(5), 419-433.
[4] Winkelman, M. A., Kim, D. Y., Kakarla, S., Grath, A., Silvia, N., & Dai, G. (2022). Interstitial flow enhances the formation, connectivity, and function of 3D brain microvascular networks generated within a microfluidic device. Lab on a Chip, 22(1), 170-192.