Tissue Damage Caused by Myeloablative, but Not Non-Myeloablative, Conditioning before Allogeneic Stem Cell Transplantation Results in Dermal Macrophage Recruitment without Active T-Cell Interaction

Abstract

Introduction: Conditioning regimens preceding allogeneic stem cell transplantation (alloSCT) can cause tissue damage and acceleration of the development of graft-versus-host disease (GVHD). T-cell-depleted alloSCT with postponed donor lymphocyte infusion (DLI) may reduce GVHD, because tissue injury can be restored at the time of DLI. In this study, we investigated the presence of tissue injury and inflammation in skin during the period of hematologic recovery and immune reconstitution after alloSCT.

Methods: Skin biopsies were immunohistochemically stained for HLA class II, CD1a, CD11c, CD40, CD54, CD68, CD86, CD206, CD3, and CD8. HLA class II-expressing cells were characterized as activated T-cells, antigen-presenting cells (APCs), or tissue repairing macrophages. In sex-mismatched patient and donor couples, origin of cells was determined by multiplex analysis combining XY-FISH and fluorescent immunohistochemistry.

Results: No inflammatory environment due to pretransplant conditioning was detected at the time of alloSCT, irrespective of the conditioning regimen. An increase in HLA class II-positive macrophages and CD3 T-cells was observed 12-24 weeks after myeloablative alloSCT, but these macrophages did not show signs of interaction with the co-localized T-cells. In contrast, during GVHD, an increase in HLA class II-expressing cells coinciding with T-cell interaction was observed, resulting in an overt inflammatory reaction with the presence of activated APC, activated donor T-cells, and localized upregulation of HLA class II expression on epidermal cells. In the absence of GVHD, patient derived macrophages were gradually replaced by donor-derived macrophages although patient-derived macrophages were detectable even 24 weeks after alloSCT.

Conclusion: Conditioning regimens cause tissue damage in the skin, but this does not result in a local increase of activated APC. In contrast to the inflamed situation in GVHD, when interaction takes place between activated APC and donor T-cells, the tissue damage caused by myeloablative alloSCT results in dermal recruitment of HLA class II-positive tissue repairing macrophages co-existing with increased numbers of patient- and donor-derived T-cells, but without signs of specific interaction and initiation of an immune response. Thus, the local skin damage caused by the conditioning regimen appears to be insufficient as single factor to provoke GVHD induction.

Keywords: HLA class II; allogeneic stem cell transplantation; graft-versus-host disease; macrophages; skin; tissue damage.

Figure 1

Pretransplant conditioning did not result in a pro-inflammatory environment in the skin at the time of transplantation. (A) In normal skin, HLA class II-expressing cells and also T-cells are present scattered over the whole dermal region. In the epidermal layer, HLA class II-expressing cells were observed sporadically. (B) In skin affected by dermatitis medicamentosa, dermal area count of HLA class II-expressing cells and T-cells is higher compared to normal skin. After non-myeloablative (NMA) conditioning (C) and myeloablative (MA) conditioning (D), the presence of HLA class II-expressing cells and T-cells at the time of allogeneic stem cell transplantation (alloSCT) resembled the situation in normal skin. Dermal area counts with median were calculated for HLA class II-expressing cells (E) and CD3 T-cells (F) in skin biopsies of normal skin, skin at the time of alloSCT after NMA and MA conditioning, and skin affected by dermatitis medicamentosa. A significant difference was detected regarding HLA class II-expressing cells between normal skin and skin affected by dermatitis medicamentosa (Kruskal–Wallis test p = 0.019, post hoc Mann–Whitney U-test p = 0.016). White line demarks the border between dermis and epidermis.

Figure 2

Remarkable increase in HLA class II-expressing cells after myeloablative (MA) allogeneic stem cell transplantation (alloSCT). Calculated dermal area counts with medians are depicted. No increase in HLA class II-expressing cells (A) or T-cells (B) was observed after autoSCT. Also, no increase in HLA class II-expressing cells (C) or T-cells (D) was observed after non-myeloablative (NMA) alloSCT. In contrast, following MA, alloSCT there was a significant increase in HLA class II-expressing cells after 12 and 24 weeks compared to T = 0 (Kruskal–Wallis test p = 0.020 and post hoc Mann–Whitney U-test p = 0.03 and 0.008) (E) and a non-significant increase in T-cells after 24 weeks (Kruskal–Wallis test p = 0.12) (F). (G–I) Results of fluorescence microscopy of illustrative examples showing HLA class II- and CD3-expressing cells 24 weeks after autologous (G), NMA (H), or MA (I) transplantation. White line demarks the border between dermis and epidermis.

Figure 3

Biopsies from patients with acute skin graft-versus-host disease (GVHD) showed massive inflammation. Both dermal area count (excluding by definition the epidermal area count) of HLA class II-positive cells (A) and CD3-positive cells (B) were not significantly different between biopsies taken during acute skin GVHD and biopsies of normal skin taken 24 weeks after myeloablative (MA) allogeneic stem cell transplantation (alloSCT). However, immunofluorescence microscopy showed honeycomb pattern of HLA class II expression on epidermal cells and upregulation of HLA class II expression on T-cells (yellow hued cells) during GVHD as signs of inflammation (C), while in normal skin biopsies 24 weeks after MA alloSCT, co-localization of HLA class II-positive cells and T-cells without interaction was observed (D). Pink hued cells represent CD8-positive CD3 T-cells. White line demarks the border between dermis and epidermis.

Figure 4

Dermal HLA class II-expressing cells seen after myeloablative (MA) allogeneic stem cell transplantation (alloSCT) were macrophages with non-professional antigen-presenting cell (APC) phenotype. (A) In normal skin, the majority of HLA class II-expressing cells were macrophages, as indicated by the expression of CD68 and CD206, while also some resident dermal dendritic cells (CD11c positive) were detected. The HLA class II-expressing cells were negative for CD40, CD54, and CD86. (B) Skin biopsies taken from patients in which an increase in HLA class II-positive cells was observed after MA alloSCT. HLA class II-expressing cells were CD68- and CD206-positive macrophages from which only a few expressed also CD11c. There was no expression of CD40, CD54, or CD86, so the HLA class II-expressing cells did not have the characteristics of activated professional APC. (C) Skin biopsies from skin affected by acute graft-versus-host disease (GVHD) showed increase in HLA class II-expressing cells with professional inflammatory APC phenotype (CD11c, CD40, CD54, and CD86 positive, while CD206 is negative). White line demarks the border between dermis and epidermis.

Figure 5

Gradual replacement of cells from recipient origin by cells of donor origin. (A) Percentage of recipient origin HLA class II-positive cells in the skin declined over time after allogeneic stem cell transplantation (alloSCT), but remained present even after 24 weeks. After autologous stem cell transplantation (autoSCT), no mixed chimera were observed, and in biopsies taken at the time of graft-versus-host disease (GVHD), the majority of HLA class II-positive cells were from donor origin. (B) Skin biopsy taken 3 weeks after alloSCT from a male patient with a female donor. All epithelial cells were XY, and nearly all HLA class II-positive cells were XY although some HLA class II-positive cells were XX (white arrows). (C) Skin biopsy taken 24 weeks after alloSCT from a male patient with a female donor. All epithelial cells were XY and all HLA class II-positive cells in this part of the biopsy were XX (white arrows). (D) Big differences were observed in the skin biopsies regarding the chimerism of CD3-positive cells. (E) 3 weeks after alloSCT, nearly all CD3-positive cells were from patient origin, except those indicated with a white arrow (male patient with female donor). (F) At the time of GVHD all CD3-positive cells were from donor origin (infiltrate of male donor CD3-positive cells in the skin of a female patient). White line demarks the border between dermis and epidermis.