An acute negative bystander effect of γ-irradiated recipients on transplanted hematopoietic stem cells

Abstract

Ultimate success of hematopoietic stem cell transplantation (HSCT) depends not only on donor HSCs themselves but also on the host environment. Total body irradiation is a component in various host conditioning regimens for HSCT. It is known that ionizing radiation exerts "bystander effects" on nontargeted cells and that HSCs transplanted into irradiated recipients undergo proliferative exhaustion. However, whether irradiated recipients pose a proliferation-independent bystander effect on transplanted HSCs is unclear. In this study, we found that irradiated mouse recipients significantly impaired the long-term repopulating ability of transplanted mouse HSCs shortly (∼ 17 hours) after exposure to irradiated hosts and before the cells began to divide. There was an increase of acute cell death associated with accelerated proliferation of the bystander hematopoietic cells. This effect was marked by dramatic down-regulation of c-Kit, apparently because of elevated reactive oxygen species. Administration of an antioxidant chemical, N-acetylcysteine, or ectopically overexpressing a reactive oxygen species scavenging enzyme, catalase, improved the function of transplanted HSCs in irradiated hosts. Together, this study provides evidence for an acute negative, yet proliferation-independent, bystander effect of irradiated recipients on transplanted HSCs, thereby having implications for HSCT in both experimental and clinical scenarios in which total body irradiation is involved.

Figure 1

Impaired long-term engraftment of bystander hematopoietic cells. (A) An overall experimental design for the bystander exposure in vivo. Recipient mice of IR received 10 Gy irradiation 1 day before transplantation. (B) A cell proliferation assay. Lin⁻c-Kit⁺CFSE⁺ cells from 20 pooled mice were sorted and injected into IR or NR recipients. Seventeen hours after transplantation, the transplanted cells were analyzed for their divisions according to the intensity of CFSE by flow cytometry. Data presented were obtained from 1 of the 2 experiments with identical results. (C) Engraftment levels at different time points in blood. Homed Lin⁻c-Kit⁺ donor cells from IR and NR recipients were sorted 17 hours after transplantation and transplanted into lethally irradiated congenic recipients (1-2 × 10⁴ cells/mouse) in the cBMT model. Engraftment levels of donor cells at different time points after transplantation were represented by the frequencies of donor cells in blood. **P < .01. n = 6. This is a representative summary from 3 experiments with consistent results. (D-E) Representative ratios in blood (D) and BM (E) 20 weeks after the transplantation. The ratios of CD45.1⁺ (IR) to CD45.2⁺ (NR) in blood at whole or different lineages (D), and in BM at whole or different hematopoietic subsets (E) are shown. The figures show the representative data from 1 of 3 mice.

Figure 2

Normal homing and localization of transplanted hematopoietic cells in recipient BM. (A) Frequencies and total numbers of homed BMNCs in recipient BM. BM cells pooled from 3 mice (CD45.1⁺) were injected into IR or NR recipients (CD45.2⁺). Seventeen hours after transplantation, BM cells were harvested and stained with anti-CD45.1 and CD45.2. The frequency of donor cells was then measured by flow cytometry. Total absolute number of donor cells was also calculated. **P < .01. n = 4. (B) Localization of the homed cells. CFSE-labeled Lin⁻c-kit⁺ cells were injected into IR or NR recipients that were scarified 17 hours after transplantation. The homed cells were identified on the sections of femurs based on CFSE staining with fluorescence microscopy. Images show the cells under a 490/20-nm filter (green, left), 555/28-nm filter (red, center column) and their overlay (right). Arrows indicate homed cells. The top panel was from a NR recipient, and the bottom panel was from an IR recipient. The photograph was taken under 40×/0.60 by a digital camera (RT Slider-Spot, model 2.3.1.1.) with Spot Version 4.0.9 software (Diagnostic Instruments). Adobe Photoshop CS Version 8.0 (Adobe System) and Microsoft Office PowerPoint 2003 software were used for image sizing and displaying. (C) Enumeration of the homed cells in proximity to the endosteal niche. Total homed cells from NR and IR recipients per section were counted, and the lodgment was quantified by the proportion of homed cells located within the endosteal region. **P < .01. n = 20 sections.

Figure 3

Increased apoptosis and proliferation without overt senescence of bystander hematopoietic cells. (A) Apoptosis of the resided hematopoietic cells. Apoptosis was measured in homed Lin⁻Sca-1⁺ cells with multicolor flow cytometry combining surface markers of CD45.1, CD45.2, lineage, Sca-1 with annexin V, and DAPI. Plots were gated on homed donor Lin⁻Sca-1⁺ cells (left). Percentage of early apoptosis (annexin V⁺DAPI⁻) and late apoptosis was compared between NR and IR recipients (right). *P < .05. n = 3 or 4. (B) Cell proliferative potential. Transplanted BMNCs were labeled with CFSE before injection. Three days later, the division of donor Lin⁻ cells in NR or IR recipients was analyzed according to the intensity of CFSE in flow cytometry. The representative plots were generated from ModFit Version 3.1 software. (C) Senescence of the bystander cells from the irradiated hosts. Homed Lin⁻Sca-1⁺ cells from NR and IR were sorted and stained with X-gal for detecting the β-gal activity. Blue color in the cells indicated positive activity of β-gal. Lin⁻Sca-1⁺ cells cultured in the medium containing 50 ng/mL of stem cell factor, 50 ng/mL of Flt3, and 10 ng/mL of thrombopoietin at 37°C, 5% CO₂ for 10 days were used as positive control. Real-time RT-PCR was performed for examining the mRNA expression of p16^INK4A and p19^ARF in homed cells 17 hours after transplantation. M indicates DNA ladder. Lanes 1 and 2 represent homed Lin⁻Sca-1⁺ cells from IR recipients; lanes 3 and 4, homed Lin⁻Sca-1⁺ cells from NR recipients; and lanes 5 and 6, positive controls.

Figure 4

Altered expression of stem cell regulators in the bystander hematopoietic cells. (A-B) Relative expression of genes in the cells from IR or NR recipients. Homed Lin⁻Sca-1⁺ cells were sorted from IR or NR recipients, and mRNA expression of selected genes was examined with real-time RT-PCR. (A) Relative expression of the genes in cells from IR or NR recipients compared with unmanipulated cells. (B) Direct comparison of the gene expression between the cells isolated from IR and NR recipients; n = 3. The primer sequences for those genes are listed in supplemental Table 1. (C) Down-regulation of c-Kit in the IR host. c-Kit expression was measured by flow cytometry 17 hours after BMT. The plots were gated on the donor Lin⁻ population. Bar graphs represents expression of the c-Kit⁺ fraction in total CD45⁺ cells homing to BM. **P < .01. n = 2 to 4.

Figure 5

Increased level of inflammatory cytokines in irradiated BM cells. (A) Cytokine antibody array analysis of irradiated and nonirradiated BM cells. Cell lysates of BM cells were collected from 10 Gy IR mice (n = 3) and NR mice 17 hours after treatment. Cell lysates were sent for membrane-based mouse cytokine antibody array. The cytokines increased more than 2-fold in irradiated cells compared with nonirradiated controls. Lower panels: Representative membrane image. (B) Increased level of activated MMP9 in irradiated BM cells. Cell lysates were also used for activated MMP9 measurement using ELISA. The fold change in irradiated BM compared with nonirradiated BM lysates is shown based on 3 experiments. (C) Down-regulation of c-Kit expression. C-kit-enriched BM cells were injected into lethally IR or NR MMP9^+/+ or MMP9^−/− mice, respectively. Mice were killed at 17 hours after transplantation, and the expression of c-Kit in the donor cells was measured by flow cytometry. Panels represent the representative flow charts gated in the donor Lin⁻ cells.

Figure 6

Increased generation of ROS in the bystander cells and the effects of NAC. (A) An elevated level of ROS in bystander cells. Lin⁻Sca-1⁺ BM cells transplanted into IR or NR mice were collected 17 hours later and subjected to ROS measurement. Top panel: Representative flow chart. Bottom panel: Fold change of mean fluorescence intensity (MFI).**P < .01. n = 3. (B-C) Down-regulation of c-Kit expression at the protein and transcript levels by exposure to hydrogen peroxide. (B) c-Kit–enriched BM cells were exposed to different doses of hydrogen peroxide. Cells collected at different time points after exposure were subjected to c-Kit analysis by flow cytometry or RT-PCR. The relative expression of c-Kit at the protein level (B) and the transcript level (C) compared with nonexposed controls was plotted. (D) An elevated level of ROS reversed by administration of NAC in vitro. Lin⁻c-kit⁺ cells plated onto an IR (12 Gy) or NR stromal cell line (AFT024) layer treated with or without NAC were collected 40 hours later. The intracellular level of ROS was compared among groups. **P < .01. n = 3. (E) Improved reconstitution of bystander hematopoietic cells by administration of NAC in vivo. Homed Lin⁻c-kit⁺ cells in NR or IR recipients treated with or without NAC were sorted 17 hours after transplantation and retransplanted into lethally irradiated congenic recipients (3000 cells/mouse). Engraftment ratios of donor cells from IR recipients to that from NR recipients were calculated; n = 3. *P < .05. n = 4.

Figure 7

Improved engraftment of bystander hematopoietic cells by ectopically expressing catalase. (A) A schematic illustration of retrovirus generation, transduction into hematopoietic cells, and transplantation. (B) Improved engraftment by ectopically expressing catalase in cBMT assay. Catalase (Cat) overexpressed HSCs showed higher engraftment potential than that of control vector transduced cells. Data are mean ± SEM of relative ratios of catalase to control in the peripheral blood of 3 independent transplantation experiments. *P < .05. n = 19 in the vector group; n = 13 in the Cat group (3 experiments in total).