Phagocytosis of apoptotic endothelial cells reprograms macrophages in skin wounds

Abstract

In healing wounds, the regression of blood vessels during the resolution phase creates a significant number of apoptotic endothelial cells (ApoECs). Surprisingly few studies have investigated the fate of apoECs in wounds, or the consequence of their removal. The current study employed both in vitro and in vivo models to investigate if macrophages ingest apoECs and to determine if such phagocytosis alters macrophage phenotype. To examine the capability of macrophages to ingest apoECs in in vivo wounds, pHrodo green labeled apoECs were injected into skin wounds 6 days after injury. The results demonstrated that 2.2% of macrophages in the wounds had engulfed apoECs 24 hours after injection. Macrophages that had engulfed apoECs expressed the markers CD80 (100%), CD86 (93.8%), and CD163 (22.8%), while no expression of CD206 marker was observed. In in vitro studies, 76.1% and 81.1% of PMA differentiated THP-1 macrophages engulfed apoECs at 6 and 24 hours, respectively. mRNA expression levels of IL-1β, iNOS, and TGF-β1 decreased in THP-1 macrophages after exposure to apoECs, while the expression of IL-6 increased. THP-1 macrophages that were incubated with apoECs for 6hours expressed CD80 (30.2%), CD163 (62.9%), and CD206 (45.3%), while expression levels in untreated group were 0.5%, 45.0%, and 2.4%, respectively. Taken together, our studies showed that macrophages phagocytize dermal apoECs both in vitro and in vivo. The engulfment of apoECs leads to a unique macrophage phenotype, which has characteristics of both M1 and M2 macrophage phenotypes. These findings provide a new mechanism by which macrophage phenotypes can be modified during wound resolution.

Keywords: apoptosis; endothelial cell; macrophage; phagocytosis; skin; wound healing.

Figure 1:. Macrophages ingest apoptotic human dermal ECs in vitro.

Serum starvation-induced apoptotic GFP-hECs were added to PMA-induced THP-1 macrophages for 6 and 24 hours. A) Flow cytometry scatter plots show percent of THP-1 macrophages engulfing apoptotic GFP-hECs. B) Confocal immunofluorescence images show macrophages (CD44+) ingesting apoptotic GFP-hECs indicated by white arrows. Representative results from two experiments are shown. Gating strategies are shown in Appendix B.

Figure 2:. Exposure of macrophages to apoECs induces macrophage phenotype changes in vitro.

Serum starvation-induced apoptotic GFP-hECs or hECs were incubated with PMA-induced THP-1 macrophages for 6 and 24 hours. A) CD80, CD163 and CD206 expression on apoptotic GFP-hECs treated THP-1 macrophages by flow cytometry analysis. Cells were pooled from three wells of a 6 well plate at each time point. The number shown in the upper right on each scatter plot is the percentage of the target on F4/80 positive macrophages. Gating strategies are shown in Appendix B. B). mRNA expression of TNF-α, IL-1β, IL-6, IL-10, iNOS, TGF- β1, VEGF-A, and IGF-1 in apoptotic hECs treated THP-1 cells. The level of relative expression for each target at zero hour without treatment was used as the baseline control. N=3 triplicate wells at each time point. Representative results from two experiments are shown. Two-way ANOVA followed by Bonferroni’s post hoc test was used for statistical analysis. *p<0.05, **p<0.01

Figure 3:. Macrophages ingest apoptotic ECs in wounds in vivio.

pHrodo green labeled apoptotic mECs were injected into day 6 mouse skin wounds. The uptake of apoptotic mECs 24 hours after injection was examined. A) Images taken using a regular fluorescence microscope. B) Confocal microscopy images. The white arrow in Z stack images at the left panel indicates a macrophage with engulfed pHrodo green labeled apoptotic mEC. The 3D image at the right panel shows the pHrodo green labeled apoptotic mECs are embedded among red stained CD68 macrophages suggesting macrophages ingested apoptotic mECs. C) Cells were isolated from the wounds on day 7 and subjected to flow cytometry to identify F4/80 macrophages that had ingested apoptotic mECs. Cells were collected from 8 pooled wounds from 4 mice (two wounds from each mouse) in each group. The number shown in the right scatter plot is the percentage of F4/80 macrophages with engulfed pHrodo green labeled apoptotic mECs. Representative results from two experiments are shown. Gating strategies are shown in Appendix C

Figure 4:. Exposure to apoptotic dermal ECs alters macrophages phenotypes in vivo.

pHrodo green labeled apoptotic mECs were injected into day 6 skin wounds. Eight wounds were harvested and pooled from 4 mice (two wounds from each mouse) in each group24 hours later. Single cell suspensions were prepared for CD80, CD86, CD163, and CD206 staining and then subjected to flow cytometry analysis. A) Expression of CD80, CD86, CD163, and CD206 on F4/80 macrophages. B) Expression of CD80, CD86, CD163, and CD206 on F4/80 macrophages with engulfed apoptotic mECs. The number showed in the upper right on each scatter plot is the percentage of the target marker on F4/80 positive macrophages. Representative results from two experiments are shown. Gating strategies are shown in Appendix C.

Figure 5:. mRNA expression of pro- and anti-inflammatory makers and growth factors in apoptotic mECs treated wounds.

A) mRNA expression of pro-inflammatory/M1 markers (IL-1β, IL-6, iNOS, and TNF-α) in day 7 wound tissue 24 hours after apoptotic mECs injection. B) mRNA expression of anti-inflammatory markers and growth factors (IL-10, TGF- β1, IGF-1, and VEGF-A) in day 7 wound tissue 24 hours after apoptotic mECs injection. The level of relative expression of the individual target in the wounds treated with PBS was used as the baseline control. N=3 mice for each treatment group, t test was used for statistical analysis,*p<0.05.

Figure 6:. Inhibition of angiogenesis leads to macrophages phenotype changes in wounds.

rPEDF or PBS control was used to treat 6 mm skin wound to up to 9 days by combination of topical administration and intradermal injection. Wounds were harvested on day 7 and 10 after wounding and treatment. Single cell suspensions were prepared from the tissues as described in Materials and Methods. Cell surface expression of CD80, CD86 and CD163 on wound F4/80 macrophages was examined by flow cytometry. The cells were collected from 16 pooled wounds (4 mice per group at each time point). The number showed in the upper right on each scatter plot is the percentage of the target on F4/80 positive macrophages. Representative results from two experiments are shown. Gating strategies are shown in Appendix E.