Modulating endothelial cells with EGFL7 to diminish aGVHD after allogeneic bone marrow transplantation in mice

Abstract

Acute graft-versus-host disease (aGVHD) is the second most common cause of death after allogeneic hematopoietic stem cell transplantation (allo-HSCT), underscoring the need for novel therapies. Based on previous work that endothelial cell dysfunction is present in aGVHD and that epidermal growth factor-like domain 7 (EGFL7) plays a significant role in decreasing inflammation by repressing endothelial cell activation and T-cell migration, we hypothesized that increasing EGFL7 levels after allo-HSCT will diminish the severity of aGVHD. Here, we show that treatment with recombinant EGFL7 (rEGFL7) in 2 different murine models of aGVHD decreases aGVHD severity and improves survival in recipient mice after allogeneic transplantation with respect to controls without affecting graft-versus-leukemia effect. Furthermore, we showed that rEGFL7 treatment results in higher thymocytes, T, B, and dendritic cell counts in recipient mice after allo-HSCT. This study constitutes a proof of concept of the ability of rEGFL7 therapy to reduce GHVD severity and mortality after allo-HSCT.

Figure 1.

Effect of EGFL7 on GVHD severity in B6→B6D2F1 mice. (A) Weights of animals that had undergone transplantation were measured daily and averaged for the group (green circles: syngeneic T cells [no GVHD]; red triangles: allogeneic treated with PBS; blue squares: allogeneic treated with rEGFL7). Daily EGFL7 treatment was initiated at day +21 post-BMT. Data were pooled from 3 experiments with 6 to 13 mice per group. (B) Survival curve of transplanted mice (dotted green line: syngeneic control [SYN], blue line: allogeneic treated with rEGFL7, and thin dotted red line: allogeneic treated with PBS). Data were pooled from 3 experiments with 6 to 13 mice per group. (C) Clinical scores of GVHD + PBS, GVHD + rEGFL7, and syngeneic mice at day +28 post-BMT. (D) Histopathology of the gut 28 days after allo-HSCT. Left: magnification ×200 and right: magnification ×400. (E) Gastrointestinal (GI) histopathology score of the gut of GVHD + rEGFL7 and GVHD + PBS-treated mice. Results show mean ± standard error of the mean (SEM). (F) Immunofluorescence analysis of the intestine from PBS or rEGFL7-treated mice transplanted with allogenic splenocytes that were stained for immunofluorescence. Cells were stained with CD31 (secondary antibody: donkey anti-goat alexa fluor 488), CD45 (secondary antibody: donkey anti-rabbit alexa fluor 647), and Ki-67 antibodies (secondary antibody donkey anti-rat alexa fluor 594). Nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI). Magnification ×400. (G) Cells were stained with CD3 (secondary antibody donkey anti-goat alexa fluor 488), CD4 (secondary antibody donkey anti-rabbit alexa fluor 647), and CD8 antibodies (secondary antibody donkey anti-rat alexa fluor 594). Nuclei were counterstained with 4′,6-diamidino-2-phenylindole. Magnification ×400. Data show 1 representative sample from each experimental group, which consists of 3 to 4 mice per group.

Figure 2.

Effect of EGFL7 on immune reconstitution and graft-versus-leukemia effect. (A-C) Splenocytes and thymocytes were collected from PBS-treated mice as they reached their endpoints (day +28 to 35) and from EGFL7-treated mice (day +30 to 35 post-BMT). (A) Absolute counts of TCR⁺, CD4⁺, and CD8⁺ lymphocytes in the spleen of Syn, GVHD + rEGFL7, and GVHD + PBS treated (B6→B6D2F1) mice. Data were pooled from 3 experiments with 4 to 7 mice per group. Gating was performed on CD45.1⁺ and CD45.2⁺ cells. (B) Absolute counts of BM-derived thymocytes (CD45.2⁺) in Syn, GVHD + rEGFL7, and GVHD + PBS treated (B6→B6D2F1) mice. Data were pooled from 3 experiments with 4 to 8 mice per group. Results show mean ± SEM. Statistical analysis compared GVHD + rEGFL7 vs GVHD + PBS treated (B6→B6D2F1) mice, and P values were determined by a Mann-Whitney U test. (C) Dot-plot analysis of thymocytes (derived from the BM cells [CD45.2⁺CD45.1⁻]), based on the expression of CD4, CD8, and CD45.2 antigens, in the thymus of SYN control, GVHD + rEGFL7, and GVHD + PBS mice. These data are representative of 3 or more experiments with 6 to 8 mice per group. Histogram showed mean ± SEM. Statistical analysis compared GVHD + rEGFL7 vs GVHD + PBS treated (B6→B6D2F1) mice, and P values were determined by a Mann-Whitney U test. Transplant for GVL was performed in BALB/c mice as described in “Methods.” (D-E) Whole-body bioluminescent signal intensity of recipient mice (n = 4-5 per cohort). Mice were imaged on indicated days. Average radiance expressed as mean ± SEM. One representative transplant experiment of 2 is shown. (F) Splenocytes were isolated at the time when mice reached their endpoints, days 28 to 33 post-BMT for PBS vehicle and days 30 to 35 post-BMT in rEGFL7-treated mice. Percentage GFP positivity representing P815 leukemic cell infiltration in the spleen. Each dot represents a single mouse. (G) Representative flow cytometric contour plots. Allo. spl., allogeneic splenocytes; GFP, green fluorescent protein; Min, minimum; Max, maximum; SSC-A, side-scatter-area; SYN, syngeneic control; TCD-BM, T cell depleted bone marrow. ***P < .001.