Optimal Biomaterials for Tracheal Epithelial Grafts: A Systematic Comparative Analysis

Varma, Ratna 1, 2; Aoki, Fabio G 2; Soon, Kayla 2; Karoubi, Golnaz 1, 2; Waddell, Thomas K 1, 2

1. Institute of Biomaterials and Biomedical Engineering, University of Toronto; 
2. Latner Thoracic Surgery Research Laboratories, University Health Network

Tracheal injury, stenosis, and malignancy demand tracheal reconstruction which often fails due to the lack of a functioning epithelium. Tracheal reconstruction using stents, decellularized scaffolds, and synthetic biomaterials has resulted in inadequate epithelial growth and function. While naturally-derived biomaterials are promising, they have not been studied systematically. Therefore, we performed the first extensive comparative analysis to determine optimal biomaterials for this context. Our goal was to determine whether these biomaterials can serve as substrates for a surgically usable graft that provides epithelial support and mechanical strength for surgical manipulation. As such, our selection criteria included tensile strength, maintenance of 50% bulk mass across 5 weeks, 80-100% cell attachment and spreading, maintained metabolic activity, focal adhesion formation within 2 to 6 hours, and differentiation into 20-25% ciliated cells and 5-10% secretory cells.  

Hyaluronan-Poly (Ethylene Glycol) (HA-PEG), Chitosan-Collagen (CC), Collagen Vitrigel Membrane (CVM), Fibrin, Silk Fibroin (SF), and Gelatin were screened using a human bronchial epithelial cell line (BEAS-2Bs) and primary human tracheal epithelial cells (HTECs). Metabolic activity was measured via PrestoBlueTM. Young’s moduli of biomaterials were determined through tensile testing. Bulk degradation was assessed in a cocktail of lysozyme (L), collagenase (C), and protease XIV (P), referred to as LCP, across 5 weeks. HTEC focal adhesion formation (vinculin) and differentiation into secretory (mucin 5AC) and ciliated (acetylated α-tubulin) cells were examined via immunocytochemistry. 
        
HA-PEG did not allow BEAS-2B monolayer formation. CC, CVM, Fibrin, SF, and Gelatin allowed 80-100% BEAS-2B attachment and spreading with increasing metabolic activity over 10 days. SF demonstrated significantly higher Young’s modulus (P<0.01) than all biomaterials. Gelatin degraded by 1 week in LCP, while CC and SF retained up to 60% and 50% of their mass by week 5, respectively. CVM, Gelatin, and Fibrin were eliminated due to either weak mechanics or their inability to be measured. HTECs formed focal adhesions by 2 hours on Collagen I controls (Col) and SF, and by 24 hours on CC. Compared to Col, SF demonstrated comparable secretory cell (5.9 ± 2.6%; P=0.79), but significantly lower ciliated cell (5.5 ± 3.9%; P<0.05) development. SF was enhanced using CVM which itself produced comparable ciliated (P=0.81) and secretory (P=0.62) cells to Col. SF-CVM composite biomaterial was developed to achieve the mechanical and differentiation benefits of its respective components. SF-CVM led to statistically similar secretory (5.6 ± 1.0%; P=0.57) and ciliated (20.6 ± 1.7%; P=0.50) cell development to Col. 

A composite SF-CVM biomaterial was developed which meets all our selection criteria and demonstrates immense potential for augmenting current state-of-the-art methods for airway surgeries and tracheal transplants.