Urethral strictures are a common problem amongst men and lead to a decreased quality of life. Management with minimally invasive procedures shows high stricture recurrence rates, which could be improved by the addition of local anti-fibrotic adjuncts. However, molecular research into urethral strictures is hampered by the lack of reliable models of the human urethra. Differences in the genitourinary tract between animals and humans result in a lack of models for studying urethral stricture disease. Rodents have a baculum which is absent in humans. Rabbits have a more similar anatomy, but their urine composition is completely different. Despite this, animals are used in the study of the male urethra, leading to problems in translation of the results and to unnecessary use of animals.

An in vitro urethral model would therefore be of great benefit. The ideal model would consist of the same combination of cells and structures as the native human urethra. In such a model, the pathophysiology of urethral stricture disease can be studied. In vitro generation of a vascular bed resembling the periurethral tissue has been reported before [1]. The aim of this study was to create an in vitro human urethra-on-a-chip by combining the previously reported vascular bed with small channels seeded with epithelial cells. This animal-free urethra model could allow  experimental pre-clinical studies of prevention of stricture or their recurrence by anti-fibrotic compounds.

Bioreactors for the human urethra-on-a-chip model were designed in Solidworks, consisting of four reservoirs with corresponding in- and outlets on top and sides of the bioreactor. A negative mold was fabricated using digital light processing (DLP) 3D printing and cast with polydimethylsiloxane (PDMS) to produce a PDMS bioreactor after curing. PDMS bioreactors were bound to glass coverslips after corona discharge treatment. Total dimensions of the PDMS bioreactors were 19 by 40 mm, dimensions of the reservoirs were 4 by 3 mm (Figure 1A). 
As the different cells that were combined in the bioreactor (endothelial, mural and epithelial cells) each require a specific medium, the optimal medium composition supporting viability, proliferation, and functionality of all three cell types needed to be established. Both metabolic activity by Alamar Blue on monocultures of cells as well as the vascular network potential of a coculture of endothelial and mural cells in hydrogel followed by analysis of the branching of the microvessels were tested in different medium compositions. 
The reservoirs in the bioreactor were filled with cross-linkable hydrogel encapsulated with endothelial and mural cells (pericytes) to form microvascular networks. A perfusable microchannel (0.3-0.6 mm) was patterned in the reservoir by needle patterning (Figure 1 B and C). Patterned channels within the hydrogel were seeded with epithelial cells to create a cellular monolayer in the microchannel. 
Interaction of the microvascular bed to the microchannels were analyzed using a THUNDER microscope (Leica, IL, USA).

Bioreactors could be fabricated and an optimal medium composition was established. Cells responded differently to different medium concentrations. Endothelial cells as well as epithelial cells did not show different metabolic activity when grown on different medium ratios. Only pericytes showed reduced metabolic activity in the Alamar Blue assay when they were grown on medium other than endothelial growth medium. The optimal medium combination was epithelial growth medium mixed with endothelial growth medium mixed in a 1:1 ratio. In a functional assay no significant difference was observed: microvessel-like structures in hydrogels under coculture medium conditions did not show significant differences when compared to those that formed under conditions ideal for vascular networks. The microvessel-like structures were generated along the x-, y-, and z- axis, which implies that the structures were formed in a 3D orientation. Pericytes maintained their functionality since they embrace and support the endothelial surface of the microvessel-like structures. 
Microchannels in the fabricated bioreactors were made by needle patterning and could be seeded with cells in the 1:1 medium combination. Confluent coverage of the channels was observed by epithelial cells (Figure 1D). Bioreactors were viable under static conditions up to 7 days after seeding. Endothelial cells showed sprouting potential towards the other cell types in the bioreactor.

The innovative character of this project is the development of a small bioreactor to produce a urethra-on-a-chip. With this model, an in vitro, human-like model of the biological complexity of the urethra is recreated in an animal-free setup. 
Several reservoirs can be cultured simultaneously to study variations of dynamic microenvironmental cues, such as flow to induce wall shear stress and circumferential strain. Both the flow can be variated (continuous, intermittent or pulsatile) as the composition of the fluid. 
Future goals are to create an inflammatory environment as well as strictures by damaging the epithelial layer. Thereafter it will be possible to screen for several anti-fibrotic compounds for their use in prevention o urethral stricture formation or recurrence.

We found that it is feasible to create a urethra-on-a-chip model. Such a model can reduce the use of laboratory animals and can overcome translational problems. It may offer the possibility of testing different physical conditions, such as flow, and chemical components on the development and/or prevention of urethral stricture formation and recurrence. Furthermore, these bioreactors provide us insight in the requirements for in vitro urethra’s that can ultimately assist in producing larger grafts to be used in urethral reconstruction surgery.

van Velthoven MJJ et al. Gel Casting as an Approach for Tissue Engineering of Multilayered Tubular Structures. Tissue Eng Part C Methods. 2020 Mar;26(3):190-198. doi: 10.1089/ten.TEC.2019.0280. PMID: 32089096.

Continence 7S1 (2023) 100855
DOI: 10.1016/j.cont.2023.100855