Abstract
Purpose :
Damage of the corneal nerve plexus (CNP) is one of the main causes of neurotrophic keratopathy. Spontaneous regeneration of the nerves is often incomplete and the underlying mechanisms are unknown. Therefore, stimulation of corneal nerve regeneration might be a promising approach to treat this condition. However, there is a lack of suitable in vitro models to test nerve regeneration. The aim of this study was to evaluate new 3D in vitro models in comparison to an in vivo mouse model regarding corneal nerve regeneration.
Methods :
For analyzing nerve regeneration in vitro, a collagen-gel based 3D model was used. Dorsal root ganglion cells (DRGs) cells from C57BL/6 mice were used and embedded into the rim of the gel. Human corneal fibroblasts were seeded into the center and limbal epithelial stem cells on top of the gel. Furthermore, a model with only DRGs or TGs seeded into the gel as well as 2D assays were used. Cells were incubated with 100µM Y27632 for different time points. QrtPCR, staining and fiber length measurement were used to evaluate the effects. For the in vivo experiments, the CNP of the right eyes of mice (n=24) was cut. Eyes were treated twice a day with eye drops containing either Y27632 or the vehicle as control. At D7, D14, D21 and D28 measurements of corneal nerves using confocal microscope was performed.
Results :
Measurement of neurite length showed a significant effect of ROCK inhibition in PCC gels with 5.17 ± 1.10 mm in treated and 1.62 ± 0.21 mm length in control gels (p = 0.0001). Comparable results were obtained in the 2D assays. Measurement in the in vivo experiments revealed a significant increase of mean fiber length for all timepoints. After 7 days, fiber length was reduced to 4.45 mm/HPF in the control and to 5.936 mm/HPF in the treated group. The fiber length in the treated group reached 8.59 mm/HPF compared to 3.72 mm/HPF in the controls at day 28. Qrt-PCR showed changes in expression in the TG of the treated compared to non-treated or native side. E.g. ROCK2 expression was reduced at D14 to 89.1% ±0.87% (p=0.007) compared to native control (100%).
Conclusions :
The data showed that using this new 3D corneal model led to similar results regarding nerve length compared to the in vivo experiments. Therefore, this model seems to provide a suitable, in-vivo like environment for specific research questions, with the possibility to test a high number of different substances in a short period of time.
This is a 2020 ARVO Annual Meeting abstract.