Poster N7 - A Microfluidic Approach to Investigate Combinatorial Effects of Decorin and Topography on TGF-Β1 Mediated Corneal Keratocytes Differentiation

Introduction: Within the corneal stroma, proteoglycans such as decorin play an important role in the development and maintenance of the extracellular matrix (ECM). During corneal wound healing, transforming growth factor-β1 (TGF-β1) is released into the stroma where it activates quiescent corneal keratocytes into highly contractile and ECM-secreting myofibroblasts which can disrupt tissue’s transparency and result in corneal scarring and blindness1. Previous studies have found that decorin exhibits high affinities for many molecules including type I collagen, TGF-β1, and various cell surface receptors2. Through its interaction with TGF-β1 and cell receptors, decorin has shown promise in reducing myofibroblast differentiation as a potential therapeutic against fibrosis. In the cornea, specifically, it is important to consider the context of the aligned collagen fibrils which make up the corneal stroma ECM. While it has previously been shown that aligned topographical cues affect keratocyte behavior, the relationship between aligned collagen fibrils and biochemical cues and their combined effect on cell behavior remains largely unknown. In this study we present a microfluidic method of fabricating and coating aligned type I collagen fibrils with the proteoglycan decorin to investigate the combined effect of decorin and aligned topography TGF-β1 mediated myofibroblast differentiation. Understanding the interplay between the biochemical cues and aligned topography present during corneal wound healing will provide valuable insight into pathological processes to enable the development of treatment and therapies for corneal disease such as corneal hazing or keratoconus.

Materials and

Methods: As described previously, random collagen fibrils and aligned collagen fibril micropatterns were fabricated by polymerizing a drop of collagen in a polydimethylsiloxane (PDMS) well or perfusing a solution of type I collagen through PDMS microchannels onto glass cover slips coated with a thin layer of PDMS3. Monomeric collagen substrates were also prepared to compare cell response to the decorin coating in the absence of fibrillar topography. To coat aligned collagen fibril micropatterns with decorin, a decorin solution was perfused through PDMS microchannels over aligned collagen fibrils and allowed to bind for 1 hour at 37 °C. To coat the random fibril and monomeric collagen substrates, the decorin solution was instead added onto the collagen coated surface within a PDMS well and allowed to bind for 1 hour at 37 °C. Normal Rabbit Keratocytes (NRK) were then seeded onto these substrates in serum free (SF) basal media with or without TGF-β1. NRK were then fixed, permeabilized, and blocked with 1% BSA before fluorescent labeling of actin (Phalloidin), nucleus (DAPI), and α-smooth muscle actin (α-SMA primary antibodies followed by Alexa-Fluor conjugated secondary antibodies) as a marker of myofibroblast differentiation.

Results, Conclusions, and Discussions:

Results: NRK were cultured for 120 hours in SF media with various concentrations of TGF-β1 (0, 5, 10 ng/mL) on type I collagen monomer, random fibril, or aligned fibril substrates with or without a decorin coating. Fluorescent imaging of Alexa-Fluor labeled decorin confirmed its attachment to the collagen substrates. Exposure to TGF-β1 resulted in α-SMA expression in NRK, with an increase from 25% to 30% of cells expressing α-SMA with 5 ng/mL versus 10 ng/mL of TGF-β1 added to the media (Figure 1). The addition of the decorin coating yielded a decrease in expression for both concentrations of TGF-β1. Interestingly, the percentage of α-SMA expressing cells decreased by similar margins for both concentrations of TGF-β1 to about 80% of the α-SMA expressing cells on uncoated substrates. NRK cultured in SF media without TGF-β1 maintained their quiescent phenotypes and did not express α-SMA irrespective of the presence of a decorin coating.

Discussion: Decreased α-SMA expression in cells cultured on decorin coated collagen substrates suggests that the decorin bound to the collagen can inhibit TGF-β1 signaling and myofibroblast differentiation. This effect may be a result of decorin sequestering growth factor to the collagen surface or by detaching from the collagen to inhibit TGF-β1 in solution before it can bind to cell receptors. However, the observation in this study that α-SMA expression was decreased by similar relative percentages for both high and low concentrations of TGF-β1 suggest that the decorin coating may also be acting at the cell-ECM interface to interfere with myofibroblast differentiation irrespective of the increased growth factor concentration. Confirmation of this interaction and further investigation of the mechanism by which decorin coated aligned collagen fibrils inhibit myofibroblast differentiation are therefore warranted and my lead to breakthroughs in the development of novel treatments using decorin for corneal fibrosis.

Conclusions: Taken together, these data suggest that (i) decorin bound to collagen can inhibit TGF-β1 signaling and myofibroblast differentiation in corneal keratocytes and that (ii) this inhibition is possible in both high and low concentrations of TGF-β1, suggesting the potential for multiple mechanisms for regulating the TGF-β1 signaling pathway in corneal keratocytes.

Acknowledgements (Optional): This work was supported by the National Institutes of Health Grant R01 EY030190.