Biology of Blood and Marrow Transplantation
Volume 12, Issue 4 , Pages 397-407, April 2006

Graft-versus-Leukemia and Graft-versus-Host Reactions after Donor Lymphocyte Infusion Are Initiated by Host-Type Antigen-Presenting Cells and Regulated by Regulatory T Cells in Early and Long-Term Chimeras

Department of Pediatrics, Medical College of Wisconsin, and Children’s Research Institute, Children’s Hospital of Wisconsin, Milwaukee, Wisconsin

Received 7 July 2005; accepted 30 November 2005.

Article Outline

Abstract 

Regulatory T (Treg) cells and host antigen-presenting cells (APCs) have been implicated in graft-versus-host disease (GVHD) and the graft-versus-leukemia (GVL) effect after donor lymphocyte infusion (DLI), but their relative contributions remain unclear in early versus long-term complete donor or mixed murine allogeneic hematopoietic stem cell (HSC) chimeras. We have previously demonstrated that donor HSC-derived Thy1+ Treg cells, consisting primarily of CD4+CD25+ cells, play an important role in the suppression of graft-versus-host (GVH) reactivity when DLI is given to complete donor chimeras 28 days after HSC transplantation. Data presented here demonstrate that protection against GVHD exerted by Thy1+ Treg cells is less evident with time and eventually is not required in long-term complete donor chimeras because of an absence of host-type APCs to activate alloreactive T cells. Lethal GVHD was observed when Thy1+ Treg cells were depleted from complete donor chimeras given by DLI at day 28, 35, or 42; however, Treg cell depletion and DLI at day 70 no longer induced GVHD-associated mortality. Moreover, the failure of DLI to induce GVHD with Treg depletion correlated with a loss of the DLI-induced GVL effect in long-term (day 100) complete donor chimeras. In contrast to the results from complete donor chimeras, GVL reactivity in day 100 mixed chimeras was robust after DLI. Loss of a DLI-induced GVL effect in long-term complete donor chimeras was attributed to the absence of host APCs because the infusion of exogenous host-type dendritic cells partially restored both DLI-induced GVL and GVH reactions in day 100 complete donor chimeras. The GVL and GVH reactions restored by infusion of host dendritic cells in day 100 complete donor chimeras were at least partially regulated by Treg cells because transient depletion of CD25+ cells increased both the GVL effect and the severity of GVHD after DLI. Taken together, these data suggest that Treg cells can regulate DLI-induced GVL and GVH reactions in both early and long-term complete donor chimeras, and a state of mixed chimerism is superior to complete donor chimerism because host-type APCs facilitate a DLI-induced GVL effect without severe GVHD.

Key words:  Donor lymphocyte infusion , Graft-versus-leukemia effect , Graft-versus-host disease , Bone marrow transplantation , Regulatory T cells , CD4+CD25+ , Antigen-presenting cells , Dendritic cells

 

Back to Article Outline

Introduction 

Donor lymphocyte infusion (DLI) has become a standard form of immunotherapy for treating leukemia relapse after hematopoietic stem cell transplantation (HSCT) in humans. It is particularly effective for inducing remission in chronic myeloid leukemia patients but is less beneficial for other types of leukemia [1, 2, 3, 4, 5]. The ability of immune cells in the DLI inoculum to react against leukemia cells is referred to as the graft-versus-leukemia (GVL) effect [2]. DLI has also been used to facilitate immune reconstitution, treat chronic viral infections such as Epstein-Barr virus and cytomegalovirus, correct immunodeficiencies, and increase donor chimerism in mixed donor/host chimeric animals or humans [3].

Graft-versus-host disease (GVHD), a major complication of DLI therapy, typically correlates with desirable GVL reactivity and response rates [1, 2, 3, 4, 5]. Hence, common target antigens (Ags) and/or effector cells have been suspected in the processes of GVHD and GVL after DLI, although the GVL effect has been separable from GVHD in experimental animal models [6, 7, 8, 9]. In a large retrospective analysis, acute or chronic GVHD occurred in up to 60% of DLI-treated patients and accounted for 10% of deaths in these patients [4]. More than 90% of complete responders developed acute GVHD, and 88% of responders developed chronic GVHD [4]. In contrast, remissions have been rarely observed in patients without GVHD after DLI. Fortunately, GVHD in this setting has frequently been mild to moderate and responsive to immunosuppressive therapy. Moreover, adoption of a dose-escalation DLI strategy has been shown to decrease the risk for developing life-threatening GVHD [4].

DLI has allowed for the infusion of donor T cells at numbers that would typically induce lethal GVHD if given at the time of HSCT. Elucidation of the immunologic mechanisms involved in regulating graft-versus-host (GVH) reactivity is needed to optimize clinical usage for DLI-based immunotherapy. As possible mechanisms, findings from clinical and experimental studies have implicated avoidance of the cytokine storm initiated by pretransplantation conditioning, decreased numbers of host Ag-presenting cells (APCs) at the time of DLI, and development of regulatory T (Treg) cells [10, 11, 12, 13], but APC and Treg dynamics (eg, short-, intermediate-, or long-term after transplantation) and their relative contributions at each stage have not been fully addressed. Recently, diminished DLI-induced GVL reactivity was seen in long-term major histocompatibility complex (MHC)–matched or mismatched complete hematopoietic stem cell (HSC) chimeras [14, 15]. Similar findings have also been made in long-term MHC-mismatched chimeras that had converted from mixed to complete donor chimerism after an earlier DLI [16]. In this study, MHC-mismatched C57BL/6 (B6)–into-AKR complete donor chimeras and B6-into-AKR mixed donor/host chimeras were used to address Treg cell dynamics and DLI-induced GVL/GVH effects in early versus long-term HSC chimeras. First, we examined DLI-induced GVHD at several time points after HSCT (eg, days 28, 35, 42, 49, and 70) and the influence of Thy1+ Treg cell depletion on GVHD at each time point. Second, we compared DLI-induced GVL in early (day 28) versus late (day 100) complete donor chimeras. Third, we examined GVL reactivity in long-term complete donor and mixed donor/host chimeras. Finally, we investigated the effect of in vivo CD25+ Treg cell depletion on DLI-induced GVL/GVHD and whether host-type APCs, in the form of dendritic cells (DCs), given to long-term complete donor chimeras could influence GVL and GVH reactions.

Back to Article Outline

Materials and methods 

Mice 

AKR/J (H-2k, Thy1.1+), AKR/Cum (H-2k, Thy1.2+), C57BL/6 (B6; H-2b, Thy1.2+), and B6.PL-Thy1a/Cy (B6.PL; H-2b, Thy1.1+) mice were purchased from the Jackson Laboratory (Bar Harbor, ME) and housed in the Biomedical Resource Center at the Medical College of Wisconsin (Milwaukee, WI).

Reagents and Monoclonal Antibodies 

The following monoclonal antibodies (mAbs) were purchased from BD Pharmingen (San Diego, CA): fluorescein isothiocyanate (FITC)–conjugated anti–H-2Kb (clone AF6-88.5), FITC–anti-CD11c (clone HL3), phycoerythrin (PE)–conjugated anti-CD3ϵ (clone 145-2C11), PE–anti-CD25 (clone PC61), PE–anti-CD45R/B220 (clone RA3-6B2), PE–anti-CD11b (clone M1/70), PE–anti-Ly6G/6C (clone RB6-8C5), allophycocyanin-conjugated anti-CD4 (clone RM4-5), and PE–Cy5–anti-CD8a (clone 53-6.7). In vivo–depleting anti-CD25 (clone PC61) and anti-Thy1.2 (clone 53-2.1) mAbs were produced from hybridoma cells obtained from the American Type Culture Collection (Bethesda, MD). These mAbs were produced in our laboratory by using a Vivascience bioreactor system (Hanover, Germany). Rabbit complement was purchased from Cedarlane Laboratories Ltd. (Hornby, Ontario, Canada). Anti-CD11c (N418)–conjugated immunomagnetic beads were purchased from Miltenyi Biotec Inc. (Auburn, CA). Collagenase D was obtained from Roche Applied Sciences (Indianapolis, IN). Bovine serum albumin and 2-mercaptoethanol were obtained from Sigma (St. Louis, MO). Fetal bovine serum, phosphate-buffered saline, high-glucose Dulbecco’s modified eagle medium, 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid buffer, sodium pyruvate, l-glutamine, l-arginine, l-asparagine, folic acid, nonessential amino acids, and penicillin/streptomycin were obtained from Invitrogen (Carlsbad, CA).

In Vitro Isolation of DCs 

Spleens harvested from AKR mice were digested with collagenase D (1 mg/mL) for 45 minutes at 37°C in a 5% carbon dioxide incubator. The splenocytes were then washed, incubated with anti-CD11c–conjugated immunomagnetic beads for 15 minutes at 4°C, and purified by positive selection with an automated Miltenyi Biotec immunomagnetic cell sorter. The purity of CD11c+ DCs was >98%.

Establishment of Complete Donor and Mixed Donor/Host HSC Chimeras 

To establish complete donor HSC chimeras, AKR/J hosts were lethally irradiated with 1100 cGy 1 day before HSC. Fresh bone marrow (BM) cells harvested from B6 donors were depleted of mature T cells in vitro by using anti-Thy1.2 mAb and rabbit complement. A total of 107 T cell–depleted nucleated BM cells were given intravenously (IV) to each recipient via the tail veins. To establish mixed donor/host HSCT chimeras, AKR/Cum mice were lethally irradiated with 1100 cGy 1 day before HSCT, and a single intraperitoneal (IP) injection of anti-Thy1.2 mAb (500 μg) was given on the day of HSCT to prevent graft rejection. Fresh BM cells harvested from B6 and AKR/Cum donor mice were mixed at a donor-host ratio of 3:1, and a total of 107 non–T cell–depleted BM cells (ie, 7.5 × 106 from B6 and 2.5 × 106 from AKR/Cum) were injected IV. Levels of donor HSC engraftment were confirmed 3 weeks after transplantation by flow cytometric analysis of peripheral blood leukocytes by using the following cell-surface markers: FITC–anti–H-2Kb, FITC–anti–H-2Kk, PE–anti-CD3, PE–anti-B220, PE–anti–Mac-1, PE–anti-Ly6G/6C, PE–Cy5–anti-CD8, and allophycocyanin–anti-CD4.

DLI and Leukemia Challenge 

In all DLI studies, 3 × 107 splenocytes isolated from B6.PL mice (Thy1.1+) were administered IV at various time points after HSCT (days 28, 35, 42, 49, 70, and 100) to complete donor chimeras or to long-term mixed chimeras (day 100). Some DLI recipients were injected IP with anti-Thy1.2 mAb (500 μg) on the day of DLI to deplete donor BM-derived Thy1.2+ T cells in vivo. Treatment with anti-Thy1.2 mAb results in >98% depletion of Thy1.2+ cells (Figure 1). Other recipients were treated IP with anti-CD25 mAb 7 days (500 μg at day 93) and 4 days (250 μg at day 96) before DLI to deplete CD25+ cells in vivo. The optimal dose of anti-CD25 mAb needed to achieve maximal depletion of CD25+ T cells was determined in preliminary experiments (data not shown). In some day 28 or day 100 complete donor HSC chimeras, 2.5 × 106 or 5 × 106 purified AKR DCs were infused alone or together with a DLI. To assess GVL reactivity, mice were challenged IV with 103 AKR-derived M2 leukemia/lymphoma cells 1 week after DLI (ie, day 35 or day 107). The origin and characteristics of the M2 tumor, a T-cell leukemia/lymphoma that spontaneously arose in a male AKR/J mouse, have been previously described [17]. Mice that died from GVHD or tumor progression were differentiated from one another on the basis of clinical symptoms and systemic autopsy findings.

  • View full-size image.
  • Figure 1. 

    The effect of Thy1+ Treg cell depletion on DLI-induced GVHD waned with time. In the first column, survival curves are shown for Thy1.2-depleted (n = 12 per group for each time point) or -nondepleted (n = 10-11 per group for each time point) B6-into-AKR/J complete donor HSC chimeras given a DLI with 3 × 107 Thy1.1+ B6.PL splenocytes at day 28 (A), day 35 (B), day 42 (C), day 49 (D), or day 70 (E) after BM transplantation. The second and third columns contain flow cytometric histograms depicting the percentages of splenic Thy1.2+ cells in representative animals at each time point that had been treated or not treated with anti-Thy1.2 mAb. The survival data represent the combined results of 2 independent experiments, all time points were included in each of the experiments, and the mice in each experiment were given BM transplants on the same day and then given DLI at the indicated time points. Statistical comparisons (P values) for the 2 survival curves in each graph are shown (ns = not significant).

T-Cell and DC Phenotyping 

Peripheral blood was collected from early (day 28) and long-term (day 100) complete donor HSC chimeras to assess the presence of donor BM-derived CD4+ and CD8+ T cells, as well as CD4+CD25+ Treg cells, by flow cytometry. Peripheral blood leukocytes were stained with the following panel of cell-surface markers: FITC–anti–H-2Kb, PE–anti-CD3 or PE–anti-CD25, PE–Cy5–anti-CD8, and allophycocyanin–anti-CD4. To assess the purity of isolated DCs, the cells were stained with a combination of FITC–anti-CD11c and PE–anti-CD11b. The samples were analyzed on a Becton Dickinson FACSCalibur flow cytometer (San Jose, CA).

Statistics 

Survival curves were compared by log-rank analysis. Relative risk comparisons were performed by using Fisher exact tests. Body weight loss values were compared by using the t test. Significance was defined as P < .05.

Back to Article Outline

Results 

Induction of DLI-Induced GVHD after Treg Depletion Waned with Time 

Previous studies from our group showed that significant lethal acute GVHD occurs after DLI in MHC-mismatched recipients until the fourth week after HSCT and that new thymus-derived donor Treg cells are in large part responsible for suppressing the GVH reactivity at 4 weeks after HSCT [11]. However, it was unclear whether Treg cells continue to play an important role at later time points. Therefore, experiments were performed to elucidate the kinetics of Thy1+ Treg cells in suppressing GVHD development in longer-term chimeras. DLI given to MHC-mismatched recipients at days 28, 35, 42, 49, or 70 resulted in little or no GVH-related mortality (Figure 1; DLI groups). In vivo depletion of Thy1+ Treg cells by treatment of BM chimeras with anti-Thy1.2 mAb immediately before DLI resulted in severe and lethal GVHD when DLI was given at days 28, 35, and 42 (Figure 1; anti-Thy1.2 + DLI groups). In contrast, after Treg depletion, DLI induced relatively mild GVHD at day 49 and no GVHD at day 70. Flow cytometric analysis of splenocytes from BM chimeras at each DLI time point indicated that anti-Thy1.2 mAb treatment resulted in >98% depletion of peripheral Thy1.2-expressing T cells (Figure 1).

The DLI-Induced GVL Effect Observed Early (Day 28) after HSCT Was Absent in Long-Term (Day 100) Complete Donor Chimeras 

To study whether the DLI-induced GVL effect correlated with development of GVHD, early (day 28) and long-term (day 100) complete donor chimeras were treated with DLI and challenged with M2 leukemia cells 1 week later, ie, at days 35 and 107, respectively. As shown in Figure 2, the GVL effect was robust in day 28 chimeras (Figure 2A), but it was completely absent in day 100 chimeras (Figure 2B). In DLI-treated day 28 chimeras, 11 of 14 mice were tumor free for more than 70 days after challenge, with some long-term body weight loss (Figure 2C). Among the 3 deaths that occurred in this group, 2 (days 14 and 39 after tumor challenge) resulted from tumor progression, whereas the third (day 56) was attributed to GVHD. Akin to non–DLI-treated controls (BM transplantation only), all DLI-treated day 100 chimeras (12/12) died rapidly from tumor progression after challenge (Figure 2B), and corresponding acute body weight loss was observed in these animals (Figure 2D).

  • View full-size image.
  • Figure 2. 

    The DLI-induced GVL effect in day 28 complete donor HSC chimeras was absent in day 100 complete chimeras. Survival (A and B) and body weight changes (C and D) are shown for DLI-treated or -nontreated day 28 (A and C) or day 100 (B and D) complete donor HSC chimeras that were subsequently challenged with 103 AKR M2 leukemia cells 1 week after DLI (ie, day 35 and day 107, respectively). Animals that were given HSCT only (no DLI or leukemia) served as controls. The data represent the combined results of 4 independent experiments for the day 100 chimeras and 3 independent experiments for the day 28 chimeras. Each experimental group was repeated at least once. The total numbers of mice in each group are shown in (C) and (D). BMT, bone marrow transplantation.

Depletion of CD25+ Treg before DLI Failed to Restore the GVL Effect in Long-Term Complete Donor HSC Chimeras 

Previous studies from our group revealed that CD4+CD25+ cells comprise the most potent Thy1+ Treg population in experimental DLI models [17, 18]. When donor-derived T cells in the peripheral blood of day 28 and day 100 complete donor BM chimeras were phenotyped for CD3, CD4, CD8, and CD25 expression, the percentage of CD25+ cells within the CD4 population (ie, putative Treg cells) was relatively higher in day 100 chimeras (9.7%-12.3%; n = 3) than in day 28 chimeras (5.7%-7.1%; n = 3) or naive B6 mice (7.7%-9.5%; n = 5), even though similar percentages of CD3+CD4+ (15.5%-27.7% versus 17.4%-28.8%) and CD3+CD8+ (5.7%-7.1% versus 5.7%-6.7%) were detected in day 28 and day 100 chimeras, respectively. To determine whether the presence of CD4+CD25+ Treg contributed to the disappearance of DLI-induced GVL in long-term chimeras, day 100 chimeras were depleted of CD25+ cells in vivo before DLI and subsequently challenged with M2 leukemia cells 1 week after DLI (ie, day 107). CD25+ cell depletion failed to restore GVL reactivity (Figure 3). All anti-CD25/DLI–treated mice (n = 11) died from tumor growth after challenge (Table 1), and the kinetics of mortality in this group were similar to those of non–CD25-depleted mice given DLI (Figure 3A). Tumor progression was associated with cachexia (Figure 3B).

  • View full-size image.
  • Figure 3. 

    Depletion of CD25+ Treg was unable to restore the DLI-induced GVL effect in day 100 complete donor HSC chimeras. Survival (A) and body weight changes (B) are shown for day 100 complete donor chimeras treated or not treated with in vivo–depleting anti-CD25 mAb and/or DLI and then challenged with 103 AKR M2 leukemia cells 1 week later (day 107). The data for the anti-CD25 + M2 and anti-CD25 + DLI + M2 experimental groups represent the combined results of 4 independent experiments. Each group was included in at least 2 of the 4 experiments, which were all completed within a 9-month time period. The survival curves for the BM transplantation only, M2, and DLI + M2 groups are the same as in Figure 2, and they are shown simply to provide comparative survival data for the anti-CD25 + M2 and anti-CD25 + DLI + M2 experimental groups. The total numbers of mice in each group are shown in (B) BMT, bone marrow transplantation.

Table 1. Outcomes for DLI-Treated and Tumor-Challenged Day 100 Complete Donor Chimeras
GroupAnti-CD25Host DCsMortality⁎MST (d)†Causes of Death (n)
1NoNo100%(12/12)15Tumor (12), GVHD (0)
2No2.5×10681.8%(9/11)27Tumor (9), GVHD (0)‡
3No5.0×10680%(8/10)38Tumor (1), GVHD (7)
4YesNo100%(11/11)15Tumor (11), GVHD (0)
5Yes2.5×10666.7%(8/12)25Tumor (2), GVHD (6)‡
6Yes5.0×106100%(10/10)41Tumor (0), GVHD (10)

⁎ These data were obtained from the experiments in Figure 3, Figure 5, Figure 6.

† MST, median survival time.

‡ As analyzed by using Fisher exact tests, the relative risk of dying from tumor was significantly greater in group 2 than in group 5 (P = .003), and the relative risk of dying from GVHD was significantly greater in group 5 than in group 2 (P = .014). Log-rank comparisons of the survival curves for groups: 1 vs. 2, P = .019; 1 vs. 3, P = .0014; 1 vs. 4, P = NS (not significant); 4 vs. 5, P = .00488; 4 vs. 6, P = .0022; 2 vs. 5, P = NS; 3 vs. 6, P = NS.

DLI-Induced GVL Activity Occurred in Long-term Mixed HSC Chimeras without GVHD 

To investigate whether the disappearance of a DLI-induced GVL effect in long-term complete donor chimeras is related to a deficiency of stimulatory host alloantigens, B6/AKR long-term mixed HSC chimeras were generated. By transplanting recipients with a mixture of donor and host BM at a 3:1 ratio, 75% donor hematopoietic chimerism could be achieved (data not shown). The mixed chimeras were given (DLI + M2 group) or not given (M2 group) a DLI at day 100 after BM transplantation and then challenged with AKR M2 leukemia cells 1 week later (day 107). In contrast to complete chimeras, a robust GVL effect was observed in mixed chimeras given DLI, with 13 of 16 mice surviving tumor free for more than 70 days after tumor inoculation (Figure 4A). An absence of body weight loss in these mice correlated with an absence of GVH-associated mortality (Figure 4B), and it was confirmed that the 3 deaths were due to tumor progression. Long-term mixed chimeras not given a DLI were unable to resist the leukemia challenge, and all mice in this group died between 15 and 30 days after tumor challenge (Figure 4).

  • View full-size image.
  • Figure 4. 

    A DLI-induced GVL effect was observed in day 100 mixed chimeras without concurrent GVHD. Survival (A) and body weight changes (B) are shown for DLI-treated or -nontreated day 100 mixed HSC chimeras that were subsequently challenged with 103 AKR M2 leukemia cells 1 week after DLI (day 107). Controls that received BM only (no tumor) were included in these experiments. The data represent the combined results of 3 independent experiments. Each group was included in at least 2 of the 3 experiments. The total numbers of mice in each group are shown in (B). BMT, bone marrow transplantation.

Host-Type DCs Coinfused with DLI Partially Restored GVL and GVH Effects in Long-Term Complete Donor Chimeras 

Results in previous experiments suggested that the persistence of host-type APCs in day 100 mixed HSC chimeras facilitated a GVL effect in DLI-treated animals and that a lack of host APCs in day 100 complete donor chimeras was responsible for the absence of a GVL effect. To address this further, host-type DCs (2.5 or 5 × 106 per mouse) were coinfused with a DLI into day 100 chimeras to determine whether a DLI-induced GVL effect could be restored by infusion of host APCs. Although coadministration of 2.5 × 106 DCs with the DLI did not result in significantly better survival after tumor challenge as compared with non–DLI-treated chimeras (Figure 5A), the median survival time in DLI-treated mice was increased from 15 to 27 days (Table 1, groups 1 and 2). When the DC dose was increased to 5 × 106 cells, DLI coadministration resulted in a significant survival benefit as compared with non–DLI-treated chimeras (Figure 5B), and the median survival time in DLI-treated mice was further increased to 38 days (Table 1, group 3). It is interesting to note that although all deaths in the chimeras given 2.5 × 106 DCs and DLI (n = 9) were attributed to tumor progression, most deaths in chimeras given 5 × 106 DCs and DLI (7 of 8 deaths) were attributed to GVHD (Table 1). Significantly increased body weight loss in the long-term chimeras given 5 × 106 DCs and DLI (DLI + DC + M2 group; Figure 5D) as compared with chimeras not given DLI (DLI + M2 group; Figure 5D) correlated with the increased GVH reactivity. However, no enhancement of GVHD was seen in DLI-treated day 28 complete donor chimeras coinfused with 5 × 106 fresh host-type DCs (Figure 6).

  • View full-size image.
  • Figure 5. 

    Infusion of host-type DCs partially restored DLI-induced GVH/GVL reactivity in long-term complete donor chimeras. Survival (A and B) and body weight changes (C and D) are shown for day 100 complete donor chimeras given 2.5 × 106 (A and C) or 5 × 106 (B and D) host-type DCs with or without DLI followed by challenge with 103 AKR M2 leukemia cells at day 107. The data represent the combined results of 3 independent experiments. Each group was included in at least 2 of the 3 experiments that were all performed within a 3-month time period. The total numbers of mice in each group are shown in (C) and (D). The survival curve for the BM transplantation only (no tumor) group is the same as that shown in Figure 2B, and it is shown to provide comparative survival data for the other experimental groups. For the mice given 5 × 106 DCs (D), body weight loss in the DLI + DC + M2 group was significantly greater than that in the DC + M2 group at days 7 and 14 after tumor challenge, as indicated by the asterisks, (P < .03). BMT, bone marrow transplantation.

  • View full-size image.
  • Figure 6. 

    Infusion of host-type DCs did not enhance DLI-induced GVHD in short-term day 28 complete donor chimeras. Survival (A) and body weight changes (B) are shown for day 28 complete donor chimeras given DLI with or without 5 × 106 host-type DCs. Controls that received BM transplantation only were included in these experiments. The data represent the combined results of 3 independent experiments. Each group was included in at least 2 of the 3 experiments, and the total numbers of mice in each group are shown in (B). The survival curve for the BM transplantation–only group is the same as that shown in Figure 2A, and it is shown to provide comparative survival data for the other experimental groups. BMT, bone marrow transplantation.

Long-Term Chimeras Depleted of CD25+ Treg Cells before Coinfusion of Host DCs and DLI Had Increased GVL Reactivity at the Expense of GVHD 

We next examined whether DLI-mediated GVH/GVL reactions initiated by host DC administration are influenced by donor CD25+ Treg cells. To address this, day 100 complete donor chimeras were treated with 2 injections of anti-CD25 mAb at days 93 and 96 after HSCT to deplete CD25+ cells in vivo. CD25-depleted or -nondepleted mice were treated either with DCs alone (2.5 or 5 × 106 cells) or DCs plus DLI at day 100, and all mice were subsequently challenged with 103 AKR M2 leukemia cells at day 107. In vivo depletion of CD25+ cells before DLI (no host DCs) had no effect on GVL reactivity (Table 1, group 1 versus 4). As shown in Figure 7A and 7B, DLI/DC coadministration modestly increased 70-day survival as compared with chimeras given DCs without DLI, but the increases in survival were not significant, and depletion of CD25+ cells had no effect on survival. However, depletion of CD25+ cells did significantly increase GVL reactivity in DLI/DC-treated chimeras (Table 1, group 2 versus 5; P = .003), but any survival benefit was negated by significantly increased GVH reactivity in these animals (P = .014).

  • View full-size image.
  • Figure 7. 

    Depletion of CD25+ cells increased GVH/GVL reactivity in long-term BM chimeras coinfused with host DCs and DLI. Survival (A and B) and body weight changes (C and D) are shown for CD25-depleted or -nondepleted day 100 complete donor chimeras given 2.5 × 106 (A and C) or 5 × 106 (B and D) host-type DCs with or without DLI followed by challenge with AKR M2 leukemia cells at day 107. The data for the anti-CD25–treated groups represent the combined results of 3 independent experiments. Each group was included in at least 2 of the 3 experiments that were all performed within a 6-month time period. The total numbers of mice in each group are shown in (C) and (D). Survival curves for the DC + M2 and DLI + DC + M2 groups are the same as those shown in Figure 5, and they are shown to provide comparative survival data for the other experimental groups.

Back to Article Outline

Discussion 

In early animal studies, immune-suppressive/regulatory cells were implicated as playing a role in the suppression of GVH reactivity after DLI [10]. Subsequent studies from our group and others demonstrated that various types of donor-derived cells and even residual host cells, including non-T cells, may be involved in regulating development of DLI-induced GVHD [11, 18, 19, 20, 21, 22, 23, 24]. However, donor Thy1+CD4+CD25+ cells seem to be the most potent Treg population according to studies in our laboratory examining the relationship of DLI-induced GVHD and deficiency of particular donor-derived T-cell subsets [11, 18]. DLI is typically administered weeks or months after HSCT to treat tumor relapse in humans [4]. Therefore, it is conceivable that donor reconstitution of the host thymus has occurred in some DLI-treated patients and that donor-derived Treg cells play a role in blunting DLI-induced GVH reactivity. This may in part account for the ability to give patients relatively large numbers of donor T cells with less severe GVHD than would be anticipated if similar numbers of T cells were given at the time of HSCT.

In the experiments shown in Figure 1, we extended our previous findings in day 28 complete donor HSC chimeras to examine the importance of Thy1+ Treg cells at later time points. We confirmed our previous observations in day 28 chimeras demonstrating suppression of DLI-induced GVH reactivity by Thy1+ regulatory cells, and depletion of donor Thy1+ cells in day 35 and day 42 chimeras similarly resulted in lethal acute GVHD, thus indicating that Thy1+ Treg cells still played an important role in suppressing DLI-induced GVH reactivity at these later time points. However, by day 49 after HSCT, the role of Thy1+ Treg cells was no longer evident. This could be due to turnover of APCs from host to donor type during this time period, thus resulting in decreased activation of alloreactive T cells given as DLI. The presence of host-type APCs has been shown by others to be an absolute requirement for the induction of GVHD in murine HSCT models [12, 15, 25, 26], and DLI in humans and dogs with mixed chimerism has resulted in GVHD [27, 28]. Alternatively, it has been suggested that the persistence of Treg cells requires the ongoing presence of cognate Ags [29]. Thus, it is possible that Treg cells specific for host allo-Ags decrease in frequency as APCs turn over. However, the GVH reactivity induced in our long-term complete donor chimeras by coadministration of host-type DCs and DLI seemed to be regulated in part by CD25+ Treg cells (Table 1), thus indicating that these cells persisted in the long-term chimeras. CD4+CD25+ Treg cells have been used to effectively prevent or treat GVHD after HSCT or DLI [18, 19, 30, 31, 32, 33, 34]. A potential problem with this approach is that accumulating evidence has suggested that CD4+CD25+ Treg cells can promote tumor growth/progression [35, 36, 37, 38]. Therefore, depletion of CD4+CD25+ cells may have a synergistic benefit with DLI therapy to clear tumor cells, particularly when the DLI-induced anti-tumor response alone is too weak to completely eradicate the tumor cells.

An absence of GVHD in DLI-treated long-term complete chimeras after Thy1+ Treg cell depletion (Figure 1E) correlated with an absence of GVL reactivity in these animals (Figure 2B). These results are consistent with findings from other investigators who showed that GVL reactivity after DLI was absent in MHC-matched (day 84 after transplantation) or -mismatched (day 56 after transplantation) HSCT models, as well as in complete donor chimeras that had converted from a state of mixed chimerism to complete chimerism several weeks after DLI treatment [14, 15, 16]. We also observed a robust DLI-induced GVL effect in day 28 complete donor chimeras (Figure 2A) and in day 100 mixed chimeras (Figure 4A). Mapara et al. [15, 16] similarly reported DLI-induced GVL reactivity without GVHD in day 56 mixed chimeras and a superior GVL effect in mixed chimeras as compared with complete donor chimeras. Other data have indicated that GVH reactions in mixed chimeric mice after DLI may be confined to the lymphohematopoietic tissues, which may account for the absence of DLI-induced GVHD in these animals [39]. Clinically, acute myeloid leukemia patients have responded better to DLI therapy than patients with acute lymphocytic leukemia, possibly as a result of direct tumor Ag presentation by myeloid leukemia cells that express immune costimulatory molecules [1]. It is interesting to note that we found that deficient GVL reactivity in long-term complete HSC chimeras could be partially restored by coadministration of host-type DCs and DLI (Figure 5A and B), but the increased GVL effect occurred at the expense of more GVHD (Table 1). This suggests the involvement of common target Ags. However, it needs to be pointed out that our studies did not distinguish whether GVHD in the long-term chimeras actually involved tissue infiltration by donor T cells or was due to a cytokine storm. Clinically, the GVL effects of DLI therapy have also been closely associated with GVHD [1, 2, 3, 4, 5]. Thus, separation of GVL from GVH may be complicated because of the nature of the target Ags. Our experimental results seem to favor the hypothesis that turnover of APCs from host type to donor type results in diminished GVH and GVL reactivities after DLI therapy, which is consistent with findings from other research groups [25, 26]. Moreover, the absolute requirement for host-type APCs to elicit a GVL effect was elegantly demonstrated by the work of Mapara et al. [15] showing that GVL was robust when host-type DCs expressed MHC class I Ags but absent when host-type DCs in mixed chimeric mice were deficient in MHC class I. In one clinical study involving the use of prophylactic DLI to treat tumors after T cell–depleted BM transplantation, myeloablated patients were shown to tolerate higher dose of T cells and develop less severe GVHD when DLI was performed at 45 days instead of 30 days after transplantation [40]. However, DLI has mostly been used to treat residual or relapsed malignancies, and there are no conclusive data regarding the timing of administration and DLI-induced toxicity and/or GVL efficacy. Furthermore, benefits from DLI treatment are complicated by patient heterogeneity, tumor burden, the T-cell dose(s), and other therapies [1, 2, 3, 4, 5]. In practice, restoration of GVL/GVH activities by coinfusion of fresh host-type DCs would be difficult in humans because the DCs would need to be generated and cryopreserved before transplantation.

Data in Figure 4, as well as results from other investigators, have suggested that a state of mixed chimerism is important to facilitate a GVL effect without significant GVHD after DLI therapy in long-term murine chimeras [15, 16]. Nonmyeloablative conditioning regimens can facilitate the establishment of mixed chimerism with less toxicity than myeloablative regimens, but mixed chimerism is associated with a higher risk for tumor relapse because of the development of donor/host T-cell tolerance [41]. Fortunately, antitumor responses have been effectively generated after DLI therapy in both myeloablative and nonmyeloablative settings [41]. Unlike mice, mixed chimeric patients frequently develop GVHD after DLI, and the incidence and severity of GVHD in these patients has been lower than or equivalent to that of complete chimeric patients [27, 42]. In contrast to the results from complete chimeric patients, no significant differences in GVHD development were observed in mixed chimeric patients whether DLI was performed within the first 100 days after transplantation or beyond 180 days after transplantation [27, 40]. These observations could be explained by the continuous presence of host APCs in mixed chimeric patients. Studies are needed to address the discrepancy between the findings in mice and humans, and a thorough clinical analysis is needed to examine the relationship between DLI-induced GVL/GVHD and the timing of DLI, as well as the relationship of DLI-induced GVL and GVHD in patients with mixed multilineage chimerism.

In summary, a state of mixed chimerism was superior to complete donor chimerism, because preservation of host-type APCs seemed to be essential for initiating DLI-induced GVL/GVH reactions in our studies. These results are consistent with previous findings by other investigators [15]. Infusion of host-type DCs partially restored a DLI-induced GVL effect in long-term complete chimeras, but restoration of GVL reactivity was accompanied with an increased risk for GVHD. Our data indicate that donor-derived Treg cells can regulate DLI-induced GVL and GVH reactions in both early and long-term complete donor chimeras. The role of donor BM–derived Treg cells in our mixed chimeric DLI model is under investigation, and we are interested in testing whether modulating the functional activity of Treg cells can optimize GVL reactions in mixed chimeras.

Back to Article Outline

Acknowledgments 

This work was supported by US Public Health Services grant no. CA90286, the Midwest Athletes against Childhood Cancer (MACC) Fund (Milwaukee, WI), and the Children’s Research Institute, Children’s Hospital of Wisconsin (Milwaukee, WI).

Back to Article Outline

References 

  1. Kolb HJ , Schmid C , Barrett AJ , Schendel DJ . Graft-versus-leukemia reactions in allogeneic chimeras . Blood . 2004;103:767–776
  2. Johnson BD , Hanke CA , Truitt RL . The graft-versus-leukemia effect of post-transplant donor leukocyte infusion . Leuk Lymphoma . 1996;23:1–9
  3. Dazzi F , Goldman JM . Adoptive immunotherapy following allogeneic bone marrow transplantation . Annu Rev Med . 1998;49:329–340
  4. Porter DL , Antin JH . The graft-versus-leukemia effects of allogeneic cell therapy . Annu Rev Med . 1999;50:369–386
  5. Luznik L , Fuchs EJ . Donor lymphocyte infusions to treat hematological malignancies in relapse after allogeneic blood or marrow transplantation . Cancer Control . 2002;9:123–137
  6. Michalek J , Collins RH , Durrani HP , et al.   Definitive separation of graft-versus-leukemia- and graft-versus-host-specific CD4+ T cells by virtue of their receptor beta loci sequences . Proc Natl Acad Sci U S A . 2003;100:1180–1184
  7. Patterson AE , Korngold R . Infusion of select leukemia-reactive TCR Vbeta+ T cells provides graft-versus-leukemia responses with minimization of graft-versus-host disease following murine hematopoietic stem cell transplantation . Biol Blood Marrow Transplant . 2001;7:187–196
  8. Kim YM , Sachs T , Asavaroengchai W , Bronson R , Sykes M . Graft-versus-host disease can be separated from graft-versus-lymphoma effects by control of lymphocyte trafficking with FTY720 . J Clin Invest . 2003;111:659–669
  9. Amrolia PJ , Muccioli-Casadei G , Yvon E , et al.   Selective depletion of donor alloreactive T cells without loss of antiviral or antileukemic responses . Blood . 2003;102:2292–2299
  10. Weiden PL , Storb R , Tsoi MS , Graham TC , Lerner KG , Thomas ED . Infusion of donor lymphocytes into stable canine radiation chimeras (implications for mechanism of transplantation tolerance) . J Immunol . 1976;116:1212–1219
  11. Johnson BD , Becker EE , LaBelle JL , Truitt RL . Role of immunoregulatory donor T cells in suppression of graft-versus-host disease following donor leukocyte infusion therapy . J Immunol . 1999;163:6479–6487
  12. Clark FJ , Chakraverty R . Role of dendritic cells in graft-versus-host disease . J Hematother Stem Cell Res . 2002;11:601–616
  13. Zeiser R , Marks R , Bertz H , Finke J . Immunopathogenesis of acute graft-versus-host disease (implications for novel preventive and therapeutic strategies) . Ann Hematol . 2004;83:551–565
  14. Billiau AD , Fevery S , Rutgeerts O , Landuyt W , Waer M . Crucial role of timing of donor lymphocyte infusion in generating dissociated graft-versus-host and graft-versus-leukemia responses in mice receiving allogeneic bone marrow transplants . Blood . 2002;100:1894–1902
  15. Mapara MY , Kim YM , Wang SP , Bronson R , Sachs DH , Sykes M . Donor lymphocyte infusions mediate superior graft-versus-leukemia effects in mixed compared to fully allogeneic chimeras (a critical role for host antigen-presenting cells) . Blood . 2002;100:1903–1909
  16. Mapara MY , Kim YM , Marx J , Sykes M . Donor lymphocyte infusion-mediated graft-versus-leukemia effects in mixed chimeras established with a nonmyeloablative conditioning regimen (extinction of graft-versus-leukemia effects after conversion to full donor chimerism) . Transplantation . 2003;76:297–305
  17. Johnson BD , Drobyski WR , Truitt RL . Delayed infusion of normal donor cells after MHC-matched bone marrow transplantation provides an antileukemia reaction without graft-versus-host disease . Bone Marrow Transplant . 1993;11:329–336
  18. Johnson BD , Konkol MC , Truitt RL . CD25+ immunoregulatory T cells of donor origin suppress alloreactivity after BMT . Biol Blood Marrow Transplant . 2002;8:525–535
  19. Xia G , Kovochich M , Truitt RL , Johnson BD . Tracking ex vivo-expanded CD4+CD25+ and CD8+CD25+ regulatory T cells after infusion to prevent DLI-induced lethal acute GVHD . Biol Blood Marrow Transplant . 2004;10:748–760
  20. Johnson BD , Hanke CA , Becker EE , Truitt RL . Sca1(+)/Mac1(+) nitric oxide-producing cells in the spleens of recipients early following bone marrow transplant suppress T cell responses in vitro . Cell Immunol . 1998;189:149–159
  21. Billiau AD , Fevery S , Rutgeerts O , Landuyt W , Waer M . Transient expansion of Mac1+Ly6G+Ly6C+ early myeloid cells with suppressor activity in spleens of murine radiation marrow chimeras (possible implication for the graft-versus-host and graft-versus-leukemia reactivity of donor lymphocyte infusions) . Blood . 2003;102:740–748
  22. Johnson BD , Dagher N , Stankowski WC , Hanke CA , Truitt RL . Donor natural killer (NK1.1+) cells do not play a role in the suppression of GVHD or in the mediation of GVL reactions after DLI . Biol Blood Marrow Transplant . 2001;7:589–595
  23. Zhang ZX , Young K , Zhang L . CD3+CD4−CD8− αβ-TCR+ T cell as immune regulatory cell . J Mol Med . 2001;79:419–427
  24. Blazar BR , Lees CJ , Martin PJ , et al.   Host T cells resist graft-versus-host disease mediated by donor leukocyte infusions . J Immunol . 2000;165:4901–4909
  25. Shlomchik WD , Couzens MS , Tang CB , et al.   Prevention of graft-versus-host disease by inactivation of host antigen-presenting cells . Science . 1999;285:412–415
  26. Teshima T , Ordermann R , Reddy P , et al.   Acute graft-versus-host disease does not require alloantigen expression on host epithelium . Nat Med . 2002;8:575–581
  27. Bethge WA , Hegenbart U , Stuart MJ , et al.   Adoptive immunotherapy with donor lymphocyte infusions after allogeneic hematopoietic cell transplantation following nonmyeloablative conditioning . Blood . 2004;103:790–795
  28. Georges GE , Storb R , Thompson JD , et al.   Adoptive immunotherapy in canine mixed chimeras after nonmyeloablative hematopoietic cell transplantation . Blood . 2000;95:3262–3269
  29. Van Santen HM , Benoist C , Mathis D . Number of Treg cells that differentiate does not increase upon encounter of agonist ligand on thymic epithelial cells . J Exp Med . 2004;200:1221–1230
  30. Taylor PA , Lees C , Blazar BR . The infusion of ex vivo activated and expanded CD4+CD25+ immune regulatory cells inhibits graft-versus-host disease lethality . Blood . 2002;99:3493–3499
  31. Hoffmann P , Ermann J , Edinger M , Fathman CG , Strober S . Donor-type CD4+CD25+ regulatory T cells suppress lethal acute graft-versus-host disease after allogeneic bone marrow transplantation . J Exp Med . 2002;196:389–399
  32. Jones SC , Murphy GF , Korngold R . Post-hematopoietic cell transplantation control of graft-versus-host disease by donor CD4+CD25+ T cells to allow an effective graft-versus-leukemia response . Biol Blood Marrow Transplant . 2003;9:243–256
  33. Edinger M , Hoffmann P , Erman J , et al.   CD4+CD25+ regulatory T cells preserve graft-versus-tumor activity while inhibiting graft-versus-host disease after bone marrow transplantation . Nat Med . 2003;9:1144–1150
  34. Trenado A , Charlotte F , Fisson S , et al.   Recipient-type specific CD4+CD25+ regulatory T cells favor immune reconstitution and control graft-versus-host disease while maintaining graft-versus-leukemia . J Clin Invest . 2003;112:1688–1696
  35. North RJ , Bursuker I . Generation and decay of the immune response to a progressive fibrosarcoma. I. Ly-1+2- suppressor T cells down-regulate the generation of Ly-1-2+ effector T cells . J Exp Med . 1984;159:1295–1311
  36. Sutmuller RP , van Duivenvoorde LM , van Elsas A , et al.   Synergism of cytotoxic T lymphocyte-associated antigen 4 blockade and depletion of CD25(+) regulatory T cells in antitumor therapy reveals alternative pathways for suppression of autoreactive cytotoxic T lymphocyte responses . J Exp Med . 2001;194:823–832
  37. Morse MA , Clay TM , Mosca P , Lyerly HK . Immunoregulatory T cells in cancer immunotherapy . Expert Opin Biol Ther . 2002;2:827–834
  38. Terabe M , Berzofsky JA . Immunoregulatory T cells in tumor immunity . Curr Opin Immunol . 2004;16:157–162
  39. Pelot MR , Pearson DA , Swenson K , et al.   Lymphohematopoietic graft-vs.-host reactions can be induced without graft-vs-host disease in murine mixed chimeras established with a cyclophosphamide-based nonmyeloablative conditioning regimen . Biol Blood Marrow Transplant . 1999;5:133–143
  40. Barrett AJ , Mavroudis D , Tisdale J , et al.   T cell-depleted bone marrow transplantation and delayed T cell add-back to control acute GVHD and conserve a graft-versus-leukemia effect . Bone Marrow Transplant . 1998;21:543–551
  41. Verdonck LF , Petersen EJ , Lokhorst HM , et al.   Donor leukocyte infusions for recurrent hematologic malignancies after allogeneic bone marrow transplantation (impact of infused and residual donor T cells) . Bone Marrow Transplant . 1998;22:1057–1063
  42. Marks DI , Lush R , Cavenagh J , et al.   The toxicity and efficacy of donor lymphocyte infusions given after reduced-intensity conditioning allogeneic stem cell transplantation . Blood . 2002;100:3108–3114

PII: S1083-8791(05)01360-1

doi:10.1016/j.bbmt.2005.11.519

Biology of Blood and Marrow Transplantation
Volume 12, Issue 4 , Pages 397-407, April 2006

Article Tools

* Image Usage & Resolution