DELAYING TRANSPLANTATION AFTER TOTAL BODY IRRADIATION IS A... : Transplantation (original) (raw)
Acute graft-versus-host disease (GVHD*) remains a major complication of bone marrow transplantation (BMT). Although T cells play an important role in the initiation of acute GVHD, there is a large body of evidence that other cells and inflammatory cytokines are involved in its pathogenesis (1). We have previously reported that total body irradiation (TBI) conditioning induced significant inflammatory cytokine release, e.g., tumor necrosis factor-α (TNF-α), interleukin (IL)-1β, and IL-6 in severe combined immunodeficiency (SCID) mice (2). The TBI related inflammatory cytokine release was markedly augmented by major histoincompatible allogeneic cell transplantation, and the animals all died of acute GVHD within 2 weeks. When we separated the TBI (day 0) and transplantation (day 4), acute GVHD rates, mortality was markedly reduced. Because these experiments were performed in genetically T- and B-cell deficient SCID mice, it was felt the results might not mimic clinical bone marrow transplantation, particularly with respect to engraftment in immunocompetent hosts. We now report the effect of day 0 TBI and day 4 transplantation on acute GVHD mortality, engraftment and a mechanism whereby delaying transplantation reduced acute GVHD in immunological intact BALB/c mice transplanted with major H2 incompatible C57BL/6 bone marrow and spleen cells.
MATERIALS AND METHODS
Mice. Donor C57BL/6 (H2b) and recipient BALB/c (H2d) were purchased from Harlan Sprague Dawley Co., Indianapolis, IN (8-12 weeks old) and kept in-house for 1 week before experiments were initiated.
Cell preparation. Bone marrow cells were obtained by flushing the femoral bones with normal saline and spleen cells were obtained by crushing and flushing the excised spleen through fine steel sieves. The preparations were performed under sterile conditions. The cells were suspended in saline and adjusted to the concentration of 1×108/ml. The bone marrow cells and spleen cells were mixed 1:1 before injection.
Conditioning and transplantation. Although the studies reported here are derived from two large experiments, multiple pilot studies preceded these definitive studies. BALB/c mice were conditioned with lethal dose (8.5 Gy) TBI (Mark I137 cesium irradiator, JL Shepherd, Glendale, CA) at 1 Gy/min on day 0 and injected i.v. with 0.5 ml of saline or 0.5 ml of bone marrow and spleen cell mixture (2.5×107 bone marrow cells plus 2.5×107 spleen cells) as follows: (1) TBI on day 0 and transplantation of allogeneic C57BL/6 cells on the same day (TBI+day 0 allogeneic cells), n=15; (2) TBI on day 0 and transplantation of C57BL/6 cells on day 4 after TBI (TBI+day 4 allogeneic cells), n=15; (3) TBI and transplantation of syngeneic BALB/c cells on the same day (TBI+syngeneic cells), n=5; and (4) TBI without transplantation (TBI only), n=5. All experimental animals were housed in the nonsterile open breeding conditions routinely used in the animal facility. Body weight and mortality were monitored every other day for 60 days. However, the animals were kept for 100 days after transplantation if they were still alive at day 60 for evaluation of long-term engraftment. Engraftment was measured by flow cytometric analysis of the recipient bone marrow and spleen either at the time of animal death or on day +60 or +100 after transplantation. Because of the time difference of the TBI+day 0 and TBI+day 4 protocols, we defined day 1 as 1 day after TBI, day +1 as 1 day after transplantation; day 15 as 15 days after TBI, day +15 as 15 days after transplantation and thereafter.
Cytokine assay. Other groups of BALB/c mice (n=20 in each group) were treated with the above protocols or TBI without cell transplantation as controls (TBI only). Sterile blood was obtained by percutaneous cardiac puncture after CO2 euthanasia from these animals with or without transplantation before and after conditioning at 4 hr, 24 hr, 48 hr, 72 hr, day 5, and day 10 (n=3 at each time point). TNF-α (mini-kit from Pharmingen, San Diego, CA; sensitivity 25 pg/ml), IL-1β (mini-kit from Endogen, Inc., Cambridge, MA; sensitivity 3 pg/ml), interferon gamma (IFN-γ) (mini-kit from Endogen; sensitivity 50 pg/ml), and IL-6 (mini-kit from Endogen; sensitivity 5 pg/ml) were measured by enzyme-linked immunosorbent assays. The quick and nontraumatic cardiac punctures were important to increase the TNF sensitivity. We have found that open chest or any other traumatic punctures resulted in a large release of TNF receptor, which interfered with the TNF-α assay (data not shown). There are no detectable TNF-α, IL-1, IL-6, or IFN-γ in naive normal animals.
Histology. The liver, spleen, skin, and small and large bowel were obtained at the time of animal death or day +60 after transplantation. The tissues were fixed in 10% formalin for 48 hr and embedded in paraffin, sectioned, and stained with hematoxylin and eosin (Sigma Chemical Co., St. Louis, MO).
Flow cytometry to detect engraftment. The bone marrow and spleen cells were washed twice with saline and then stained with anti-H2b (staining for C57BL/6 cells) or H2d (staining for BALB/c cells) monoclonal antibody conjugated with phycoerythrin and anti-CD3 monoclonal antibody conjugated with fluorescein isothiocyanate for 20 min on ice. Mouse IgG-phycoerythrin and IgG- fluorescein isothiocyanate were used for control antibody staining (all antibodies used in the staining were purchased from Pharmingen). The cells were washed 2 times after staining and two color fluorescence was measured by a FACScan flow cytometer (Becton Dickinson, Sunnyvale, CA).
Statistical analysis. Double-tailed Student's t tests were used for all data analysis.
RESULTS
Mortality. All BALB/c mice treated with TBI+day 0 allogeneic C57BL/6 transplantation developed classic acute GVHD and died by day +10. The survival rate on day +60 was 0% in the TBI+day 0 group versus 66% in the TBI+day 4 group after major H2 incompatible C57BL/6 transplantation (P<0.001) (Fig. 1). As expected, all syngeneic BALB/c transplanted control mice survived over 60 days. All mice received TBI only treatment died between day 16 and 30 after the lethal radiation.
Body weight. Aside from mortality, body weight is another sensitive parameter of acute GVHD. All mice lost weight in the first week after TBI (Fig. 2). The TBI+day 0 allogeneic C57BL/6 transplanted animals had the greatest weight loss in the first week and continued to lose weight to the time of death. The syngeneic control animals had the least weight loss in the first week after TBI, began to recover in the second week and completely recovered their weight by day +30. The TBI+day 4 allogeneic C57BL/6 transplanted animals developed weight loss similar to that of the syngeneic controls in the first 10 days after TBI. However, these animals began to lose weight again from day 10 to day 14 after TBI (day +6 to day +10 after transplantation). Although the body weight in this group of animals recovered somewhat in the next 1 and a half months, unlike the syngeneic control animals, they never quite recovered to baseline body weight (P<0.05). Histology of bowel, skin, and liver by day +60 showed chronic GVHD changes (data not shown). The five animals that died within 2 weeks after TBI+day 4 allogeneic transplantation lost more weight compared with that of surviving animals at the time of death. However, the difference did not achieve statistical significance due to the small number of animals. Autopsy and histology showed typical acute GVHD in the bowel and liver.
Engraftment. Flow cytometry of BALB/c bone marrow cells revealed 95-98% donor H2b + C57BL/6 cells in TBI+day 0 allogeneic transplanted animals by the time of death (day +7 to +10 after transplantation). In the TBI+day 4 allogeneic transplanted animals, 55-60% donor H2b + cells were present in the recipient marrow by day +10 to +15 after transplantation and increased to 70-80% by day +60 and 89-98% by day +100. (Table 1). Their spleens were enlarged and infiltrated with multilineage hematopoietic cells (myeloid cells, erythroid cells, and megakaryocytes) and lymphocytes by 2 weeks after transplantation. No animal with allogeneic transplantation failed to engraft. Of note, the percentage of CD3+ T cells in the engrafted donor cell population demonstrated by two-color flow cytometry was virtually identical between the TBI+day 0 and TBI+day 4 allogeneic transplant animals (Table 1). Both groups received essentially the same amount of T cell inoculation. Because it is technically very difficult to wash all the cells out of the femur or the spleen, it is virtually impossible to make accurate absolute cell yields from a femur or spleen. The percentage of donor cells in the marrow and spleen from the same animals was almost identical at all time point (spleen data not shown).
Inflammatory cytokine release in TBI conditioning and transplantation. Serum TNF-α, IL-1β, IL-6, and IFN-γ were measured in the four groups of animals: (1) TBI only; (2) TBI+day 0 allogeneic transplantation; (3) TBI+day 4 allogeneic transplantation; and (4) TBI + syngeneic transplanted controls. In the TBI only animals, serum TNF-α rose 4 hr after TBI conditioning, followed by IL-1β (24 hr) and IL-6 (48 hr). These cytokines rapidly declined within 72 hr and almost disappeared by 4 days after TBI conditioning if transplantation did not occur (Fig. 3). However, when the animals received TBI+day 0 allogeneic histoincompatible transplantation, serum TNF-α continued to rise 10-fold higher compared with that of the TBI only or syngeneic control groups (Fig. 4A). IL-1β and IL-6 levels in TBI+day 0 allogeneic transplant animals were also markedly elevated similar to TNF-α (Fig. 4, B and C). The differences of TNF-α, IL-1β and IL-6 levels between the syngeneic control and TBI+day 0 allogeneic groups were all very significant (P<0.01). When BALB/c hosts received histoincompatible transplantation 4 days after TBI conditioning (TBI+day 4 allogeneic), we did not see the very marked cytokine elevations after transplantation that occurred in the TBI+day 0 allogeneic transplanted animals. Although IL-1β remained somewhat high compared with the syngeneic control animals (P<0.05), the overall kinetic patterns of TNF-α, IL-1β, and IL-6 in the TBI+day 4 group were similar to that for the TBI only and syngeneic control groups (Fig. 4, A-C). The kinetic pattern of IFN-γ was different from TNF-α, IL-1β, and IL-6. There was no detectable serum IFN-γ in the animals treated with TBI only or syngeneic transplantation. Serum IFN-γ was elevated only after allogeneic transplantation and peaked about day +5 after transplantation (Fig. 4D). Although IFN-γ was higher in TBI+day 4 compared with the TBI+day 0 allogeneic transplant animals, it did not achieve statistical significance.
DISCUSSION
Conditioning regimens in BMT are necessary for successful engraftment and may also have some antitumor effect. Unfortunately, many data in humans and animals have shown that acute GVHD is closely associated with the intensity and toxicity of the conditioning regimens (3). The key is how to use the conditioning regimens properly. TBI and large doses of cyclophosphamide (4, 5) have been shown to induce host inflammatory cytokine release. These cytokines cause or aggravate host tissue damage (6) and modulate major histocompatibility complex antigen expression (7). Histoincompatible transplantation during this sensitive period may further activate effector cells, e.g., monocytes, macrophages, and natural killer cells and augment the inflammatory cytokine release (could be from both host and donor cells). This synergistic inflammatory cell cytokine reaction cycle induced by conditioning (host origin) and incompatible transplantation (donor and/or host) may result in further tissue damage and lethal acute GVHD. Giving recipients a 4-day complete rest between conditioning and histoincompatible transplantation to interrupt the synergistic host-donor cytokine reaction cycle seems to be a simple and effective method reducing tissue damage and acute GVHD mortality.
In human BMT, fractional TBI is widely used now and many transplantation centers have switched TBI to the front end of conditioning followed by chemotherapy to decrease the toxicity of conditioning. Theoretically, this may relate to a decreased host-donor cytokine response induced by TBI and allogeneic transplantation. Unfortunately, there are no published clinical data that compares outcome or cytokine release of TBI administered at the front versus later in the conditioning protocol. In addition to TBI, large doses of cyclophosphamide have also been shown to induce host serum TNF-α elevation in human pretransplant conditioning (4). Schwaighofer et al. observed in humans that serum IFN-γ increased after cyclophosphamide (but not after TBI) on day 6/8 after BMT in T-depleted hosts but returned to base line levels, whereas it did not rise until day 17/17 and remained high in non-T-depleted hosts (5). Lehnert and Rybka reported that separating cyclophosphamide (60 mg/kg) conditioning and parental murine allogeneic lymphoid cell injection by a 7-day interval significantly reduced acute GVHD and mortality (8). Whether other common conditioning drugs, e.g., busulfan and etoposide, may also induce host inflammatory cytokine release in human BMT is not clear. Although our previous animal study showed that BuCy2 (busulfan 16 mg/kg plus cyclophosphamide 100 mg/kg) conditioning regimen did not induce host inflammatory cytokine elevation in SCID mice (2), the body surface converting factors are very different between the human and mouse. The relative BuCy doses we used in mice were lower than that are used in humans.
Delaying transplantation by 4 days did not affect the final engraftment. This finding has been proven in both our allogeneic and xenogeneic (human-mouse) models (9). Delaying transplantation slowed but did not impair eventual complete engraftment compared with immediate transplantation. This might be related to the lower levels of TNF-α and IL-6, both of which have been shown to stimulate lymphohematopoiesis (10, 11). In human BMT, the speed of engraftment is roughly correlated with the degree of donor-host disparity, and mismatched allogeneic engraftment is usually faster than autologous engraftment. Nevertheless, the somewhat slower engraftment seen in autologous transplantation is not associated with an unusually high frequency of infection. In our experiment, the slower partial engraftment (mix chimeras) in the early months but steadily increasing the engraftment in TBI+day 4 allogeneic transplanted animals was associated with much higher survival rates compared with early aggressive complete engraftment (complete chimeras) in TBI+day 0 transplanted animals. TBI+day 0 allogeneic transplanted animals, with 97% donor cell engraftment on day +7 to +10, died even faster than that of the animals who received TBI without transplantation. Whereas TBI+day 4 transplanted animals, with about 50% donor cell engraftment in first 2 weeks and mixed chimerism for about 2-3 months after transplantation, had much higher survival rates. Huss et al. (12) demonstrated that the TBI dosage used for conditioning was inversely correlated with the development of mixed chimerism among chronic myeloid leukemia (CML) patients. The incidence of grade II-IV acute GVHD was lower and survival was significantly higher in mixed chimeras than in complete chimeras after allogeneic BMT in aplastic anemia patients given single agent GVHD prophylaxis.
Whether the mixed chimerism in the early months after TBI+day 4 allogeneic transplantation would affect the graft-versus-leukemic effect (GVL) is another issue. Many studies have demonstrated that GVL is more associated with chronic GVHD (13) than acute GVHD and that GVHD and GVL are associated but separable phenomenon (14, 15). Among the patients with CML, both overall survival and relapse-free survival were significantly superior in mixed than in complete chimeras. The mixed chimerism was not uniformly associated with graft failure or leukemic relapse (12). In our experiment, delaying transplantation did not prevent chronic GVHD after the transplantation across a major histocompatibility barrier.
Although T cells have been thought to be very important in the pathogenesis of acute GVHD and graft failure after T-cell depletion, we did not see any differences in the percentage of T-cell engraftment in recipient marrow and spleen between the TBI+day 0 and TBI+day 4 allogeneic transplantation nor was there any significant difference in the marked elevation of IFN-γ. Neither TBI alone nor TBI + syngeneic transplantation stimulated IFN-γ release indicating IFN-γ correlated with alloantigen induced proliferation and activation and was not directly related to acute GVHD mortality. This suggested that other cells may be involved in the final tissue damage in acute GVHD. Delaying transplantation markedly reduced the TBI-induced TNF-α, IL-1β, and IL-6 inflammatory cytokine response made primarily by non-T cells, because nearly identical elevations after TBI were observed in T- and B-cell deficient SCID mice (1). It has been reported that IFN-γ may have a priming or enhancing effect on lipopolysaccharide (LPS)-induced TNF-α lethality (16), and macrophage transcription of the TNF-α and IL-1 genes (17). That the outpouring of IFN-γ did not enhance TNF-α, IL-6, or IL-1β release in the TBI+day 4 transplanted mice suggests that the early response of monocytes-macrophages to TBI may have made them less responsive to subsequent IFN-γ priming. This phenomenon could be similar to endotoxin tolerance whereby monocytes become relatively refractory to endotoxin rechallenge if they have been exposed to this agent before and the expected stimulation of macrophage production of TNF-α is markedly diminished (18, 19). Finally, it is important to note that IFN-γ may be a major factor in the development of the immune suppression and lymphoid hypoplasia associated with chronic GVHD (20). TBI+day 4 allogeneic transplantation did not prevent the late development of this reaction.
In conclusion, we have demonstrated that a 4-day rest between conditioning and transplantation is a simple and effective method to reduce acute GVHD mortality. During this period, the elevated host inflammatory cytokines, TNF-α, IL-1, and IL-6, stimulated by TBI begin to decline toward baseline. The marked elevation after TBI and allogeneic cell transplantation on the same day is largely aborted. The delayed transplantation with lower levels of TNF-α and IL-6 produced early mixed chimeras then complete chimeras within 2-3 months after transplantation and were associated significantly greater survival than immediate transplantation and early complete chimerism
Survival rate. BALB/c mice were conditioned with 8.5 Gy TBI and transplanted with 2.5×107 bone marrow plus 2.5×107 spleen cells as follows: (1) TBI on day 0 and transplanted with C57BL/6 allogeneic cells on the same day (TBI+day 0 allogeneic cells); (2) TBI on day 0 and transplanted with C57BL/6 allogeneic cells on day 4 after TBI (TBI+day 4 allogeneic cells); (3) TBI on day 0 and transplanted with syngeneic BALB/c cells on the same day (TBI+syngeneic cells); (4) TBI without transplantation (TBI only): 15 of 15 TBI+day 0 allogeneic transplanted mice died of acute GVHD by day +10, whereas only 5 of 15 TBI+day 4 allogeneic and 0 of 5 syngeneic transplanted mice died of acute GVHD within the 60 day observation period. Survival rates were 66% in TBI+day 4 allogeneic transplanted animals versus 0% in TBI+day 0 allogeneic transplanted animals (P<0.001) (combination of two experiments).
Body weight changes. The same experiment as described in Figure 1. All animals lost weight during the first week after TBI conditioning. The TBI only animals lost their most weight during first week after the TBI, and continued losing weight slowly until death of bone marrow failure. The syngeneic controls began to recover on the second week and completely recovered to baseline by day +30. TBI+day 0 allogeneic transplanted mice continued to lose weight rapidly to the time of death. TBI+day 4 animal weight loss paralleled that of the syngeneic controls in the first 10 days, but they began to lose more weight 10 days after TBI (+6 days after allogeneic transplantation) and never recovered to baseline during the 60-day experimental period, P<0.05 compared with syngeneic animals.
Serum TNF-α, IL-1β, IL-6, and IFN-γ levels after TBI conditioning only. TBI induced host TNF-α (peak at 4 hr), IL-1β (peak at 24 hr), and IL-6 (peak at 48 hr) elevations. These elevated cytokines quickly declined nearly to baseline by 72 hr. TBI did not induce IFN-γ elevation.
Serum TNF-α, IL-1β, IL-6, and IFN-γ levels after TBI and transplantation. Serum TNF-α (A), IL-1β (B), and IL-6 (C) were markedly elevated in TBI+day 0 allogeneic transplanted animals compared with the syngeneic or TBI only controls (P<0.01 with all three cytokines and also note of the different scale on the y axis). This phenomenon was not found in TBI+day 4 allogeneic transplanted animals. IFN-γ (D) was elevated only in the animals with allogeneic transplantation and peaked in 5 days after transplantation. There was no significant difference in the TBI+day 0 vs. TBI+day 4 allogeneic transplanted animals. IFN-γ was not detectable in TBI only or syngeneic transplanted animals.
Footnotes
This work was supported by a Veterans Affairs merit review grant.
Abbreviations: BMT, bone marrow transplantation; CML, chronic myeloid leukemia; GVHD, graft-versus-host-disease; GVL, graft-versus-leukemia; IFN, interferon; IL, interleukin; LPS, lipopolysaccharide; SCID, severe combined immunodeficiency; TBI, total body irradiation; TNF, tumor necrosis factor.
REFERENCES
1. Antin JH, Ferrara JLM. Cytokine dysregulation and acute graft versus host disease. Blood 1992; 80: 2964.
2. Xun CQ, Thompson JS, Jennings CD, Brown SA, Widmer MB. Effect of total body irradiation, busulfan-cyclophosphamide or cyclophosphamide conditioning on inflammatory cytokine release and development of acute and chronic graft versus host disease in H-2 incompatible transplanted SCID mice. Blood 1994; 83: 2360.
3. Clift RA, Buckner CD, Appelbaum FR, et al. Allogeneic marrow transplantation in patients with acute myeloid leukemia in first remission: a randomized trial of two irradiation regimens. Blood 1990; 76: 1867
4. Holler E, Kolb HJ, Mittermuller J, et al. Modulation of acute graft-versus-host-disease after allogeneic bone marrow transplantation by tumor necrosis factor alpha (TNF alpha) release in the course of pre-transplant conditioning: role of conditioning regimens and prophylactic application of a monoclonal antibody neutralizing human TNF alpha (MAK 195F). Blood 1995; 86: 890.
5. Schwaighofer H, Kernan NA, O'rally RJ, et al. Serum levels of cytokines and secondary messages after T-cell-depleted and non-T-depleted bone marrow transplantation: influence of conditioning and hematopoietic reconstruction. Transplantation 1996; 62: 947.
6. Wang JH, Redmond HP, Wastson RW, Bouchier-Hayes D. Role of lipopolysaccharide and tumor necrosis factor alpha in induction of hepatocyte necrosis. Am J Physiol 1995; 269: G297.
7. Gabbianelli M, Boccole G, Cianetti L, Russo G, Testa U, Peschle C. HLA expression in hematopoietic development. Class I and II antigens are induced in the definitive erythroid lineage and differentially modulated by fetal liver cytokines. J Immunol 1990; 144: 3354.
8. Lehnert S, Rybka WB. Amplification of the graft-versus-host reaction by cyclophosphamide: dependence on timing of drug administration. Bone Marrow Transplant 1994; 13: 473.
9. Tsuchida M, Brown SA, Seehafer D, Tan J, Thompson JS. In vivo xenogeneic murine model: response and immunosuppressive effect of anti-T cell antibody on human skin rejection. Clin Transpl (in press).
10. Kishimoto T. The biology of interleukin-6. Blood 1989; 74: 1.
11. Kossodo S, Grau GE, Daneva T, et al. Tumor necrosis factor alpha is involved in mouse growth and lymphoid tissue development. J Exp Med 1992; 176: 1259.
12. Huss R, Deeg HJ, Gooley T, et al. Effect of mixed chimerism on graft-versus-host disease, disease recurrence and survival after HLA-identical marrow transplantation for aplastic anemia or chronic myelogenous leukemia. Bone Marrow Transplant 1996; 18: 767.
13. Sullivan KM, Weiden PL, Storb R, et al (The Seattle bone marrow transplant team). Influence of acute and chronic GVHD on relapse and survival after BMT from HLA-identical siblings as treatment of acute and chronic leukemia. Blood 1989; 73: 1720.
14. Xun CQ, Thompson JS, Jennings CD, Brown SA. The effect of human IL-2 activated natural killer and T cells on graft-versus-host disease and graft-versus-leukemia in SCID mice bearing human leukemic cells. Transplantation 1995; 60: 821.
15. Horowitz MM, Gale RP, Sondel PM, et al. Graft-versus-leukemia reaction after bone marrow transplantation. Blood 1990; 75: 555.
16. Nestel FP, Price KS, Seemayer TA, Lapp WS. Macrophage priming and lipopolysaccharide-triggered release of tumor necrosis factor alpha during graft-vs-host disease. J Exp Med 1992; 175: 405.
17. Collart MA, Belin D, Vassallli JD, Kossodo S, Vassalli P. Gamma interferon enhanced macrophage transcription of the tumor necrosis factor/cachectin, interleukin 1 and urokinase genes which are controlled by short-live repressors. J Exp Med 1986; 164: 2113.
18. Neta R, Oppenheim JJ, Schrieber RD, Chizzonite R, Ledney GD, MacVittie TJ. Role of cytokines (IL-1, tumor necrosis factor and transforming growth factor B) in natural and lipopolysaccharide enhanced radioresistance. J Exp Med 1991; 173: 1177.
19. Ziegler-Heitbrock HW, Wedel A, Schraut W, et al. Tolerance to lipopolysaccharide involves of P50 homodimers. J Biol Chem 1994; 269: 1700.
20. Klimpel GR, Annable CR, Cleveland MG, Jerrells TR, Patterson JC. Immunosuppression and lymphoid hypoplasia associated with chronic graft versus host disease is dependent upon IFN gamma production. J Immunol 1990; 144: 84.
© Williams & Wilkins 1997. All Rights Reserved.