In murine models of allogeneic bone marrow transplantation (BMT), MHC-mismatched recipients given a delayed infusion of donor leukocytes (DLI) at 21 days posttransplant develop significant GVHD whereas MHC-matched recipients do not. The current study was initially designed to test the hypothesis that small numbers of T cells in the MHC-mismatched donor bone marrow (BM) graft exacerbated graft-vs-host disease (GVHD) when DLI was administered at 21 days after BMT. Ex vivo depletion of Thy1+ cells from the donor BM had no impact on the severity of GVHD after DLI. However, depletion of donor T cells in vivo with a Thy1 allele-specific mAb given after BMT resulted in significantly more severe GVHD after DLI. Similar results were obtained in a MHC-matched model of allogeneic BMT, indicating that this was a general phenomenon and not model dependent. These results indicated that a population of donor-derived Thy1+ cells suppressed graft-vs-host reactivity after DLI. Results of experiments with thymectomized recipients demonstrated that an intact thymus was required for generation of the immunoregulatory donor cells. Experiments using TCR β-chain knockout mice as BM donors indicated that the immunosuppressive Thy1+ cells coexpressed αβTCR heterodimers. Similar experiments with CD4 and CD8 knockout donor BM suggested that the immunoregulatory Thy1+αβTCR+ cells consisted of two subpopulations: a CD4+CD8 subpopulation and a CD4CD8 subpopulation. Together, these results show that thymus-derived, Thy1+αβTCR+ donor cells generated early after allogeneic BMT suppress the graft-vs-host reactivity of T cells given as DLI. These cells may mediate dominant peripheral tolerance after allogeneic BMT.

Cellular immunotherapy with infusions of donor lymphocytes, referred to as delayed (or donor) leukocyte infusion (DLI)3 therapy, has been used to induce stable remissions in patients with posttransplant leukemia relapse (1, 2, 3). Despite the large numbers of donor T cells given during DLI therapy, graft-vs-host disease (GVHD) has typically been less severe than that observed when similar numbers of donor T cells were infused at the time of bone marrow transplant (BMT). Reasons for the decreased severity of GVHD following DLI therapy are unknown.

Experimental data from our laboratory indicated that the generation of alloreactive Th cells was inhibited following DLI in murine MHC-matched bone marrow (BM) chimeras (4). Mechanisms responsible for decreased GVHD following DLI might include 1) avoidance of the “cytokine storm” induced by pretransplant conditioning (5, 6, 7); 2) presence of fewer host-type APC due to the switch in hematopoiesis from host to donor (8); and/or 3) induction of immunosuppressive regulatory cells in the newly developing immune system (9, 10, 11).

While examining mechanisms for decreased graft-vs-host (GVH) reactivity following DLI, we serendipitously observed that administration of a Thy1 allele-specific mAb against donor BM-derived T cells increased the severity of GVHD in full MHC-mismatched murine recipients given DLI. Similar results were obtained in a MHC-matched model. The results led us to test the hypothesis that immunoregulatory Thy1+ cells of donor origin developed early posttransplant and suppressed the GVH reactivity of T cells given as DLI. Using thymectomized hosts as well as marrow from donors lacking αβTCR+, CD4+, or CD8+ T cells, we determined that the immunoregulatory cell population was 1) αβTCR+CD4+/−CD8; 2) derived from the donor BM; and 3) educated in the repopulating host thymus. The origin and regulatory properties of this population suggest that dominant peripheral tolerance contributes to the suppression of GVH reactivity after DLI therapy.

C57BL/6 (H-2b; Thy1.2+), B6.PL-Thy1a (H-2b; Thy1.1+), B10.BR (H-2k; Thy1.2+), AKR (H-2k; Thy1.1+), C57BL/6-Cd4tm1MAK(CD4 knockout (ko)), C57BL/6-Cd8atm1MAK (CD8 ko), and C57BL/6-Tcrbtm1Mom (TCR β-chain ko) mice (4–6-wk old) were purchased from The Jackson Laboratory (Bar Harbor, ME). B10.BR-Thy1.1 congenic mice were obtained from Bruce Blazar (University of Minnesota, Minneapolis, MN) and bred at the Medical College of Wisconsin (Milwaukee, WI). All mice were housed in the Medical College of Wisconsin’s American Association for the Accreditation of Laboratory Animal Care approved Animal Resource Center.

Normal or surgically thymectomized (Tx) AKR recipient mice were preconditioned with a lethal dose of 1100 cGy of total body irradiation 2–24 h before BMT. BM cells were obtained from healthy C57BL/6, C57BL/6 transgenic (CD4 ko, CD8 ko, or βTCR ko), or B10.BR donor mice by flushing the femurs and tibias with DMEM (Life Technologies, Grand Island, NY). The irradiated AKR recipients were injected i.v. with either 107 non-T cell-depleted or 107 T cell-depleted (TCD) donor BM cells. For T cell depletion, the BM cells were treated with anti-Thy1.2 mAb (clone 30-H12; obtained from the American Type Culture Collection, Manassas, VA) and Low-Tox-M complement (Accurate Chemical and Scientific, Westbury, CT). Some chimeras were treated with i.p. injections of anti-Thy1.2 mAb before DLI. Anti-Thy1.2 mAb was used in the form of bioreactor-derived culture supernatant or partially purified from bioreactor culture supernatants by ammonium sulfate precipitation. For DLI, 3 × 107 B10.BR, congenic B10.BR-Thy1.1, C57BL/6, or congenic B6.PL-Thy1a spleen cells were injected i.v. at day 21 or 28 after BMT.

Spleen cells and purified T cells were stained with various combinations of anti-Thy1.2 mAb (clone 30-H21; FITC-conjugated), anti-H-2Kb mAb (clone AF6-88.5; FITC-conjugated), anti-Thy1.1 mAb (clone OX-7; FITC- or PE-conjugated), anti-CD4 mAb (clone GK1.5; PE-conjugated; PharMingen, San Diego, CA), and anti-CD8 (clone CT-CD8a; PE-conjugated; Caltag, Burlingame, CA). Two-color analysis was performed using a Becton Dickinson FACScan flow cytometer (San Jose, CA).

Limiting dilution analysis (LDA) assays were used to determine Th cell and CTL precursor frequencies as described previously with minor modifications (4). Briefly, the spleen cells from five or six individual BM chimeras given DLI 7 days previously were isolated and pooled. The cells were first treated with anti-Thy1.2 mAb and complement to eliminate thymus-derived donor T cells. Responder Thy1.1+ T cells (i.e., donor cells from the DLI) were then enriched by positive selection using a MACS cell separator with a combination of anti-CD4-conjugated and anti-CD8-conjugated microbeads. Enrichment of H-2b/Thy1.1+ cells was confirmed by two-color flow cytometric analysis. To assess alloreactive Th cell frequencies, the Thy1.1-enriched responder cells were cocultured at 10 different concentrations in 24-replicate U-bottom microwells with 3 × 104 irradiated (3000 cGy) AKR dendritic cells and 105 irradiated (700 cGy) AKR B cells (200 μl/well total volume). Dendritic cells and B cells were enriched by MACS cell separation using CD11c-conjugated or B220-conjugated microbeads, respectively. After 4 days of culture in a 37°C, 10% CO2 incubator, 150 μl of culture supernatant was removed from each well and transferred to a new microwell plate containing 2500 IL-2-dependent SAC 9.12 (12) indicator cells and 1 μCi of [3H]thymidine. The plates were cultured for an additional 18–24 h, and thymidine uptake was assessed. Spontaneous thymidine uptake was determined from the supernatants of 24 wells that had contained stimulator cells only.

To assess CTL frequencies, various numbers of Thy1.1-enriched responder cells were cocultured in U-bottom microwells (24 wells per responder concentration) with 3 × 104 irradiated (3000 cGy) AKR dendritic cells and 2 × 105 irradiated (700 cGy) AKR B cells. Con A-stimulated rat spleen cell-conditioned medium was added to the wells at a concentration of 50% (v/v) (200 μl/well total volume). The plates were cultured for 8 days. On day 8, 50 μl of culture supernatant was removed from each well and replaced with 50 μl of 51Cr-labeled AKR Con A-stimulated lymphoblast targets (5000/well). 51Cr release was assessed after a 3.5-h incubation. Spontaneous and maximum 51Cr release values were determined from 24 wells containing target cells plus stimulators only or detergent (2.5% 7× detergent; Flow Laboratories, McLean, VA), respectively.

Individual wells in the Th cell and CTL assays were scored as positive when experimental values (cpm [3H]thymidine uptake or percent specific 51Cr release) exceeded the spontaneous control values by at least 3 SDs. The frequencies of alloreactive Th cells and CTL was calculated by χ2 minimization as described by Taswell (13).

Kaplan-Meier survival curves were compared using log-rank statistical analysis. The Student t test was used to analyze the percentages and numbers of donor T cells in the spleens of BMT chimeras given DLI.

MHC-matched BM chimeras (B10.BR-into-AKR) given DLI at 21 days after marrow transplantation survived without GVHD (Fig. 1,A). In contrast, MHC-mismatched (C57BL/6-into-AKR) chimeras developed severe GVHD after DLI, which resulted in 67% GVH-associated mortality (Fig. 1,A). GVHD in the MHC-mismatched chimeras was reflected by a sharp loss in body weight within the first week after donor cell infusion (Fig. 1 B), followed by continued body weight loss during the remainder of the experiment. These results confirmed earlier work from our laboratory (4, 14, 15).

FIGURE 1.

GVHD following DLI at 21 days after BMT is significantly more severe in MHC-mismatched chimeras than in MHC-matched chimeras. Irradiated AKR recipients were transplanted with 107 non-T-depleted B10.BR or C57BL/6 BM cells. The recipients were randomized to receive no further treatment (BM only) or DLI with 3 × 107 B10.BR or C57BL/6 spleen cells at 21 days after BMT. The mice were followed for survival (A) and body weight changes (B). Survival curves are the combined data of two experiments, and the body weight curves are from one representative experiment.

FIGURE 1.

GVHD following DLI at 21 days after BMT is significantly more severe in MHC-mismatched chimeras than in MHC-matched chimeras. Irradiated AKR recipients were transplanted with 107 non-T-depleted B10.BR or C57BL/6 BM cells. The recipients were randomized to receive no further treatment (BM only) or DLI with 3 × 107 B10.BR or C57BL/6 spleen cells at 21 days after BMT. The mice were followed for survival (A) and body weight changes (B). Survival curves are the combined data of two experiments, and the body weight curves are from one representative experiment.

Close modal

To explain the disparate results between the MHC-matched and mismatched BM chimeras, we initially hypothesized that small numbers of donor T cells in the MHC-mismatched BM grafts (∼1–3%) augmented the alloreactivity of donor T cells contained in the DLI inoculum, thereby increasing the severity of DLI-induced GVHD. We used two different experimental approaches to test this hypothesis. First, we depleted the donor BM of T cells ex vivo with anti-Thy1 mAb and complement before BMT. If mature donor T cells in the marrow graft were promoting GVH reactivity of T cells given as DLI, we speculated that ex vivo depletion of the donor T cells would result in less severe GVHD after DLI. AKR recipients were transplanted with untreated or TCD C57BL/6 BM and then randomized to receive no further treatment (BM controls) or to be given DLI at 21 days after BMT (Fig. 2). Control mice given T-depleted or nondepleted BM alone (no DLI) did not develop clinically evident GVHD and maintained their body weight after an initial loss attributed to the total body irradiation conditioning. Both groups of mice given DLI on day 21 developed GVHD as indicated by increased weight loss and mortality following infusion. Thus, ex vivo T cell depletion of the marrow graft had no effect on GVHD-related mortality and did not support the working hypothesis.

FIGURE 2.

Ex vivo depletion of Thy1.2+ cells in the donor BM inoculum had no effect on GVHD after DLI with MHC-mismatched spleen cells. Irradiated AKR recipients were transplanted with 107 T-depleted or non-T-depleted C57BL/6 BM cells. The recipients were then randomized to receive no further treatment (BM only) or DLI with 3 × 107 congenic B6.PL-Thy1a spleen cells at 21 days after BMT. The mice were followed for survival (A) and body weight changes (B). Survival curves are the combined data of two experiments, and the body weight curves are from one representative experiment. Key statistics: ○ vs •, p = 0.023; □ vs ▪, p = 0.037; • vs ▪, p = NS; ○ vs □, p = NS.

FIGURE 2.

Ex vivo depletion of Thy1.2+ cells in the donor BM inoculum had no effect on GVHD after DLI with MHC-mismatched spleen cells. Irradiated AKR recipients were transplanted with 107 T-depleted or non-T-depleted C57BL/6 BM cells. The recipients were then randomized to receive no further treatment (BM only) or DLI with 3 × 107 congenic B6.PL-Thy1a spleen cells at 21 days after BMT. The mice were followed for survival (A) and body weight changes (B). Survival curves are the combined data of two experiments, and the body weight curves are from one representative experiment. Key statistics: ○ vs •, p = 0.023; □ vs ▪, p = 0.037; • vs ▪, p = NS; ○ vs □, p = NS.

Close modal

As a second approach, we treated BM chimeras before DLI with a Thy1.2-specific mAb to deplete donor T cells in vivo. This would determine whether donor T cells generated de novo from BM precursors contributed to GVH reactivity after DLI. In the initial experiments, we administered anti-Thy1.2 mAb on days 10 and 12 after BMT followed by DLI at 21 days after BMT. The mAb persisted in the circulation of treated animals for several weeks (data not shown). Therefore, to avoid potential problems with residual circulating anti-Thy1.2 mAb, spleen cells from Thy1.1+ congenic H-2b donors (B6.PL-Thy1a) were given as DLI. Surprisingly, as shown in Fig. 3,A, chimeras pretreated with anti-Thy1.2 mAb in vivo showed accelerated and increased GVHD-associated mortality after DLI as compared with non-Ab-treated chimeras. Mice in both experimental groups had a sharp loss in body weight following DLI (Fig. 3 B) and developed clinically evident signs of GVHD including diarrhea and ruffled fur; however, clinical symptoms in the anti-Thy1.2-treated chimeras were more severe. Control mice given MHC-mismatched BM alone (no DLI) survived without any clinical evidence of GVHD and showed only a transient 10% body weight loss between 3 and 5 wk posttransplant. All BM-only control mice treated with anti-Thy1.2 survived >90 days and did not show any significant body weight loss, indicating that the mAb treatment alone was not toxic. Taken together, these results contradicted our original hypothesis and suggested instead that a population of donor BM-derived Thy1.2+ cells suppressed, rather than increased, the severity of GVHD after DLI.

FIGURE 3.

In vivo depletion of donor BM-derived Thy1.2+ cells resulted in exacerbated GVHD after DLI with MHC-mismatched cells. Lethally irradiated AKR hosts were transplanted with 107 non-T-depleted C57BL/6 BM cells. The hosts were randomized to receive no treatment or injected i.p. with 250 μl of anti-Thy1.2 mAb on days 10 and 12 posttransplant. For DLI, 3 × 107 B6.PL-Thy1a spleen cells (Thy1.1+) were administered at day 21 after BMT. Survival curves (A) represent the combined data from two experiments; body weight curves (B) are from one representative experiment.

FIGURE 3.

In vivo depletion of donor BM-derived Thy1.2+ cells resulted in exacerbated GVHD after DLI with MHC-mismatched cells. Lethally irradiated AKR hosts were transplanted with 107 non-T-depleted C57BL/6 BM cells. The hosts were randomized to receive no treatment or injected i.p. with 250 μl of anti-Thy1.2 mAb on days 10 and 12 posttransplant. For DLI, 3 × 107 B6.PL-Thy1a spleen cells (Thy1.1+) were administered at day 21 after BMT. Survival curves (A) represent the combined data from two experiments; body weight curves (B) are from one representative experiment.

Close modal

Time after BMT is an important variable for avoiding GVHD after DLI therapy (4). We previously reported that GVHD was avoided in MHC-mismatched BM chimeras if DLI was delayed from 21 days until 28 days posttransplant (15). Because of the surprising results described above, experiments were done to test whether depletion of donor marrow-derived Thy1.2+ T cells would result in increased GVHD when DLI was given at 28 days posttransplant. AKR recipients were transplanted with untreated or TCD BM. Anti-Thy1.2 mAb was given to chimeras on days 10, 12, 24, and 26 after BMT and was followed with DLI at 28 days after BMT. The additional Ab treatments given on days 24 and 26 were to ensure that Thy1.2+ cells were depleted at the time of DLI. T depletion was confirmed by flow cytometric analysis (data not shown). As shown in Fig. 4,A, 97% (30/31) of chimeras treated with anti-Thy1.2 mAb died within 2 wk after DLI. Death was paralleled by a significant decrease in body weight (Fig. 4,B). Ex vivo T depletion of the BM inoculum had no impact on the timing of mortality. Untreated BM chimeras had a significantly higher survival rate than the anti-Thy1.2-treated chimeras, and body weight loss was decreased. Differences in GVH-associated mortality between the BM chimeras transplanted with nondepleted or TCD BM were not significant (Fig. 4,A), confirming the results of Fig. 2. These data supported the hypothesis that a donor marrow-derived population of Thy1+ “immunoregulatory” cells suppressed GVHD following DLI in MHC-mismatched BM chimeras. The suppressive effect of the cells was more pronounced at 28 days than at 21 days posttransplant, but, in both instances, their removal resulted in intensified GVHD after DLI.

FIGURE 4.

Treatment of MHC-mismatched BM chimeras with anti-Thy1.2 mAb in vivo resulted in severe and lethal GVHD after DLI. Irradiated AKR recipients were transplanted with 107 T-depleted or non-T-depleted C57BL/6 BM cells. The transplanted recipients were randomized to receive no treatment or injected i.p. with 250 μl of anti-Thy1.2 mAb on days 10, 12, 24, and 26 after BMT. For DLI, 3 × 107 B6.PL-Thy1a spleen cells were administered at day 28 after BMT. Survival curves (A) represent the combined data from two experiments; body weight curves (B) are from one representative experiment.

FIGURE 4.

Treatment of MHC-mismatched BM chimeras with anti-Thy1.2 mAb in vivo resulted in severe and lethal GVHD after DLI. Irradiated AKR recipients were transplanted with 107 T-depleted or non-T-depleted C57BL/6 BM cells. The transplanted recipients were randomized to receive no treatment or injected i.p. with 250 μl of anti-Thy1.2 mAb on days 10, 12, 24, and 26 after BMT. For DLI, 3 × 107 B6.PL-Thy1a spleen cells were administered at day 28 after BMT. Survival curves (A) represent the combined data from two experiments; body weight curves (B) are from one representative experiment.

Close modal

The presence of Thy1+ regulatory cells was examined in an established MHC-matched BMT model (B10.BR-into-AKR) to determine whether the observations made in the MHC-mismatched combination were model specific. AKR recipients, transplanted with non-T-depleted B10.BR BM, were treated with anti-Thy1.2 mAb on days 10, 12, 17, and 19 posttransplant. DLI with B10.BR-Thy1.1 congenic spleen cells was given at 21 days after BMT. Control non-Ab-treated BM chimeras given DLI did not develop lethal GVHD and maintained their body weights after DLI (Fig. 5). In contrast, chimeras treated with anti-Thy1.2 mAb developed GVHD after DLI, resulting in 75% mortality (Fig. 5,A). GVH reactivity in Ab-treated chimeras was accompanied by a significant loss of body weight (Fig. 5 B). These results indicated that suppression of DLI-mediated GVH reactivity by donor-derived Thy1+ immunoregulatory cells was a general phenomenon and not model dependent.

FIGURE 5.

Treatment of MHC-matched BM chimeras with anti-Thy1.2 mAb in vivo resulted in severe GVHD after DLI. Lethally irradiated AKR hosts were transplanted with 107 non-T-depleted B10.BR BM cells. The hosts were randomized to receive no treatment or injected i.p. with 500 μg of anti-Thy1.2 mAb on days 10, 12, 17, and 19 after BMT. For DLI, 3 × 107 B10.BR-Thy1.1 spleen cells were administered at day 21 after BMT. Survival curves (A) represent the combined data from two experiments; body weight curves (B) are from one representative experiment.

FIGURE 5.

Treatment of MHC-matched BM chimeras with anti-Thy1.2 mAb in vivo resulted in severe GVHD after DLI. Lethally irradiated AKR hosts were transplanted with 107 non-T-depleted B10.BR BM cells. The hosts were randomized to receive no treatment or injected i.p. with 500 μg of anti-Thy1.2 mAb on days 10, 12, 17, and 19 after BMT. For DLI, 3 × 107 B10.BR-Thy1.1 spleen cells were administered at day 21 after BMT. Survival curves (A) represent the combined data from two experiments; body weight curves (B) are from one representative experiment.

Close modal

BM chimeras pretreated with anti-Thy1.2 mAb developed significantly more severe GVHD after DLI, as indicated by increased mortality and body weight loss. To determine whether the frequencies of host-reactive CTL and Th cells increased as a result of depletion of the immunoregulatory Thy1+ cells, spleen cells were collected from MHC-mismatched BM chimeras 7 days after DLI. Flow cytometric and LDA assays were performed. Thy1.1+ T cells in the DLI inoculum could be distinguished from residual host T cells by differential MHC class I expression (H-2Kb). As shown in Table I, the anti-Thy1.2-treated chimeras had significantly increased percentages and absolute numbers of splenic T cells derived from the DLI inoculum as compared with non-mAb-treated controls. Infused CD8+ T cells accounted for the majority of this increase (data not shown). Results of LDA assays performed on splenocyte preparations from anti-Thy1.2-treated chimeras 7 days after DLI indicated that significantly greater numbers of both anti-host-reactive CTL and Th cells were present as compared with the untreated BM chimeras (Table II). Anti-Thy1.2-treated chimeras had 15-fold and 3-fold increases in anti-host-reactive CTL and Th cells, respectively. Together, the data in Tables I and II provided further evidence that donor marrow-derived Thy1+ cells down-regulated anti-host (GVHD) alloreactivity after DLI.

Table I.

Elimination of donor BM-derived Thy1+ cells before DLI resulted in higher percentages and greater numbers of infused donor T cells in the spleens of DLI-treated chimeras

Anti-Thy1.2an% Infused Donor T Cells/SpleenbNo. of Infused T Cells/Spleen (×106)b
No 18 6.1 ± 1.8 4.5 ± 1.6 
Yes 15 16.0 ± 3.3* 8.5 ± 2.2* 
Anti-Thy1.2an% Infused Donor T Cells/SpleenbNo. of Infused T Cells/Spleen (×106)b
No 18 6.1 ± 1.8 4.5 ± 1.6 
Yes 15 16.0 ± 3.3* 8.5 ± 2.2* 
a

AKR hosts, preconditioned for BMT with 1100 cGy, were transplanted with 107 T-depleted C57BL/6 BM cells. On days 24 and 26 posttransplant, designated chimeras were treated with 500 μg anti-Thy1.2 mAb i.p. On day 28 posttransplant, all chimeras were given DLI with 3 × 107 B6.PL spleen cells.

b

The chimeras were sacrificed 7 days post-DLI and individual spleens were tested by flow cytometric analysis to determine the percentage and absolute number of infused donor T cells. The data is from the combined results of two experiments and the values shown are the averages per spleen ± SD.

c

, p < 0.05.

Table II.

Elimination of donor BM-derived Thy1+ cells before DLI resulted in significantly greater numbers of anti-host-reactive CTL and Th cells in the spleens of DLI-treated chimeras

Anti-Thy1.2aDays after DLICytolytic T Cell FrequenciesTh Cell Frequencies
CTL/106 cellsFold increase over DLI controlTH/106 cellsFold increase over DLI control
No 30  471  
Yes 443* 15 1587* 
Anti-Thy1.2aDays after DLICytolytic T Cell FrequenciesTh Cell Frequencies
CTL/106 cellsFold increase over DLI controlTH/106 cellsFold increase over DLI control
No 30  471  
Yes 443* 15 1587* 
a

AKR hosts, preconditioned for BMT with 1100 cGy, were transplanted with 107 T cell-depleted C57BL/6 BM cells. On days 24 and 26 posttransplant, designated chimeras were treated with 500 μg anti-Thy1.2 mAb i.p. On day 28 posttransplant, all chimeras were given DLI with 3 × 107 B6.PL spleen cells. Seven days after DLI, the chimeras (5–6 per group) were killed, and the pooled splenocytes were tested in LDA assays (see Materials and Methods for details).

b

, significantly greater value than the non-Ab-treated control (i.e., nonoverlapping 95% confidence intervals).

Because pretransplant depletion of T cells had no effect on DLI-induced GVHD, but in vivo depletion of Thy1+ cells increased GVHD after DLI, we hypothesized that the BM-derived immunoregulatory cells were generated de novo in the repopulating host thymus. Emigration of Thy1.2+ T cells from the donor-reconstituted host thymus begins ∼14 days after BMT (B. Johnson, unpublished data). To prove that the immunoregulatory donor cells were derived from the repopulated host thymus after BMT, AKR recipients were thymectomized ∼2 wk before BMT. Tx or normal euthymic AKR recipients were transplanted with T-depleted C57BL/6 BM and then given DLI at day 28 posttransplant. Separate groups of Tx and normal recipients were treated with anti-Thy1.2 mAb on days 24 and 26 after BMT and then given DLI on day 28. All nonthymectomized chimeras treated with anti-Thy1.2 mAb died from severe GVHD within 16 days after DLI (Fig. 6,A). The Tx chimeras with or without anti-Thy1.2 mAb treatment developed severe GVHD after DLI. The kinetics of body weight loss and mortality paralleled that of the anti-Thy1.2-treated nonthymectomized group (Fig. 6). These results demonstrated that an intact thymus was required for generation of the donor-derived Thy1+ immunoregulatory cells. The nonthymectomized chimeras given DLI had an overall survival rate of 78% and maintained their average body weight near pretransplant levels (Fig. 6).

FIGURE 6.

An intact host thymus was required for de novo generation of the donor BM-derived immunoregulatory cells. Irradiated normal or Tx AKR recipients were transplanted with 107 T-depleted C57BL/6 BM cells. The hosts were randomized to receive no treatment or injected i.p. with 500 μg of anti-Thy1.2 mAb on days 24 and 26 posttransplant. All chimeras were given DLI with 3 × 107 B6.PL-Thy1a spleen cells at 28 days after BMT. Survival curves (A) represent the combined data from two experiments; body weight curves (B) are from one representative experiment.

FIGURE 6.

An intact host thymus was required for de novo generation of the donor BM-derived immunoregulatory cells. Irradiated normal or Tx AKR recipients were transplanted with 107 T-depleted C57BL/6 BM cells. The hosts were randomized to receive no treatment or injected i.p. with 500 μg of anti-Thy1.2 mAb on days 24 and 26 posttransplant. All chimeras were given DLI with 3 × 107 B6.PL-Thy1a spleen cells at 28 days after BMT. Survival curves (A) represent the combined data from two experiments; body weight curves (B) are from one representative experiment.

Close modal

Experiments were designed to further characterize the donor-derived host-educated Thy1+ immunoregulatory cells. The first set of experiments addressed whether the immunoregulatory cells coexpressed αβ heterodimers of the TCR by using mice with a ko mutation for the β-chain of the TCR as BM donors. These mice do not produce T cells expressing αβTCR heterodimers (16). AKR recipients were transplanted with TCD BM from normal C57BL/6 mice or TCD BM from C57BL/6 βTCR ko mice. If the immunoregulatory cells were αβTCR+, the severity of DLI-induced GVHD in chimeras transplanted with βTCR ko BM should be comparable to that of chimeras treated with anti-Thy1.2 mAb. The transplanted chimeras were randomized to be treated or not treated with anti-Thy1.2 mAb on days 24 and 26 posttransplant. DLI was administered on day 28 posttransplant. Groups of chimeras given BM alone served as GVH-negative controls. Non-Ab-treated chimeras transplanted with normal BM developed only mild GVHD after DLI and had a 70% survival rate (Fig. 7). In contrast, chimeras transplanted with βTCR ko BM developed severe GVHD, and the kinetics of mortality and body weight loss (data not shown) was similar to the groups treated with anti-Thy1.2 mAb. As described previously, C57BL/6-into-AKR chimeras pretreated with anti-Thy1.2 mAb rapidly developed severe and lethal GVHD following DLI. These results indicated that the donor Thy1+ immunoregulatory cells were αβTCR+.

FIGURE 7.

Transplantation of BM from βTCR ko donor mice resulted in lethal GVHD after DLI. Irradiated AKR recipients were transplanted with 107 T-depleted normal or βTCR ko C57BL/6 BM cells. The hosts were randomized to receive no treatment or injected i.p. with 500 μg of anti-Thy1.2 mAb on days 24 and 26 posttransplant. For DLI, chimeras were given 3 × 107 B6.PL-Thy1a spleen cells at 28 days after BMT. Survival curves represent the combined data from two experiments.

FIGURE 7.

Transplantation of BM from βTCR ko donor mice resulted in lethal GVHD after DLI. Irradiated AKR recipients were transplanted with 107 T-depleted normal or βTCR ko C57BL/6 BM cells. The hosts were randomized to receive no treatment or injected i.p. with 500 μg of anti-Thy1.2 mAb on days 24 and 26 posttransplant. For DLI, chimeras were given 3 × 107 B6.PL-Thy1a spleen cells at 28 days after BMT. Survival curves represent the combined data from two experiments.

Close modal

In a second set of experiments, we addressed whether the Thy1+αβTCR+ immunoregulatory donor cells belonged to either the CD4 or CD8 T cell subset. For these experiments, mice with ko mutations for CD4 or CD8 were used as BM donors. AKR recipients were transplanted with T-depleted BM from either normal, CD4 ko, or CD8 ko C57BL/6 mice. The transplanted recipients were again randomized to be treated or not treated with anti-Thy1.2 mAb on days 24 and 26 posttransplant, followed by DLI on day 28 after BMT. A group of recipients given normal T-depleted BM only was included as a GVH-negative control. Mice in the experimental groups treated with anti-Thy1.2 mAb (filled symbols in Fig. 8) developed severe and lethal GVHD after DLI regardless of the BM source. Recipients of normal BM and DLI (but no mAb) lost body weight after DLI (data not shown), but there were no deaths (Fig. 8), indicating that GVH reactivity was relatively mild when the immunoregulatory cells were not depleted. Chimeras transplanted with CD8 ko BM also developed only mild GVHD after DLI as indicated by low mortality (18%) and modest weight loss. Survival of these chimeras was not significantly different from that of the GVH-negative controls (p > 0.05). These results indicate that the Thy1+αβTCR+ immunoregulatory cells do not belong to the CD8 T subset.

FIGURE 8.

Transplantation of BM from CD4 ko donor mice, but not CD8 ko mice, resulted in severe GVHD after DLI. Irradiated AKR recipients were transplanted with 107 T-depleted normal, CD4 ko, or CD8 ko C57BL/6 BM cells. The hosts were randomized to receive no treatment or injected i.p. with 500 μg of anti-Thy1.2 mAb on days 24 and 26 posttransplant. For DLI, chimeras were given 3 × 107 B6.PL-Thy1a spleen cells at 28 days after BMT. Survival curves represent the combined data from two experiments with the exception of the groups transplanted with CD4 ko and CD8 ko BM plus anti-Thy1.2 mAb (n = 5 mice/group), where the data was obtained from one experiment.

FIGURE 8.

Transplantation of BM from CD4 ko donor mice, but not CD8 ko mice, resulted in severe GVHD after DLI. Irradiated AKR recipients were transplanted with 107 T-depleted normal, CD4 ko, or CD8 ko C57BL/6 BM cells. The hosts were randomized to receive no treatment or injected i.p. with 500 μg of anti-Thy1.2 mAb on days 24 and 26 posttransplant. For DLI, chimeras were given 3 × 107 B6.PL-Thy1a spleen cells at 28 days after BMT. Survival curves represent the combined data from two experiments with the exception of the groups transplanted with CD4 ko and CD8 ko BM plus anti-Thy1.2 mAb (n = 5 mice/group), where the data was obtained from one experiment.

Close modal

Chimeras transplanted with CD4 ko BM developed severe GVHD after DLI as indicated by acute body weight loss (data not shown) and 88% mortality (Fig. 8). However, although survival of these chimeras was significantly worse than that of AKR hosts given normal BM and DLI (p < 0.00008), the kinetics of death was delayed as compared with the anti-Thy1.2-treated chimeras, and median survival time was longer by ∼20 days. Because BMT with CD4 ko BM did not completely eliminate the suppression of DLI-induced GVHD, we speculate that there may be two subpopulations of Thy1+αβTCR+ immunoregulatory cells: one of which is CD4+ and the other CD4.

Results shown in Fig. 1 of this paper confirm our previous report (15) that DLI given 21 days after BMT induced lethal GVHD in MHC-mismatched recipients but did not induce GVHD in MHC-matched recipients. Increased immunogenetic disparity in the MHC-mismatched model (i.e., greater alloreactivity) may contribute to the disparate results in these two models. We hypothesized that because of the higher frequency of alloreactive T cells in the MHC-mismatched BMT model, the small numbers of mature T cells in non-T-depleted donor BM intensified GVH reactivity after DLI in MHC-mismatched BM chimeras. This hypothesis was based on our observation that although the number of T cells in the MHC-mismatched marrow grafts were below the threshold needed to induce acute GVHD, transient body weight loss (5–15%) often occurred 3–6 wk after BMT (for examples, see Figs. 1,B and 3B). In contrast, significant body weight loss did not occur in MHC-matched recipients given BM only (Fig. 1,B). The early and transient body weight loss in MHC-mismatched recipients was eliminated if the donor BM was depleted of mature T cells, indicating that T cells were involved. Two different strategies were used to address our hypothesis: 1) ex vivo depletion of donor T cells from the BM graft and 2) in vivo depletion of BM-derived donor T cells with anti-Thy1 mAb. Ex vivo depletion of T cells from the BM graft had no effect on the severity of GVHD after DLI (Fig. 2). To our surprise, when recipient mice were treated in vivo with anti-Thy1.2 mAb after BM but before DLI, GVH-associated mortality and body weight loss were significantly more severe when DLI therapy was given (Figs. 3 and 4). This phenomenon was not unique to the MHC-mismatched donor-host combination. Similar results were observed in an MHC-matched BMT model (Fig. 5). These serendipitous results led us to hypothesize that a population of Thy1+ donor cells developed early after BMT and suppressed GVH reactivity of the infused donor cells. Subsequent studies revealed that the immunoregulatory Thy1+ cells coexpressed αβTCR heterodimers and that an intact host thymus was required for their development (Figs. 6 and 7). Results from experiments using CD4 and CD8 ko mice as BM donors established that αβTCR+CD4+CD8 and αβTCR+CD4CD8 regulatory cells played key roles in suppressing GVH reactivity after DLI (Fig. 8).

Immunoregulatory CD4+ cells have been shown to be involved in BMT tolerance (17), solid organ allograft tolerance (18), and skin graft tolerance (19, 20). There are numerous reports describing the role of CD4+CD8 immunoregulatory cells in various models of autoimmunity (reviewed in Ref. 21). Typically, these cells have been found to express the CD45RB molecule at low levels, which is a characteristic of memory T cells. Unlike typical memory T cells, these cells fail to proliferate in response to polyclonal activation and have tolerogenic properties (22, 23). They become detectable in peripheral lymphoid tissues shortly after birth, and an intact thymus is required for their development (24, 25, 26, 27). Elimination of the CD4+ regulatory cells in both normal and autoimmune-susceptible animals has resulted in the development and/or acceleration of various autoimmune diseases, suggesting that these cells actively suppress autoreactive T cells that escape thymic selection.

To our knowledge, T cells with a CD4+ phenotype have not been previously described as playing an important role in suppressing GVHD after allogeneic BMT. Tutschka et al. (9, 10, 11) published a series of papers in the early 1980s describing the involvement of suppressor cells in immune tolerance after BMT. They reported that histoincompatible rat BM chimeras that survived acute GVHD demonstrated tolerance to host alloantigens and possessed T lymphocytes that suppressed donor-anti-host mixed lymphocyte reactions in vitro (10). Furthermore, chimeric splenocytes containing the suppressor T cells could adoptively transfer suppression of GVHD to secondary hosts. Long-term BM chimeras (250 days after BMT) possessing the suppressor cells did not develop GVHD when given DLI with small numbers of donor-type cells. However, if the chimeras were treated with immunosuppressive agents such as cyclophosphamide or radiation before DLI, tolerance was disturbed and the mice developed GVHD. Perhaps the immunoregulatory Thy1+αβTCR+CD4+/− cells identified by us are similar to the immunosuppressive T cells described in those earlier studies.

Hess and colleagues (28, 29) have demonstrated that elimination of a peripheral regulatory mechanism is required for the development of cyclosporin A (CSA)-induced autologous GVHD. The primary autoregulatory activity appears to reside within the CD4+ T subset. Recently, Wu and Goldschneider (30) demonstrated that the immunoregulatory CD4+CD8 cells are recent thymic emigrants that are exported to the peripheral lymphoid organs within a few days after cessation of CSA treatment. Their results suggest that the regulatory T cells actively suppress autoeffector T cells that have escaped intrathymic negative selection during CSA treatment. Similarly, development of the immunosuppressive regulatory cells described in our studies required an intact thymus (Fig. 6). Modigliani et al. (20, 31) have shown that the thymic epithelium is involved in generating CD4+ regulatory T cells that maintain peripheral tolerance to Ags not present in the thymus. This form of tolerance has been referred to as active or dominant peripheral tolerance (21, 32). Perhaps the αβTCR+CD4+CD8 regulatory cells identified from our work emerge from the reconstituting thymus after BMT to help maintain dominant peripheral tolerance over anti-host-reactive T cells that have escaped thymic selection. Donor leukocytes infused into chimeric hosts after BMT may be seen by the donor-derived, host thymus-selected immunoregulatory T cells as “autoreactive” since they respond to “host-self” but express “donor-self” histocompatibility Ags.

Although administration of CD4 ko donor BM significantly reduced the suppression of DLI-induced GVHD (Fig. 8), the suppressive effect was not completely eliminated. These results suggest that αβTCR+CD4CD8 cells also are involved in the suppression of GVHD after DLI. Cells with this phenotype have been previously found to exhibit suppressor activity, inhibit GVHD, and facilitate tolerance in mice (33, 34, 35). Cells expressing the NK1 Ag with an αβTCR+CD4CD8 phenotype have been shown by others to exhibit immunoregulatory properties (36, 37). The NK1+ cells described by Sykes et al. (36) were able to suppress GVHD. Experiments are in progress to examine whether NK1+αβTCR+ cells contribute to the suppression of GVHD in our DLI models.

It is noteworthy that when the immunoregulatory Thy1+αβTCR+ cells were eliminated before DLI, the kinetics of GVHD-related mortality in MHC-mismatched chimeras was similar to that seen when irradiated hosts were given the donor spleen cells at the time of BMT (15). This suggests that, at 3–4 wk after BMT in mice, active suppression mediated by the immunoregulatory cells described in this report may be the principal mechanism responsible for suppressing GVHD after DLI. We speculate that these immunoregulatory cells are particularly important because they are being actively produced by the newly donor-reconstituted host thymus to mediate dominant peripheral tolerance.

In summary, we have identified an active regulatory/suppressor mechanism in mice which plays an important role in suppressing GVH reactivity after DLI. The regulatory activity appears to be mediated by at least two populations of donor BM-derived, host thymus-selected cells: Thy1+αβTCR+CD4+CD8 cells and Thy1+αβTCR+CD4CD8 cells. These immunoregulatory cells may be important for maintaining dominant peripheral tolerance. If regulatory cells similar to those described in this report can be identified in humans, their immunosuppressive properties could be exploited to reduce the incidence and severity of GVHD following DLI. The possible role of thymic-derived regulatory T cells raises questions about age-related intensification of GVHD after BMT (38, 39) and about the more severe level of GVHD that occurs after DLI in humans (1, 3).

1

This work was supported by U.S. Public Health Service Grants CA71559 and CA39854 and the Midwest Athletes Against Childhood Cancer Fund (Milwaukee, WI).

3

Abbreviations used in this paper: DLI, delayed leukocyte infusion; GVH, graft-vs-host; GVHD, graft-vs-host disease; BM, bone marrow; BMT, bone marrow transplantation; ko, knockout; TCD, T cell depleted; LDA, limiting dilution analysis; Tx, thymectomized; CSA, cyclosporin A.

1
Kolb, H. J., A. Schattenberg, J. M. Goldman, B. Hertenstein, N. Jacobsen, W. Arcese, P. Ljungman, A. Ferrant, L. Verdonck, D. Niederwieser, et al
1995
. Graft-versus-leukemia effect of donor lymphocyte transfusions in marrow grafted patients.
Blood
86
:
2041
2
Slavin, S., E. Naparstek, A. Nagler, A. Ackerstein, S. Samuel, J. Kapelushnik, C. Brautbar, R. Or.
1996
. Allogeneic cell therapy with donor peripheral blood cells and recombinant human interleukin-2 to treat leukemia relapse after allogeneic bone marrow transplantation.
Blood
87
:
2195
3
Collins, R. H., O. Shpilberg, W. R. Drobyski, D. L. Porter, S. Giralt, R. Champlin, S. A. Goodman, S. N. Wolff, W. Hu, C. Verfaillie, et al
1997
. Donor leukocyte infusions in 140 patients with relapsed malignancy after allogeneic bone marrow transplantation.
J. Clin. Oncol.
15
:
433
4
Johnson, B. D., R. L. Truitt.
1995
. Delayed infusion of immunocompetent donor cells after bone marrow transplantation breaks graft-host tolerance and allows for persistent antileukemic reactivity without severe graft-versus-host disease.
Blood
85
:
3302
5
Antin, J., J. Ferrara.
1992
. Cytokine dysregulation and acute graft-versus-host disease.
Blood
80
:
2964
6
Ferrara, J., S. Abhyankar, D. Gilliland.
1993
. Cytokine storm of graft-versus-host disease: a critical effector role for interleukin-1.
Transpl. Proc.
25
:
1216
7
Hill, G. R., J. M. Crawford, K. R. Cooke, Y. S. Brinson, L. Pan, J. L. M. Ferrara.
1997
. Total body irradiation and acute graft-versus-host disease: the role of gastrointestinal damage and inflammatory cytokines.
Blood
90
:
3204
8
Shlomchik, W. D., M. S. Couzens, C. B. Tang, J. McNiff, M. E. Robert, J. Liu, M. J. Shlomchik, S. G. Emerson.
1999
. Prevention of graft versus host disease by inactivation of host antigen-presenting cells.
Science
285
:
412
9
Tutschka, P. J., P. F. Ki, W. E. Beschorner, A. D. Hess, G. W. Santos.
1981
. Suppressor cells in transplantation tolerance. II. Maturation of suppressor cells in the bone marrow chimera.
Transplantation
32
:
321
10
Tutschka, P. J., A. D. Hess, W. E. Beschorner, G. W. Santos.
1981
. Suppressor cells in transplantation tolerance. I. Suppressor cells in the mechanism of tolerance in radiation chimeras.
Transplantation
32
:
203
11
Tutschka, P. J., A. D. Hess, W. E. Beschorner, G. W. Santos.
1982
. Suppressor cells in transplantation tolerance. III. The role of antigen in the maintenance of transplantation tolerance.
Transplantation
33
:
510
12
Shih, C. Y., R. L. Truitt.
1987
. Prophylaxis of graft-vs-host disease by adoptive transfer of a class II antigen-specific murine T-cell clone. R. L. Truitt, and R. P. Gale, and M. M. Bortin, eds.
Cellular Immunotherapy of Cancer
439
Alan R. Liss, New York.
13
Taswell, C..
1981
. Limiting dilution assays for the determination of immunocompetent cell frequencies. I. Data analysis.
J. Immunol.
126
:
1614
14
Johnson, B. D., W. R. Drobyski, R. L. Truitt.
1993
. Delayed infusion of normal donor cells after MHC-matched bone marrow transplantation provides an antileukemia reaction without graft-versus-host disease.
Bone Marrow Transplant.
11
:
329
15
Johnson, B. D., E. E. Becker, R. L. Truitt.
1999
. Graft-vs.-host and graft-vs.-leukemia reactions after delayed infusions of donor T-subsets.
Biol. Blood Marrow Transplant.
5
:
123
16
Mombaerts, P., A. R. Clarke, M. A. Rudnicki, J. Iacomini, S. Itohara, J. J. Lafaille, L. Wang, Y. Ichikawa, R. Jaenisch, M. L. Hooper, et al
1992
. Mutations in T-cell antigen receptor genes α and β block thymocyte development at different stages.
Nature
360
:
225
17
Bemelman, F., K. Honey, E. Adams, S. Cobbold, H. Waldmann.
1998
. Bone marrow transplantation induces either clonal deletion or infectious tolerance depending on the dose.
J. Immunol.
160
:
2645
18
Yin, D., C. G. Fathman.
1995
. CD4-positive suppressor cells block allotransplant rejection.
J. Immunol.
154
:
6339
19
Qin, S., S. P. Cobbold, H. Pope, J. Elliott, D. Kioussis, J. Davies, H. Waldmann.
1993
. “Infectious” transplantation tolerance.
Science
259
:
974
20
Modigliani, Y., A. Cuntinho, P. Pereira, D. L. Douarin, V. Thomas-Vaslin, O. Burlen-Defranoux, J. Salaun, A. Bandeira.
1996
. Establishment of tissue-specific tolerance is driven by regulatory T cells selected by thymic epithelium.
Eur. J. Immunol.
26
:
1807
21
Saoudi, A., B. Seddon, V. Heath, D. Fowell, D. Mason.
1996
. The physiological role of regulatory T cells in the prevention of autoimmunity: the function of the thymus in the generation of the regulatory T cell subset.
Immunol. Rev.
149
:
195
22
Thornton, A. M., E. M. Shevach.
1998
. CD4+CD25+ immunoregulatory T cells suppress polyclonal T cell activation in vitro by inhibiting interleukin 2 production.
J. Exp. Med.
188
:
287
23
Itoh, M., T. Takahashi, N. Sakaguchi, Y. Kuniyasu, J. Shimizu, F. Otsuka, S. Sakaguchi.
1999
. Thymus and autoimmunity: Production of CD25+CD4+ naturally anergic and suppressive T cells as a key function of the thymus in maintaining immunologic self-tolerance.
J. Immunol.
162
:
5317
24
Boitard, C., R. Yasunami, M. Dardenne, J. F. Bach.
1989
. T cell-mediated inhibition of the transfer of autoimmune diabetes in NOD mice.
J. Exp. Med.
169
:
1669
25
Smith, H., Y. Sakamoto, K. Kasai, K. S. K. Tung.
1991
. Effector and regulatory cells in autoimmune oophoritis elicited by neonatal thymectomy.
J. Immunol.
147
:
2928
26
Asano, M., M. Toda, N. Sakaguchi, S. Sakaguchi.
1996
. Autoimmune disease as a consequence of developmental abnormality of a T cell subpopulation.
J. Exp. Med.
184
:
387
27
Suri-Payer, E., A. Z. Amar, A. M. Thornton, E. M. Shevach.
1998
. CD4+CD25+ T cells inhibit both the induction and effector function of autoreactive T cells and represent a unique lineage of immunoregulatory cells.
J. Immunol.
160
:
1212
28
Fischer, A. C., W. E. Beschorner, A. D. Hess.
1989
. Requirements for the induction and adoptive transfer of cyclosporine-induced syngeneic graft-vs-host disease.
J. Exp. Med.
169
:
1031
29
Hess, A. D., A. C. Fischer, L. R. Horwitz.
1993
. Cyclosporine-induced autoimmunity: critical role of autoregulation in the prevention of major histocompatibiltiy class II-dependent autoaggression.
Transplant. Proc.
25
:
2811
30
Wu, D. Y., I. Goldschneider.
1999
. Cyclosporin A-induced autologous graft-versus-host disease: a prototypical model of autoimmunity and active (dominant) tolerance coordinately induced by recent thymic emigrants.
J. Immunol.
162
:
6929
31
Modigliani, Y., P. Pereira, V. Thomas-Vaslin, J. Salaun, O. Burlen-Defranoux, A. Coutinho, N. L. Douarin, A. Bandeira.
1995
. Regulatory T cells in thymic epithelium-induced tolerance. I. Suppression of mature peripheral non-tolerant T cells.
Eur. J. Immunol.
25
:
2563
32
Modigliani, Y., A. Bandeira, A. Coutinho.
1996
. A model for developmentally acquired thymus-dependent tolerance to central and peripheral antigens.
Immunol. Rev.
149
:
155
33
Bruley-Rosset, M., I. Miconnet, C. Canon, O. Halle-Pannenko.
1990
. Mlsa generated suppressor cells. I. Suppression is mediated by double-negative (CD3+CD5+CD4CD8) α/β T cell receptor-bearing cells.
J. Immunol.
145
:
4046
34
Abraham, V. S., D. H. Sachs, M. Sykes.
1992
. Mechanism of protection from graft-versus-host disease mortality by IL-2. III. Early reductions in donor T subsets and expansion of a CD3+CD4CD8 cell population.
J. Immunol.
148
:
3746
35
Strober, S., L. Cheng, D. Zeng, R. Palathumpat, S. Dejbakhsh-Jones, P. Huie, R. Sibley.
1996
. Double negative (CD4CD8αβ+) T cell which promote tolerance induction and regulate autoimmunity.
Immunol. Rev.
149
:
217
36
Sykes, M., K. A. Hoyles, M. L. Romick, D. Sachs.
1990
. In vitro and in vivo analysis of bone marrow-derived CD3+,CD4,CD8,NK 1.1+ cell lines.
Cell. Immunol.
129
:
478
37
Tamada, K., M. Harada, K. Abe, T. Li, H. Tada, Y. Onoe, K. Nomoto.
1997
. Immunosuppressive activity of cloned natural killer (NK1.1+) T cells established from murine tumor-infiltrating lymphocytes.
J. Immunol.
158
:
4846
38
Weisdorf, D., R. Hakke, B. Blazar, W. Miller, P. McGlave, N. Ramsay, J. Kersey, A. Filipovich.
1991
. Risk factors for acute graft-versus-host disease in histocompatible donor bone marrow transplantation.
Transplantation
51
:
1197
39
Bryson, J. S., C. D. Jennings, B. E. Caywood, A. R. Dix, D. M. Lowery, A. M. Kaplan.
1997
. Enhanced graft-versus-host disease in older recipient mice following allogeneic bone marrow transplantation.
Bone Marrow Transplant.
19
:
721