Amniotic membrane traps and induces apoptosis of inflammatory cells in ocular surface chemical burn

Abstract

Purpose: Severe chemical burns can cause necrosis of ocular surface tissues following the infiltration of inflammatory cells. It has been shown that amniotic membrane transplantation (AMT) is an effective treatment for severe chemical burns, but the phenotypes of cells that infiltrate the amniotic membrane and the clinical significance of these cellular infiltrations have not previously been reported. The present work studies the inflammation cell traps and apoptosis inducing roles of the amniotic membrane after AMT in patients with acute chemical burns.

Methods: A total of 30 patients with acute alkaline burns were classified as having either moderate or severe burns. In all participants, AMT was performed within one week of his/her injury. After 7-9 days, the transplanted amniotic membranes were removed. Histopathological and immunohistochemical techniques were used for the examination and detection of infiltrating cells, and tests for the expression of CD (cluster of differentiation)15, CD68, CD3, CD20, CD57, CD31, CD147, and CD95 (Fas) were performed. A TUNEL (TdT-mediated dUTP nick end labeling) assay was used to confirm apoptosis of the infiltrating cells. Three patients with herpes simplex-induced keratitis who had undergone AMT to treat persistent epithelium defects were used as a control group. Amniotic membrane before transplantation was used as another control.

Results: After amniotic membrane transplantation, the number of infiltrating cells in patients with severe burns was significantly higher than in patients with moderate burns or in control patients (p<0.05). Among the severe burns patients, CD15 and CD68 were widely expressed in the infiltrating cells, and CD3, CD20, and CD57 were only found in a small number of cells. Occasionally, CD31-positive cells were found in the amniotic membranes. More cells that were CD147, Fas, and TUNEL positive were found in patients with severe burns than in patients with moderate burns or in control patients.

Conclusions: Neutrophils and macrophages were the main cells that had infiltrated into the amniotic membrane during the acute phase of healing from a chemical burns. AMT can trap different inflammatory cells and induce apoptosis of inflammatory cells in acute ocular chemical burns.

Figure 1

Hematoxylin and eosin staining of ammiotic membrane stromal infiltrating cells in patients with chemical burns. In the severe burns group, there were many neutrophils in the amniotic membrane (A). In the moderate burns group, there were clearly fewer neutrophils in the amniotic membrane than in the severe burns group (B). In the control HSK group, there were only a few neutrophils on the surface of the amniotic membrane (C). In the amniotic membrane before transplantation, no neutrophils were found (D). Scale bar: 20 μm.

Figure 2

The expression of CD15, CD68, CD3, CD20, CD57, and CD31 in the severe burn group. In the severe burns group, CD15- and CD68-positive cells were the main types of cells that infiltrated the stroma of the amniotic membrane (A and B). CD3-, CD20-, or CD57-positive cells could also be found in the amniotic membrane (C, D, and E). A few CD31-positive cells could also be found in the amniotic membrane (F). Scale bar: 20 μm.

Figure 3

The expression of CD15, CD68, CD3, CD20, CD57, and CD31 in the control HSK group. In the control HSK group, CD15- and CD68-positive cells were the main types of cells that infiltrated the stroma of the amniotic membrane (A and B). No CD3-, CD20-, CD57- or CD31-positive cells could be found in the amniotic membrane (C, D, E, and F). These cell markers could not be found in the amniotic membrane before transplantation (G). Scale bar: 20 μm.

Figure 4

CD147 staining of amniotic membrane stromal infiltrating cells in patients with chemical burns. In the severe burns group, there were many CD147-positive cells in the amniotic membrane (A). In the moderate burns group, CD147-positive cells were only observed in the bottom of the amniotic membrane (B). In the control HSK group, the cells of the amniotic membrane were CD147-negative (C). In the amniotic membrane before transplantation, the epithelium cells of the amniotic membrane were CD147-positive (D). Scale bar: 20 μm.

Figure 5

CD95 (Fas) staining of amniotic membrane stromal infiltrating cells in patients with chemical burns. In the severe burns group, there were many Fas-positive cells in the amniotic membrane (A). In the moderate burns group, Fas-positive cells were observed in the bottom of the amniotic membrane (B). In the control HSK group, a few Fas-positive cells were observed in the surface of the amniotic membrane (C). In the amniotic membrane before transplantation, the epithelium cells of the amniotic membrane were Fas-positive (D). Scale bar: 20 μm.

Figure 6

TUNEL staining of amniotic membrane stromal infiltrating cells in patients with chemical burns. In the severe burns group, there were many TUNEL-positive cells in the amniotic membrane (A). In the moderate burns group, TUNEL-positive cells were observed in the stroma of the amniotic membrane (B). In the control HSK group, a few TUNEL-positive cells were observed in the stroma of the amniotic membrane (C). In the amniotic membrane before transplantation, a few epithelium cells of the amniotic membrane were TUNEL-positive (D). Scale bar: 20 μm.

Figure 7

Evaluation of amniotic membrane stromal infiltrating cells in patients with chemical burns. Significant differences were found between the severe burns group and the moderate burns group (p<0.05) in (A) the number of inflammation-related cells per high-power field, (B) the number of CD147-positive cells per high-power field, (C) the number of Fas-positive cells per high-power field, and (D) the number of TUNEL-positive cells per high-power field.