Acacia honey accelerates in vitro corneal ulcer wound healing model

Acacia honey (AH) was purchased from Ministry of Agriculture, Malaysia and gamma irradiated at 25 kGy at Ministry of Science, Technology and Innovation, Malaysia. Our previous study revealed that the optimal concentration of AH was 0.025 % [9]. The same study observed that the AH was not cytotoxic at the range of 0–3.125 % concentrations using H₂O₂ at the concentration of 1.56nM as positive control [9].

Primary cultures of corneal stromal cells were isolated and cultured as described previously [10]. Briefly, six New Zealand white strain rabbits’ corneas (2.0–2.5 kg) were excised by circular incision 2 mm beyond the corneoscleral junction. Connective tissue such as endothelium, ocular muscle, sclera, ciliary body and iris were stripped off and discarded. Following rinsing with phosphate buffered solution (Gibco Invitrogen, USA), corneas were immersed in Dispase solution 2 mg/ml (Sigma-Aldrich, USA) at 4 °C for 18 h to dissociate the stromal layer from the epithelium. Loose epithelial layer were gently removed with a scalpel (Fisher Scientific, Los Angeles, CA) and the remaining stroma was rinsed, sliced and digested with 0.3 % collagenase type I with intermittent gentle shaking at 37 °C for 90 min. The digested stroma was centrifuged at 500 x g for 10 min and the resultant pellet was washed twice with phosphate buffered solution (PBS, pH 7.2, Gibco Invitrogen, USA) to remove the remaining enzyme. Viable stromal cells were seeded in T75 cm² culture flask (BD Falcon, Franklin Lakes, NJ) with seeding density of 5 × 10⁴ cell/cm² in medium containing serum (FDS) consisting of Ham’s F-12: Dulbecco’s Modified Eagle’s Medium (Gibco), with 10 % foetal bovine serum (FBS; Gibco), 1 % of antibiotic and antimitotic (Gibco), 1 % of 50 μg/ml ascorbic acid (Sigma, St. Louis, USA). Stromal derived corneal fibroblasts were cultured in a 5 % CO₂ incubator (Jouan, Duguay Trouin, SH) under 95 % humidity at 37 °C. At 80 % confluence, the primary culture (P0) was trypsinized using 0.125 % trypsin-EDTA (Gibco) and subcultured to passage 1 (P1) in FDS with seeding density of 5 × 10³ cells/cm² in six well-plates (BD Falcon, Franklin Lakes, NJ) under the same condition as the primary culture. Medium was changed every 2 days and the cultures were monitored everyday using inverted phase contrast microscope (Carl Zeiss, Germany) to examine the morphological changes or any sign of contamination.

To evaluate the migration of corneal fibroblasts, a defect was created onto confluent monolayer culture. Corneal fibroblasts at passage 1 were seeded in the six well plates (BD Falcon, Franklin Lakes, NJ) at the density of 5 × 10³ cells/cm² in FDS medium until it reached confluence at day 3. The medium was discarded and corneal ulcer was created using 4 mm corneal trephine onto the confluent monolayer corneal fibroblasts culture. The cells within the circumference of the wound were gently scraped off using a sterilized cotton bud. The cultures were rinsed twice with PBS after wounding to remove non-adherent cells and cultured in four different media; basal medium (FD), medium containing serum (FDS), FD with supplementation of 0.025 % AH (AH-enriched FD medium), and FDS with supplementation of 0.025 % AH (AH-enriched FDS medium). Cells were cultured at 37 °C in a 5 % CO₂ incubator and media were changed every 3 days. The wound area was measured using Axiovision LE64 software and the percentage of wound closure was quantified by measuring the average percentage of wound closure (n = 6 independent experiments) by migrating cells at the wound edges in the initial day of wounding (day 0), day 3 and day 6 post wound creation (Fig. 1) using the formula below:

Formula:

Morphological assessment, gene expression and immunocytochemistry analyses were performed at the same time point of the cell migration study.

Total RNA from corneal fibroblasts cultured in four different media at day 0, day 3 and day 6 following in vitro wound creations were extracted using TRI Reagent (Molecular Research Centre, Cincinnati, USA) according to the manufacturer’s protocol. In brief, chloroform was added into the homogenate to isolate the transparent aqueous contained total RNA. Isopropanol and Polyacryl carrier (Molecular Research Centre) were added to each extraction to precipitate the total RNA. The extracted RNA pellet was washed with 75 % ethanol and air dried before dissolving it in Rnase and Dnase free distilled water (Invitrogen, Carlsbad, USA). Complementary DNA was synthesized from 100 ng of Total RNA with Super Script™ III First-Strand Synthesis Super Mix reverse transcriptase (Invitrogen, Carlsbad, USA). The protocol conditions were 10 min at 23 °C for primer annealing, 60 min at 50 °C for reverse transcription and 5 min at 85 °C for reaction termination. The expression of aldehyde dehydrogenase (ALDH), vimentin, alpha-smooth muscle actin (α-SMA), collagen type I, lumican and matrix metalloproteinase 12 (MMP12) were evaluated by two-step reverse transcriptase-polymerase chain reaction (Invitrogen, Carlsbad, USA). Positive control was observed using GAPDH primer pairs. All gene-specific primers used are shown in Table 1. The two-step RT-PCR reaction was performed using SYBR Green as the indicator in a Bio-Rad iCycler (Bio-Rad, USA). RT-PCR was carried out using 12.5 μl of iQ SYBR Supermix, 1 μl each of forward and reverse primers (GenBank, using Primer-3), 8.5 μl deionized water and 1 μl of cDNA template. The cycling conditions for all primer pairs were as follows: cycle 1: 95 °C for 3 min (1×), cycle 2: Step 1 95 °C for 10 s and Step 2 61 °C for 30 s (40×), followed by melting curve analysis. The specificity and PCR product size were visualized by 2 % agarose gel electrophoresis with ethidium bromide staining.

Corneal fibroblasts cultured on cover glass in four different media at day 0, day 3 and day 6 post wound creation were fixed with 4 % paraformaldehyde at 4 °C using standard protocol from Dako Animal Research Kit. Briefly, the slides were rinsed with running tap water for 3 min. Following incubation with blocking agent (0.03 % peroxidase block) at room temperature for 5 min, cells were labelled using the Biotinylation reagent and incubated with primary antibodies for 30 min. Primary antibodies used were anti-ALDH (1:100, Dako), anti-Vimentin (1:200, Dako) and anti-α-SMA (1:100, Dako). Biotinylation reagent was mixed to bind biotinylated secondary antibody to the primary antibody. The specimen was then incubated with streptavidin-peroxidase, followed by reaction with diaminobenzidine/hydrogen peroxide as substrate-chromogen. Nuclei were stained with haematoxylin (Sigma). Positive stained cells exhibited brownish precipitate in the cytoplasm. Coverslips were mounted using DPX mounting medium (Sigma Aldrich Co, USA) and slides were observed using confocal laser scanning microscope (LSM-510, Zeiss).

Corneal fibroblasts underwent centripetal migration within 3 h post wound creation. Cells at the wound edge changed morphology to become spindle-shaped cells resembling active fibroblast (Fig. 2).

Cell migration study signifies the centripetal migration of the corneal fibroblast in closing the wound. The cells cultured in AH-enriched FD and FDS media showed higher percentage of wound closure compared to their respective controls (Fig. 3). At day 3, cells cultured in AH-enriched FD medium showed a 27.4 % reduction in the wound size compared to 19.5 % for cells cultured in FD medium alone (Fig. 3b, e and 4). Cells cultured in AH-enriched FDS medium reduced the wound area to 49.9 % compared to that of 44.5 % for cells cultured in FDS medium alone (Fig. 3h, k and 4). At day 6, the migration of cells were significantly slower in FD medium compared to AH-enriched FD medium which showed 30.8 and 39.5 % wound closure respectively (Fig. 3c, f and 4). Cells cultured in AH-enriched FDS medium showed complete wound closure at day 6 (Fig. 3l), compared to only 81.5 % wound closure in the FDS medium alone (Fig. 3i and 4). These results indicate that AH at the concentration of 0.025 % has significant effect in accelerating the migration of the cultured corneal fibroblasts of the in vitro corneal ulcer model.

The differentiation and wound healing markers were evaluated by gene expression study. The expression of ALDH was the highest at the initial day of wound creation in all groups (Fig. 5a). At day 3, the expression of ALDH reduced significantly in FDS and AH-enriched FDS media compared to FD and AH-enriched FD media, respectively. At day 6, corneal fibroblasts cultured in AH-enriched media significantly reduced the expression of ALDH compared to their respective controls. In contrast, vimentin showed increasing expression pattern from the initial day of wound creation until day 6 in all groups (Fig. 5b). Cells cultured in AH-enriched FDS media showed the highest vimentin expression compared to FDS medium alone.

Cells showed increasing trend of α-SMA expression, compared to the initial day of wounding in all groups (Fig. 5c). At day 6, corneal fibroblasts cultured in FDS medium showed higher expression of α-SMA compared to FD medium. However, the expression of α-SMA was significantly reduced in AH-enriched FDS media. The expression of collagen type I gene was up-regulated at day 3 and day 6 in comparison to the initial day of wound creation in all groups (Fig. 5d). Corneal fibroblasts cultured in AH-enriched FDS medium showed the highest expression of collagen type I compared to control medium at day 6 of wound creation.

The expression of lumican was up-regulated for cells cultured in all media (Fig. 5e). At day 6, expression of lumican increased significantly (p < 0.05) when corneal fibroblasts were cultured in AH-enriched media compared to their respective controls. Gene expression of MMP12 was the lowest at the initial day of wound creation in all media (Fig. 5f). At day 3, MMP12 gene expression was up-regulated, especially in cells cultured in AH-enriched FDS medium compared to FDS medium alone. At day 6, corneal fibroblasts cultured in FDS groups showed reduction in the gene expression of MMP12 compared to the FD groups.

Gel electrophoresis showed specific product size with the presence of single band for gene expression of ALDH, vimentin, α-SMA, collagen type I, lumican and MMP12 in corneal fibroblasts cultured with or without supplementation of 0.025 % AH (Fig. 6).

The protein expression during wound healing was ascertained through immunohistochemistry. Abundant positive brown stained cells for ALDH protein was detected at the wound edge at the initial day of wounding in all groups (Fig. 7a, d, g, j). The protein expression was reduced during day 3 (Fig. 7b, e h, k) and day 6 in AH-enriched media compared to their respective controls (Fig. 7c, f, i, l).

Vimentin was highly expressed throughout the study in all culture media (Fig. 8). The density of positive stained cells was increased in cell cultured in FD (Fig. 8b, e, c, f) and FDS (Fig. 8h, k i, l) media supplemented with 0.025 % AH compared to control media at day 3 and day 6 post wounding.

At the initial day of wounding, α-SMA protein expression was reduced in all groups (Fig. 9a, d, g, j). At day 3 and day 6, α-SMA protein expression were increased especially in corneal fibroblasts cultured in AH-enriched media (Fig. 9e, f, k, l) compared to controls, respectively (Fig. 9b, c h, i), indicating the presence of myofibroblasts phenotype in the wound. All the immunocytochemistry results were in agreement with the gene expression analyses.