ROS-Mediated Fragmentation Alters the Effects of Hyaluronan on Corneal Epithelial Wound Healing

Abstract

A buildup of reactive oxygen species (ROS) occurs in virtually all pathological conditions. Hyaluronan (HA) is a major extracellular matrix component and is susceptible to oxidation by reactive oxygen species (ROS), yet the precise chemical structures of oxidized HA products (oxHA) and their physiological properties remain largely unknown. This study characterized the molecular weight (MW), structures, and physiological properties of oxHA. For this, high-molecular-weight HA (HMWHA) was oxidized using increasing molar ratios of hydrogen peroxide (H₂O₂) or hypochlorous acid (HOCl). ROS lead to the fragmentation of HA, with the oxHA products produced by HOCl exhibiting an altered chemical structure while those produced by H₂O₂ do not. HMWHA promotes the viability of human corneal epithelial cells (hTCEpi), while low MWHA (LMWHA), ultra-LMWHA (ULMWHA), and most forms of oxHA do not. HMWHA and LMWHA promote hTCEpi proliferation, while ULMWHA and all forms of oxHA do not. LMWHA and some forms of oxHA promote hTCEpi migration, while HMWHA does not. Finally, all native forms of HA and oxHA produced by HOCl promote in vivo corneal wound healing, while oxHA produced by H₂O₂ does not. Taken together, our results show that HA fragmentation by ROS can alter the physiological activity of HA by altering its MW and structure.

Keywords: corneal epithelium; high-molecular-weight hyaluronan; low-molecular-weight hyaluronan; oxidative stress; wound healing.

Figure 1

Schematic representation of the preparation of the oxidized forms of HA. Both H₂O₂ and HOCl were used to oxidize HA at different molar ratios. Samples were incubated at 37 °C for 4 h, and thereafter, H₂O₂ and HOCl were removed from the samples by gel filtration chromatography using a PD10 desalting column. The samples were quantified, aliquoted, and stored for use.

Figure 2

Characterization of the molecular weight of oxHA. The molecular weight of the differently prepared oxHA samples was determined by agarose gel electrophoresis (A–C) and by gel filtration using HPLC (D,E). (A) Ladder (La) and HMWHA were subjected to agarose gel electrophoresis. (B) La, HMWHA, and both H₂O₂ and HOCl at molar ratios of 1:50 and 1:250 were subjected to agarose gel electrophoresis. (C) La, LMWHA, HMWHA, and HOCl-treated HMWHA at the molar ratios of 1:1250 and 1:6250 were subjected to agarose gel electrophoresis. (D) HMWHA, LMWHA, ULMWHA, and H₂O₂-treated HMWHA were subjected to gel filtration using HPLC. (E) HMWHA, LMWHA, ULMWHA, and HOCl-treated HMWHA were subjected to gel filtration using HPLC.

Figure 3

Chemical characterization of oxHA by NMR. ULMWHA, oxHA in the LMWHA range generated by H₂O₂ oxidation and oxHA in the ULMWHA range generated by HOCl oxidation were analyzed. (A) ¹H NMR spectra of native HA. (B) ¹H NMR spectra of oxHA produced by H₂O₂. (C) ¹H NMR spectra of oxHA produced by HOCl are presented. The asterisk shows the chemical shift in acetamido protons upon chlorination.

Figure 4

Biological effects of different oxidized forms of HA on human corneal epithelial cell viability and proliferation. (A) hTCEpi were treated with HMWHA, LMWHA, ULMWHA, and both H₂O₂- and HOCl-treated HMWHA, and viability was assayed using the CCK-8 kit. Values are presented as percentage change compared to the vehicle control (PBS). (B) hTCEpi was treated with HMWHA, LMWHA, ULMWHA, and both H₂O₂- and HOCl-treated HMWHA, and proliferation was assayed using the BrdU assay kit. Values are presented as percentage change compared to the vehicle control (PBS). η represents p ≤ 0.05 in reference to HMWHA and * represents p ≤ 0.05 in reference to PBS control. Each experiment was carried out in triplicate and carried out three times.

Figure 5

Biological effects of different oxidized forms of HA on corneal epithelial cell migration in vitro. (A) Confluent hTCEpi were subjected to scratch wounds and maintained in media containing 0.02 mg/mL HMWHA, LMWHA, ULMWHA, or H₂O₂-treated HMWHA. Images were collected 0, 6, 12, 18, 24, and 30 h after injury, the wounded area was calculated using Image J, and data are presented as the percentage of wounded area remaining compared to 0 h. (B) Representative images of the wounded area over time. The boundaries of the wounded area are demarcated with a dashed white line. (C) Confluent hTCEpi were subjected to scratch wounds and maintained in media containing 0.02 mg/mL HMWHA, LMWHA, ULMWHA, or HOCl-treated HMWHA. Images were collected 0, 6, 12, 18, 24, and 30 h after injury, the wounded area was calculated using Image J, and data are presented as the percentage of wounded area remaining compared to 0 h. Each experiment was carried out in triplicate and carried out three times. * represents p ≤ 0.05 in reference to PBS control.

Figure 6

Biological effects of different oxidized forms of HA on corneal epithelial cell migration in vivo. C57BL6 mice were subjected to debridement wounds and topically treated with HMWHA, LMWHA, ULMWHA, or oxHA generated by H₂O₂ or HOCl at 2 mg/mL. (A) Debridement wounds were treated with HA of different MWs and compared to PBS. The wounded area was evidenced with fluorescein, the wounded area was calculated with imageJ, and data are presented as the percentage of wounded area remaining compared to the wounded area at 0 h. (B) Representative images of the data are presented in panel (A). (C) Debridement wounds were treated with oxHA generated by H₂O₂, compared to PBS and HMWHA. The wounded area was evidenced with fluorescein, the wounded area was calculated with imageJ, and data are presented as the percentage of wounded area remaining compared to the wounded area at 0 h. (D) Representative images of the data presented in panel (C). (E) Debridement wounds were treated with oxHA generated by HOCl, compared to PBS and HMWHA. The wounded area was evidenced with fluorescein, the wounded area was calculated with imageJ, and data are presented as the percentage of wounded area remaining compared to the wounded area at 0 h. (F) Representative images of the data are presented in panel (E). * represents p ≤ 0.05. Images for PBS and HMWHA in panel B, D, and F are adapted from [64], and copyright has been gained from the authors and the journal TVST.