Characterization of Recruited Mononuclear Phagocytes following Corneal Chemical Injury

Abstract

Mononuclear phagocytes (MP) have central importance in innate immunity, inflammation, and fibrosis. Recruited MPs, such as macrophages, are plastic cells and can switch from an inflammatory to a restorative phenotype during the healing process. However, the role of the MPs in corneal wound healing is not completely understood. The purpose of this study is to characterize the kinetics of recruited MPs and evaluate the role of macrophage metalloelastase (MMP12) in the healing process, using an in vivo corneal chemical injury model. Unwounded and wounded corneas of wild-type (WT) and Mmp12^-/- mice were collected at 1, 3, and 6 days after chemical injury and processed for flow cytometry analysis. Corneal MP phenotype significantly changed over time with recruited Ly6C^high (proinflammatory) cells being most abundant at 1 day post-injury. Ly6C^int cells were highly expressed at 3 days post-injury and Ly6C^neg (patrolling) cells became the predominant cell type at 6 days post-injury. CD11c⁺ dendritic cells were abundant in corneas from Mmp12^-/- mice at 6 days post-injury. These findings show the temporal phenotypic plasticity of recruited MPs and provide valuable insight into the role of the MPs in the corneal repair response, which may help guide the future development of MP-targeted therapies.

Keywords: MMP12; corneal injury; dendritic cells; macrophages; monocytes; mononuclear phagocytes; myeloid cells.

Figure 1

Representative gating strategy of flow cytometry used to identify recruited and resident leukocyte populations and kinetics in uninjured and injured corneas after corneal chemical injury. Corneas from wild type (WT) and Mmp12^−/− (KO) mice at 8–12 weeks old, were collected at 1, 3, and 6 days after an alkaline burn was created by applying filter paper soaked in 0.1 N NaOH. Four corneas from 2 mice were pooled per sample, cut into pieces, and digested with collagenase type I for 1.5 h at 37 °C. Cells were washed with FACS buffer and then incubated for 30 min in a FACS buffer solution containing CD16/32 Fc block. Cells were then washed with FACS buffer and stained for 30 min with the following antibodies: CD45-FITC, CD11B-Pacific Blue, F4/80-APC, Ly-6C-PerCP/Cy5, CD64-PE-Dazzle, CD11c-BrilliantViolet 605, Zombie-Acqua Viability Dye. Cells were washed and then fixed with 2% PFA in PBS overnight (at 4 °C). Cell profiles were acquired on an LSRII flow cytometer (Becton Dickinson, Franklin Lakes, NJ, USA), and were gated based on fluorescence minus one control. The gating strategy is illustrated: Cells → Live → Singlets → CD45^high leukocytes (L). Fifty-six corneas were tested in each group (WT and KO); controls (n = 3 samples; 12 corneas); day 1 (n = 4 samples; 16 corneas); day 3 (n = 4 samples, 16 corneas); day 6 (n = 3 samples; 12 corneas). CD11b⁺F4/80^high resident mononuclear phagocytes (R2) can be seen in the uninjured samples. CD11b⁺F4/80^intermediate recruited mononuclear phagocytes (R1) and other CD11b⁺F4/80⁻ leukocytes (O) are seen mainly in day 1 after the chemical injury.

Figure 2

Corneal leukocyte dynamics after chemical injury. Corneas from wild type (WT) and Mmp12^−/− (KO) mice at 8–12 weeks old, collected at 1, 3, and 6 days after an alkaline burn was created by applying filter paper soaked in 0.1 N NaOH. Four corneas from 2 mice were pooled per sample, day 1 (n = 4 samples; 16 corneas); day 3 (n = 4 samples, 16 corneas); day 6 (n = 3 samples; 12 corneas). All data show as mean ± SEM percentage of corneal leukocytes (L). (A) The frequency of CD11b⁺F4/80^intermediate recruited mononuclear phagocytes (R1) significantly changed over time (two-way ANOVA p = 0.0002) and was statistically significantly higher on day 1 after the chemical injury in WT and KO mice in comparison to other time points. (B) The percentage of CD11b⁺F4/80^high resident mononuclear phagocytes (R2) changed over time (two-way ANOVA p < 0.0001) and they were more frequent in uninjured corneas. (C) CD45⁺CD11b⁺F4/80⁻ (O) frequency changed over time (two-way ANOVA p < 0.0001) with a significantly higher percentage observed on day 1 after injury. The differences between WT and KO leukocyte populations were not statistically significant. Tukey’s multiple comparison-adjusted p-value * <0.05; ** <0.01; *** <0.001; **** <0.0001.

Figure 3

Dynamics of Ly6C expression in recruited mononuclear phagocytes (R1) after corneal chemical injury. Corneas from wild type (WT) and Mmp12^−/− (KO) mice at 8–12 weeks old, collected at 1, 3, and 6 days after an alkaline burn was created by applying filter paper soaked in 0.1 N NaOH. Four corneas from 2 mice were pooled per sample, day 1 (n = 4 samples; 16 corneas); day 3 (n = 4 samples, 16 corneas); day 6 (n = 3 samples; 12 corneas). Cell profiles were acquired on an LSRII flow cytometer (Becton Dickinson, Franklin Lakes, NJ, USA), and were gated based on fluorescence minus one control. All data are shown as the mean ± SEM. (A) Based on Ly6C-PerCp-Cy5 staining, CD11b⁺ F4/80-APC ^intermediate recruited mononuclear phagocytes (R1) were grouped into 3 groups: Ly6C negative (L1), Ly6C intermediate (L2), Ly6C high (L3). (B,C) L1, L2, L3 percentage fractions over time in WT and KO mice, and average percentage of each group are presented. (D) Average percentage of L1 significantly changed over time (two-way ANOVA, p-value < 0.0001) and is higher on day 6 in both WT and KO mice. (E) Average percentage of L2 changed over time (two-way ANOVA, p-value = 0.0023) and was significantly higher on day 3 than day 6 in both WT and KO mice. (F) Average percentage of L3 changed over time (two-way ANOVA, p < 0.0001) and was significantly higher on day 1 in both WT and KO mice. The differences between WT and KO MP populations, based on Ly6C expression, were not statistically significant. Tukey’s multiple comparison-adjusted p-value * <0.05; *** <0.001; **** <0.0001.

Figure 4

Dynamics of recruited monocytes and dendritic cells after corneal chemical injury. Corneas from wild type (WT) and Mmp12^−/− (KO) mice at 8–12 weeks old, were collected 1,3 and 6 days after an alkaline burn was created by applying filter paper soaked in 0.1 N NaOH. Four corneas from 2 mice were pooled per sample, day 1 (n = 4 samples; 16 corneas); day 3 (n = 4 samples, 16 corneas); day 6 (n = 3 samples; 12 corneas). Cell profiles were acquired on an LSRII flow cytometer (Becton Dickinson, Franklin Lakes, NJ, USA), and were gated based on fluorescence minus one control. All data are shown as the mean ± SEM. (A) Representative gating strategy of flow cytometry used to identify dendritic cells (D), macrophages (Ma), and monocytes (Mo) within the recruited mononuclear phagocyte (R1) population. (B) Average percentage of CD11b⁺ F4/80 ^intermediate CD11c⁻ CD64⁺ macrophages (Ma) was significantly lower in KO mice on day 6 when compared to days 1 and 3. (C) Average percentage of CD11b⁺ F4/80 ^intermediate CD11c⁻ CD64⁻ monocytes (Mo) was significantly higher on day 1 in both WT and KO mice. (D) Average percentage of CD11b⁺ F4/80 ^intermediate CD11c⁺ dendritic cells (D) was significantly higher in KO mice on day 6 when compared to day 1. Tukey’s multiple comparison-adjusted p-value * <0.05; ** <0.01.

Figure 5

Dynamics of Ly6C expression in recruited macrophages (Ma) after corneal chemical injury. Corneas from wild type (WT) and Mmp12^−/− (KO) mice at 8–12 weeks old were collected at 1,3 and 6 days after an alkaline burn was created by applying filter paper soaked in 0.1 N NaOH. Four corneas from 2 mice were pooled per sample, day 1 (n = 4 samples; 16 corneas); day 3 (n = 4 samples, 16 corneas); day 6 (n = 3 samples; 12 corneas). Cell profiles were acquired on an LSRII flow cytometer (Becton Dickinson, Franklin Lakes, NJ, USA), and were gated based on fluorescence minus one control. All data are shown as the mean ± SEM. (A) Based on Ly6C-PerCp-Cy5 staining, CD11b⁺F4/80^int CD11c⁻ CD64⁺ recruited macrophages (Ma) were grouped into 3 groups: Ly6C^neg (L1), Ly6C^int (L2), Ly6C^high (L3). (B,C) L1, L2, L3 percentage fractions over time in WT and KO mice, and average percentage of each group are presented. (D) Average percentage of L1 significantly changed over time (two-way ANOVA, p-value < 0.0001) and was higher on day 6 in both WT and KO mice. (E) Average percentage of L2 changed over time (two-way ANOVA, p-value = 0.0002) and was significantly higher on day 3 than on days 1 and 6 in both WT and KO mice. (F) Average percentage of L3 changes over time (two-way ANOVA, p < 0.0001) and was significantly higher on day 1 in both WT and KO mice. The differences between WT and KO MP populations, based on Ly6C expression, were not statistically significant. Tukey’s multiple comparison-adjusted p-value * <0.05; **<0.01; **** <0.0001.