(M-502) Sacrificially templated hydrogels with pre-fabricated vasculature enable artificial tissue integration with host circulation

Introduction:: Despite a plethora of impressive vessel fabrication technologies, and more popping up each month, none have been able to recapitulate physiological vessel density. Although progress has been made on expanding pre-fabricated vascular networks in vivo, these methods need to be improved because slow or incomplete vascularization of tissue engineered constructs causes a lack of oxygen and nutrient supply, resulting in necrosis [1,2]. We have previously shown that fabricating hydrogel implants (6mm) containing cords of endothelial cells in collagen promote host infiltration as well as concomitant expansion of parenchymal cells and vascularization [3,4]. We now aim to study the vascularization promoting capacity of these endothelial cords in centimeter scale tissues, as well as examine the effects of different bulk hydrogel materials and cell populations on host infiltration.


Materials and Methods:: Hydrogels (19mm x 8mm x 4mm) were fabricated via sacrificial templating. Briefly, sacrificial templates of isomalt (a bioorthogonal sugar) were formed via selective laser sintering (SLS) using a custom built SLS machine, as described previously [5]. These templates were then filled with ~500uL of 10mg/mL fibrin, 5% GelMA (synthesized in house), or 15% GelMA. For hydrogels with cells in the bulk, 12M/mL HUVECs and 12M/mL NHDFs were suspended in the hydrogel precursor. Fibrin gels were crosslinked by mixing with thrombin and GelMA gels were crosslinked by exposure to 405nm light for 2 minutes. The sugar was then removed by flushing with warm PBS.

Endothelial cords were formed by injecting 200-300uL of 2.5mg/mL collagen I (Corning) containing 10M/mL HUVECs and 200K/mL C3H10T1/2s into the vacant lumens where the sugar used to be. The cords were allowed to contract at 37C for 3 hours before implantation into the intraperitoneal space of athymic nude mice. After 7 days, the gels were then explanted and processed for 3D staining via the Ce3D clearing method [6]. Vascularization was evaluated via perfusion of lectins injected retro-orbitally into the mice and subsequent quantification of total vessel area and average vessel length using NIH Java plugin Angiotool. Additionally, immunohistological analysis was done for Ter-119 (erythroid cell marker) and CD31 (endothelial marker).


Results, Conclusions, and Discussions:: In 10mg/mL fibrin and 5% GelMA gels, perfusion of both human and mouse vessels was observed at varying degrees. Furthermore, the formation of chimeric vasculature consisting of both implant- and host- endothelium was indicated by overlapping signals of human and mouse lectins. The presence of human, mouse, and chimeric vessels demonstrates integration of the implanted sacrificially templated endothelial networks with the host vasculature. No significant differences in total vessel area between all groups was detected although a slight trend towards cords made without 10T1/2s in 5% GelMA (GelMA -) having higher average vessel area was observed. However, average perfused human vessel length in 5% GelMA – was found to be significantly longer than those found in 10mg/mL fibrin gels containing cords made with HUVECs and 10T1/2s (Fibrin +). Interestingly, when grouped by +/-10T1/2s and independent of material, constructs without 10T1/2s had significantly longer average perfused vessel length. Furthermore, when comparing 5% GelMA constructs to 10mg/mL fibrin constructs, the average perfused vessel lengths found within 5% GelMA were on average longer than those found in 10mg/mL fibrin.

Analysis is currently ongoing for 5% versus 15% GelMA gels as well as hydrogels containing cells in the bulk.



Sacrificially templated fibrin and GelMA implants offer promising approaches for generating robust networks of perfusable vessels in vivo. However, with no significant difference in total vessel area across all conditions, there is potential for both materials to be used for different applications or environments. For example, longer vessels formed in GelMA may be able to replace or support larger arteries or veins, but shorter vessels of similar total area formed in fibrin may more effectively create perfusable capillary networks in a more localized area. Furthermore, given that stromal cell support of endothelial cords in fibrin gels have repeatedly been shown to induce vascularization in vivo, it was surprising that GelMA constructs without 10T1/2s had higher average vessel area and significantly longer average vessel length. Additional research into the relationship of different materials, stromal cells, and vascularization mechanisms is required.


Acknowledgements (Optional): :

References (Optional): :

1. 
   Masson-Meyers DS, Tayebi L. Vascularization strategies in tissue engineering approaches for soft tissue repair. J Tissue Eng Regen Med. 2021 Sep;15(9):747-762. doi: 10.1002/term.3225. Epub 2021 May 31. PMID: 34058083; PMCID: PMC8419139.
   
   


2. 
   Yang, G., Mahadik, B., Choi, J. Y. & Fisher, J. P. Vascularization in tissue engineering: fundamentals and state-of-art. Prog. Biomed. Eng. 2, 012002 (2020).
   
   


3. 
   Raghavan S, Nelson CM, Baranski JD, Lim E, Chen CS. Geometrically controlled endothelial tubulogenesis in micropatterned gels. Tissue Eng Part A. 2010 Jul;16(7):2255-63. doi: 10.1089/ten.TEA.2009.0584. PMID: 20180698; PMCID: PMC2947948.
   
   


4. 
   Brady EL, Kirby MA, Olszewski E, Grosjean P, Johansson F, Davis J, Wang RK, Stevens KR. Guided vascularization in the rat heart leads to transient vessel patterning. APL Bioeng. 2020 Mar 5;4(1):016105. doi: 10.1063/1.5122804. PMID: 32161835; PMCID: PMC7058427.
   
   


5. 
   Kinstlinger IS, Saxton SH, Calderon GA, Ruiz KV, Yalacki DR, Deme PR, Rosenkrantz JE, Louis-Rosenberg JD, Johansson F, Janson KD, Sazer DW, Panchavati SS, Bissig KD, Stevens KR, Miller JS. Generation of model tissues with dendritic vascular networks via sacrificial laser-sintered carbohydrate templates. Nat Biomed Eng. 2020 Sep;4(9):916-932. doi: 10.1038/s41551-020-0566-1. Epub 2020 Jun 29. Erratum in: Nat Biomed Eng. 2021 Aug;5(8):941. PMID: 32601395.
   
   


6. 
   Li, W., Germain, R.N. & Gerner, M.Y. High-dimensional cell-level analysis of tissues with Ce3D multiplex volume imaging. Nat Protoc 14, 1708–1733 (2019). https://doi.org/10.1038/s41596-019-0156-4