The Role of Hypoxia in Corneal Extracellular Matrix Deposition and Cell Motility

Abstract

The cornea is an excellent model tissue to study how cells adapt to periods of hypoxia as it is naturally exposed to diurnal fluxes in oxygen. It is avascular, transparent, and highly innervated. In certain pathologies, such as diabetes, limbal stem cell deficiency, or trauma, the cornea may be exposed to hypoxia for variable lengths of time. Due to its avascularity, the cornea requires atmospheric oxygen, and a reduction in oxygen availability can impair its physiology and function. We hypothesize that hypoxia alters membrane stiffness and the deposition of matrix proteins, leading to changes in cell migration, focal adhesion formation, and wound repair. Two systems-a 3D corneal organ culture model and polyacrylamide substrates of varying stiffness-were used to examine the response of corneal epithelium to normoxic and hypoxic environments. Exposure to hypoxia alters the deposition of the matrix proteins such as laminin and Type IV collagen. In addition, previous studies had shown a change in fibronectin after injury. Studies performed on matrix-coated acrylamide substrates ranging from 0.2 to 50 kPa revealed stiffness-dependent changes in cell morphology. The localization, number, and length of paxillin pY118- and vinculin pY1065-containing focal adhesions were different in wounded corneas and in human corneal epithelial cells incubated in hypoxic environments. Overall, these results demonstrate that low-oxygenated environments modify the composition of the extracellular matrix, basal lamina stiffness, and focal adhesion dynamics, leading to alterations in the function of the cornea. Anat Rec, 2019. © 2019 Wiley Periodicals, Inc.

Keywords: cornea; epithelium; extracellular matrix; focal adhesions; hypoxia.

Fig. 1.

Corneal epithelial cell morphology. Epithelial cells were cultured on fibronectin- and collagen IV-coated substrates of different stiffness values under normoxic (A) and hypoxic (B) conditions. Images are representative of a minimum of five independent experiments for each time point and condition. Changes in cell morphology are substrate and environment dependent. Scale bar = 66 μm.

Fig. 2.

The effects of hypoxia, matrix protein, and stiffness on cell circumference over time. Cell shape was determined using FIJI. (A,B) The effects of stiffness and substrate over time under normoxic conditions. (C,D) The effects of stiffness and substrate over time under hypoxic conditions. Statistical analysis was conducted (ANOVA). Standard error bars are ± SEM. *P < 0.05, **P < 0.01, ***P < 0.005, and ****P < 0.005.

Fig. 2.

The effects of hypoxia, matrix protein, and stiffness on cell circumference over time. Cell shape was determined using FIJI. (A,B) The effects of stiffness and substrate over time under normoxic conditions. (C,D) The effects of stiffness and substrate over time under hypoxic conditions. Statistical analysis was conducted (ANOVA). Standard error bars are ± SEM. *P < 0.05, **P < 0.01, ***P < 0.005, and ****P < 0.005.

Fig. 3.

Hypoxia modulates the localization of laminin and Type IV collagen in wounded corneas. Epithelial abrasions were performed and corneas incubated in normoxic or hypoxic environments for 18 hours. Corneas were immunostained for laminin-5 and Type IV collagen and counterstained with DAPI. Images were obtained at 40× magnification using a Zeiss Axiovert LSM 700 confocal microscope. The region outlined is shown in inset. A, Laminin-5 is reduced along the basal lamina and in the stroma of hypoxic corneas. Arrow indicates laminin-5 localization to the basal lamina. B, Type IV collagen localization is reduced along the basal lamina and stroma after exposure to hypoxic conditions. Arrow indicates Type IV collagen localization to the basal lamina in the normoxic cornea. Scale bar = 100 μm. *, wound edge. Images are representative of a minimum of three independent experiments. Nx, normoxia; Hx, hypoxia; E, epithelium; S, stroma; BL, basal lamina.

Fig. 4.

Localization of focal adhesion proteins is altered under hypoxic conditions. A–D, Corneas were wounded, incubated in normoxic or hypoxic environments for 18 hr, and stained for focal adhesion proteins. Images were obtained at 40× magnification using a Zeiss Axiovert LSM 700 confocal microscope and are representative of a minimum of three experiments. A, Talin localizes along the basal surface (arrows) and along basolateral surface (arrowheads) of corneal epithelial cells. B, Talin is diminished after exposure to hypoxic conditions. Arrows indicate localization to the basal surface. C, Paxillin pY118 localizes to the basal surface (arrows) and the basolateral surface (arrowhead) of epithelial cells D, Paxillin pY118 localization is present along the basal surface (arrows). E, Vinculin pY1065 localizes to the basal surface (arrows). F, Vinculin pY1065 is present at the wound edge along basal and apical surface. *, wound edge. Scale bar = 50 μm.

Fig. 5.

Localization of vinculin pY822 under normoxic and hypoxic conditions. Corneas were wounded and incubated in normoxic or hypoxic environments for 18 hr. Corneas were immunostained for vinculin pY822 and counterstained with DAPI. Images were obtained at 40× magnification using a Zeiss Axiovert LSM 700 confocal microscope. (A,B) Vinculin pY822 localizes to cell–cell contacts (arrows). Images are representative of a minimum of three independent experiments. *, wound edge. Scale bar = 50 μm.

Fig. 6.

Hypoxia affects the number and length of paxillin pY118- and vinculin pY1065-positive focal adhesions. Human corneal limbal epithelial (HCLE) cells were wounded and incubated under normoxic or hypoxic conditions for 4 hr. HCLE cells were fixed and immunostained for focal adhesion proteins. Images were obtained at 40× magnification using a Zeiss Axiovert LSM 700 confocal microscope. The leading edge was analyzed in 500 μm intervals. Focal adhesion number and length were determined using FIJI. A, Talin localized to the leading edge. B, No significant change in talin-positive focal adhesion number or length under hypoxic conditions. C, Paxillin pY118-positive focal adhesions (arrows). D, Increased number and decreased length of paxillin pY118-positive focal adhesions when exposed to hypoxia. E, Vinculin pY1065-positive focal adhesions (arrows). F, Hypoxia reduced the number and length of vinculin pY1065-positive focal adhesions at the leading edge. Images are representative of a minimum of at least four independent experiments. *, wound edge. Scale bars = 50 μm. Standard error bars are ±SEM. (unpaired t-test). *P < 0.05, **P < 0.01, ***P < 0.005, ns = not significant. FAs, focal adhesions.

Fig. 7.

Vinculin pY822 localized to cell–cell contacts under normoxic and hypoxic conditions. Cells were wounded and incubated under normoxic or hypoxic conditions and immunostained for vinculin pY822. Images were obtained at 40× magnification using a Zeiss Axiovert LSM 700 confocal microscope. The leading edge was analyzed in 500 μm intervals. The focal contact number and length were determined using FIJI. A, Vinculin pY822 localized primarily to cell–cell contacts. B, Hypoxia decreased the number but increased the length of vinculin pY822-positive cell–cell contacts. Images are representative of at least four independent experiments. *, wound edge. Scale bar = 50 μm. Standard error bars are ± SEM. (unpaired t-test). *P < 0.05, ***P < 0.005.