Kinetic analysis of cytokine gene expression in patients with GVHD after donor lymphocyte infusion

Article metrics


Patients who receive a donor lymphocyte infusion (DLI) for the treatment of relapsed leukemia after allogeneic BMT (alloBMT) often developed GVHD. To determine whether cytokines might have a role in GVHD, an intensive kinetic analysis of in vivo cytokine gene expression was performed on PBMC from three such patients. Expression of IL-1β, IL-2, IFN-γ, IL-4, IL-5, IL-8, IL-10, IL-12, TNF-α, and IL-2Rα was examined using a sensitive semi-quantitative reverse transcription (RT)-PCR assay system. Six normal controls were also analyzed for comparison. Expression of type 1 T helper (Th1) cytokines, IL-2 and IFN-γ was greatly increased in all three patients. In particular, the changes in IL-2 gene expression correlated well with disease progression, suggesting that IL-2 has a critical role in the development of GVHD. Although the pattern of type 2 T helper (Th2) cytokine gene expression differed in each patient, the expression of IL-4 was inversely related to expression of Th1 cytokines. These results suggest that Th1 dominates in the development of human clinical GVHD. Bone Marrow Transplantation (2001) 27, 373–380.


GVHD, a major complication of allogeneic BMT (alloBMT), is caused by immune responses of donor lymphocytes against host alloantigens.1 It has been suggested that dysregulation of cytokines plays an important role in the development of graft-versus-host disease (GVHD).234567 Ferrara8 recently proposed that cytokine dysregulation occurs in three sequential phases. The conditioning of the host by irradiation or high-dose chemotherapy induces inflammatory processes in recipient tissues (phase 1). Donor T cells are activated by host alloantigens and secrete cytokines under the influence of the inflammatory process initiated in phase 1 (phase 2). The T cell-derived cytokines of phase 2 activate various inflammatory mediators, cytotoxic T cells, and natural killer cells which culminates in the destruction of the host tissues (phase 3).

Mature T helper cells polarize into two subsets with contrasting and cross-regulating cytokine profiles, including type 1 (IL-2, IFN-γ) and type 2 (IL-4, IL-10).910 The secretion of type 1 cytokines is generally associated with activation of macrophages, inflammatory cytokine production, and the activation of cytotoxic T cells and natural killer cells, whereas the secretion of type 2 cytokines is associated with the down-regulation of cell-mediated immune responses.11 Although the differential activation of these two subsets, type 1 T helper (Th1) and type 2 T helper (Th2) cells, has been implicated in murine models of GVHD,12 it is unclear whether the Th1/Th2 paradigm is relevant to human clinical GVHD.

Because of the extreme heterogeneity of the course of GVHD development among different BMT recipients, we performed a longitudinal analysis of constitutive cytokine expression in PBMC samples collected from the same patient at different time-points during the progression of the disease. Patients who received a donor lymphocyte infusion (DLI) for treating the recurrence of leukemia after alloBMT, were intensively analyzed since these patients were relatively free from influences of irradiation, chemotherapy and major infections. We analyzed the expression of cytokine genes (IL-1β, IL-2, IL-2R, IL-4, IL-5, IL-8, IL-10, IL-12, IFN-γ and TNF-α) using a semi-quantitative reverse transcription PCR (RT-PCR) assay system. This analysis offered an opportunity to observe changes in the pattern of cytokine gene expression occurring over time in the same individual.

Materials and methods


Three patients who received DLI for the treatment of relapsed leukemia after alloBMT from HLA-matched sibling donors were selected in the present study. Patient 1 was a 30-year-old male with AML, M2. Although he received extensive chemotherapy, he did not achieve CR. BMT was then performed. A relapse was observed after 120 days. At this time, 40.7% of the cells in the bone marrow were blastic cells. He was treated with re-induction chemotherapy, but about 1% of the blastic cells remained. He was then treated with three DLIs (a total of 2.4 × 108 T cells/kg) and developed acute GVHD, which affected the skin and gastro-intestinal tract. Patient 2, was 22-year-old female with AML, M2. After her first CR by combination chemotherapy, BMT was performed. Her disease relapsed after 150 days with 69.2% of blastic cells in the bone marrow. Although she received one course of re-induction chemotherapy, 1% of the blastic cells remained. For the treatment of residual disease, she received three DLIs (a total of 4.9 × 108 T cells/kg) and developed acute GVHD, which affected the skin and liver. Patient 3 was a 20-year-old male with ALL, L2. AlloBMT was performed during the second CR period. Cytogenetic relapse was observed 60 days after BMT. Immunosuppressive therapy was immediately discontinued and he was given three DLIs (a total of 2.5 × 108 T cells/kg) and developed GVHD, which affected the skin and liver, along with sicca syndrome. None of the patients had a hepatitis B or C viral infection or other major infections during the observation periods.

Mononuclear cell culture

PBMC from healthy donors and patients were isolated from heparinized fresh blood using Ficoll–Paque gradient centrifugation. To obtain positive controls for the expression of different cytokines, a mixture containing mononuclear cells of a healthy donor at a density of 5 × 106 cells/ml was cultured with one of three stimulants in RPMI-1640 medium, supplemented with 10% FBS for 18 h. For IL-2, IL-2R, IFN-γ, IL-4, IL-5, and IL-10, the stimulant was phytohemagglutinin (PHA, 10 μg/ml), for IL-1, IL-8, and TNF-α it was lipopolysaccharide (LPS, 1 μg/ml), and for IL-12, it was Staphylococcus aureus Cowan-I (SAC, 0.0075%). Cells were harvested and washed twice with PBS, and then total RNA was extracted.

Analysis of cytokine mRNA

Total RNA was extracted from the isolated PBMC using TRIzol (Life Technologies, Gaithersburg, MD, USA). One μg of each total RNA was then reverse transcribed using a random hexamer and a SuperScript pre-amplification system (Life Technologies). An aliquot of 1/20th of the resulting cDNA was used for semiquantitative PCR. The primers used for amplification are shown in Table 1. Selection of primers was performed using the GeneWorks software program (IntelliGenetics, Mountain View, CA, USA). The cDNA was amplified in 50 μl of PCR buffer containing 1.5 mM MgCl2, and 200 nM dNTP using Ex-Taq DNA polymerase (TaKaRa, Kyoto, Japan). The appropriate annealing temperatures for specific primers, and different numbers of cycles of amplification (25–45) were selected for the different cytokines. Other conditions were the same in all PCR reactions. After an initial denaturation for 3 min at 94°C, each cycle consisted of denaturation for 10 s at 92°C, annealing for 30 s and extension for 60 s at 72°C. A 10 μl aliquot of amplified products was then analyzed on a 2% Metaphor agarose gel. Visualization and incorporation of the gel image was performed using an ImageMaster VDS (Pharmacia Biotech, Uppsala, Sweden) after ethidium bromide staining. IFN-γ, IL-1β, TNF-α, and GAPDH could be detected in this way. The images obtained were analyzed by Phoretix software program (Phoretix, Newcastle upon Tyne, UK) to obtain densitometric data. PCR of serial two-fold dilutions of positive control cDNA was performed to obtain a standard curve for the expression of each cytokine. Based on the standard curve, the relative expression level of each cytokine in each sample was calculated and corrected by the GAPDH expression level in the respective sample.

Table 1  Primers for cytokine genes

Southern blotting

After ethidium bromide staining, the amplified cytokine gene products (IL-2, IL-2R, IL-4, IL-5, IL-8, IL-10, IL-12) were detected by Southern blotting. Briefly, the gels were alkaline denatured and the PCR products were then transferred to a nylon membrane (Boehringer Mannheim, Mannheim, Germany) using 10× SSC buffer. The membrane was then crosslinked using a UV cross linker (UV Stratalinker 1800; Stratagene, La Jolla, CA, USA). The membrane was pre-hybridized at 68°C for 1 h and hybridized overnight at the same temperature with a digoxigenin-labeled probe for each cytokine. To prepare the hybridization probes, positive control cDNA was amplified and labeled using a digoxigenin-labeled dNTP mix (Boehringer Mannheim). Primers for the probes are localized within the respective cytokine gene sequences amplified by the sample amplification primers (Table 2). The membrane was then washed and the bands were detected using substrate CDP-star (Boehringer Mannheim), according to the manufacturer's protocol. The bands were visualized and chemiluminescent signals were measured using a Luminograph (LB 980, EG&G, Berthhold, Germany). The relative expression level of each cytokine gene was calculated and corrected by the GAPDH expression level in the respective sample.

Table 2  Primers for hybridization probes


Expression of cytokine genes in PBMC from normal subjects

The expressions of cytokine genes in PBMC from six normal healthy donors are shown in Figure 1. To quantitate the expression levels, the expressions of these cytokines in serial two-fold dilutions of stimulated PBMC of a healthy donor are also shown. Significant expression of Th1 cytokines (IL-2 and IFN-γ) and IL-2Rα was not detected in any of the normal controls by this assay. Among the Th2 cytokines, IL-4 and IL-10 were detectable at various levels while IL-5 was not detectable in the normal controls. Among the monokines, IL-1β and TNF-α were detectable in all six normal controls. IL-8 was detected at various levels and IL-12 was barely detectable.

Figure 1

Electrophoretic gels showing the expression of cytokine genes in PBMC from six normal subjects. IFN-γ, IL-1β, TNF-α, and GAPDH were detected by reversed ethidium bromide staining, and the others were detected by Southern blotting. PBMC stimulated with PHA (IL-2, IL-2R, IFN-γ, IL-4, IL-5, IL-10), LPS (IL-1β, IL-8, TNF-α), or SAC(IL-12) were used as positive controls. Positive control cDNA samples were serially two-fold diluted to quantify expression levels. The fragment size of each PCR product is shown in the right column.

Expression of Th cytokines in DLI-induced GVHD patients

The levels of IL-2, IL-2Rα and IFN-γ gene expression were significantly increased in three DLI-treated patients compared with normal subjects (Figure 2). However, the pattern of cytokine gene expression differed in each patient.

Figure 2

Changes in the expression levels of Th1 cytokines (IL-2, IL-2R, IFN-γ) in three patients during GVHD induced by DLI. Patient 1, first column of panels; patient 2, second column; patient 3, third column. Relative expression levels were obtained from standard curves and then normalized to GAPDH expression. Samples were obtained indicated days after the first DLI. The original gels are shown at bottom.

In the first patient, the expression levels of IL-2, IL-2Rα, and IFN-γ were already high before DLI (Figure 2, first column). IL-2 and IL-2Rα expressions reached their maximum levels when the clinical symptoms of acute GVHD were most severe (grade III; skin 1, gut 2), and prednisolone (PSL) (60 mg/day) therapy decreased the expression and improved the clinical symptoms. IFN-γ levels were high throughout this observation period, but their relation to GVHD symptoms was unclear. Expression of IL-4 was clearly decreased as compared with that in the normal subjects throughout the clinical course, which contrasts with the high Th1 cytokine expression in this patient (Figure 3, first column). An increase in IL-10 expression was noted around the onset of GVHD.

Figure 3

Changes in the expression levels of Th2 cytokines (IL-4, IL-5, and IL-10) in three patients during GVHD induced by DLI. See Figure 2 for further details.

In the second patient, the expression of Th1 cytokines showed more dynamic changes (Figure 2, second column). IL-2 gene expression was within normal levels before DLI but sharply increased 2 weeks after the first DLI. This was accompanied by progression of acute GVHD grade III (skin 3, liver 3), which were confirmed by biopsies. IL-2 gene expression reached the maximum level on day 42. Expression of IL-2Rα changed similarly. IFN-γ expression was also clearly higher after DLI. After the start of methyl-PSL pulse therapy (1 g/day), the levels of expression of these cytokines were decreased (Figure 2, second column) and simultaneously the clinical condition of the patient was improved. As the dose of PSL was reduced, however, the levels of the cytokine gene expression started to increase again. The level of IL-4 expression decreased with the progression of GVHD. The change in IL-4 expression seemed to be inversely related to the change in IL-2 expression. The minimum expression of IL-4 occurred on day 42 when IL-2 expression was maximum (Figure 2 second column, and Figure 3 second column). The level of IL-5 gene expression in this patient sharply increased at the onset of GVHD, but was rapidly reduced after the pulse therapy. We also noted a marked increase in serum IL-5 level in this patient.13 The expression level of IL-10 did not significantly change during her clinical course.

In the third patient who showed a chronic type GVHD with sicca syndrome and liver damage, the kinetics of IL-2 gene expression were different from those in the other two patients (Figure 2, third column). An increased level of IL-2 was observed immediately after the first DLI. The level of IL-2 expression was decreased at the onset of the GVHD on day 46. The expression of IL-2R was detectable, but showed little change. IFN-γ expression gradually increased during the DLI and development of GVHD. Relatively high levels of IL-4 expression were observed. A significant increase in IL-10 expression and a decrease in IL-2 expression were noted on day 46 with the appearance of clinical symptoms of GVHD (Figure 2 third column, and Figure 3 third column).

Expression of monokines in DLI-induced GVHD patients

IL-1β and TNF-α levels were in the normal range or decreased during the clinical courses of the three patients (Figure 4). A transient increase in the IL-8 level was observed in all three patients. IL-12 gene expression tended to increase during the development of GVHD in the second and third patients.

Figure 4

Changes in the gene expression of four monokines (IL-1, IL-8, TNF-α, IL-12) during GVHD induced by DLI. See Figures 2 and 3 for further details.


The expression levels of Th1-type cytokines, IL-2 and IFN-γ greatly increased during the development of GVHD induced by DLI in the three patients examined in this study (Figure 2). These data are in accordance with earlier studies in murine models of GVHD.1415 The correlation of IL-2 gene levels with GVHD development was especially evident. High levels of soluble IL-2R have been demonstrated in patients with acute GVHD,16 and anti-IL-2R antibodies have been effective in treating GVHD.17181920 However, several studies of IL-2 in humans failed to show an alteration in IL-2 expression at either the protein level or the mRNA level2122 and only one study detected IL-2 expression more frequently in patients with acute GVHD than in those without GVHD.23 Thus, our report is the first to show a quantitative alteration in the levels of IL-2 gene expression during the development of clinical GVHD. Our finding of IL-2 mRNA in GVHD patients may be in part due to the higher sensitivity of the assay system: PCR combined with Southern blotting and detection. However, it is important to note that the use of an immunosuppressive drug such as cyclosporine or a steroid hormone, which reduces the expression of IL-2, was delayed until clinically significant GVHD was observed. Thus, these DLI-treated patients provided a valuable opportunity to observe the natural course of cytokine production during GVHD. Moreover, analysis of multiple time-points during the clinical courses showed when the changes in cytokine production occurred.

Although all three patients showed increased levels of IFN-γ expression, GVHD progression seemed to correlate less well with IFN-γ expression than with IL-2 expression (Figure 2). Several studies have reported increased IFN-γ expression in BMT patients,212223 but an association of IFN-γ expression with GVHD has not been well established. Moreover, conflicting results on its role in GVHD have been obtained in different experimental systems.242526

A study using a murine GVHD model has suggested that there are antagonistic effects between Th2 cells and Th1 cells in acute GVHD.27 Other studies using murine GVHD models have found that Th2 cytokines are predominant in chronic GVHD.1214 Our kinetic analysis revealed that the level of IL-4 decreased while that of IL-2 increased with the development of GVHD in patients 1 and 2 (Figures 2 and 3). Tanaka et al28 also noted that IL-4 expression was suppressed in patients with severe acute GVHD. This inverse correlation between IL-2 and IL-4 might reflect Th1 dominance during acute GVHD. Increased levels of IL-10 mRNA were found in two patients, as was previously found in human and murine acute GVHD.141522 In human GVHD, increased serum IL-10 levels were associated with severe GVHD.2930

Although increased levels of TNF-α and IL-1β expression have been implicated in the development of GVHD,821 such increases were not observed in PBMNC from the three patients in this study (Figure 4). The low TNF-α and IL-1β expressions in our patients may be partly the result of their not receiving any high-dose chemotherapy or irradiation, which are known to induce pro-inflammatory cytokines,3132 during the DLI treatment periods. The transient increase in IL-8 mRNA levels that we observed in each of the three patients (Figure 4) is interesting because a transient increase in serum IL-8 level has been correlated with veno-occlusive disease of the liver and acute GVHD.33 Further studies of this change are needed.

In summary, we have demonstrated the occurrence of cytokine dysregulation, especially the predominance of Th1 type cytokines, in the development of human GVHD induced by DLI. Although levels of IL-2 mRNA were correlated with GVHD progression, serum IL-2 levels were too low to be measured and measurement of IL-2 mRNA by our method is time-consuming and requires extensive labor. Thus, for the further assessment of cytokine profiles, both a quick and sensitive assay system such as a recently developed real-time quantitative PCR system34 and a larger number of clinical samples are needed.


  1. 1

    Storb R, Thomas ED . Graft-versus-host disease in dog and man: the Seattle experience Immunol Rev 1985 88: 215–238

  2. 2

    Klimpel GR, Annable CR, Cleveland MG et al. Immunosuppression and lymphoid hypoplasia associated with chronic graft-versus-host disease is dependent upon IFN-gamma production J Immunol 1990 144: 84–93

  3. 3

    Symington FW, Pepe MS, Chen AB, Deliganis A . Serum tumor necrosis factor alpha associated with acute graft-versus-host disease in humans Transplantation 1990 50: 518–521

  4. 4

    Holler E, Kolb HJ, Moller A et al. Increased serum levels of tumor necrosis factor alpha precede major complications of bone marrow transplantation Blood 1990 75: 1011–1016

  5. 5

    McCarthy PL Jr, Abhyankar S, Neben S et al. Inhibition of interleukin-1 by an interleukin-1 receptor antagonist prevents graft-versus-host disease Blood 1991 78: 1915–1918

  6. 6

    Ferrara JLM, Deeg HJ . Graft-versus-host disease New Engl J Med 1991 324: 667–674

  7. 7

    Antin JH, Ferrara JL . Cytokine dysregulation and acute graft-versus-host disease Blood 1992 80: 2964–2968

  8. 8

    Ferrara JL . The cytokine modulation of acute graft-versus-host disease Bone Marrow Transplant 1998 21: (Suppl. 3) S13–S15

  9. 9

    Mosmann TR, Cherwinski H, Bond MW et al. Two types of murine helper T-cell clone. I. Definition according to profiles of lymphokine activities and secreted proteins J Immunol 1986 136: 2348–2357

  10. 10

    Salgame P, Abrams JS, Clayberger C et al. Differing lymphokine profiles of functional subsets of human CD4 and CD8 T cell clones Science 1991 254: 279–282

  11. 11

    SAllen JE, Maizels RM . Th1-Th2: reliable paradigm or dangerous dogma? Immunol Today 1997 18: 387–392

  12. 12

    Garlisi CG, Pennline KJ, Smith SR et al. Cytokine gene expression in mice undergoing chronic graft-versus-host disease Mol Immunol 1993 30: 669–677

  13. 13

    Imoto S, Oomato Y, Murata K et al. Kinetics of serum cytokines after allogeneic bone marrow transplantation: interleukin-5 as a potential marker of acute graft-versus-host disease Int J Haematol 2000 72: 92–97

  14. 14

    Allen RD, Staley TA, Sidman CL . Differential cytokine expression in acute and chronic murine graft-versus-host disease Eur J Immunol 1993 23: 333–337

  15. 15

    Baker KS, Allen RD, Roths JB, Sidman CL . Kinetic and organ-specific patterns of cytokine expression in acute graft-versus-host disease Bone Marrow Transplant 1995 15: 595–603

  16. 16

    Siegert W, Josimovic-Alasevic O, Schwerdtfeger R et al. Soluble interleukin 2 receptors in patients after bone marrow transplantation Bone Marrow Transplant 1990 6: 97–101

  17. 17

    Herve P, Wijdenes J, Bergerat JP et al. Treatment of corticosteroid resistant acute graft-versus-host disease by in vivo administration of anti-interleukin-2 receptor monoclonal antibody (B-B10) Blood 1990 75: 1017–1023

  18. 18

    Anasetti C, Martin PJ, Hansen JA et al. A phase I–II study evaluating the murine anti-IL-2 receptor antibody 2A3 for treatment of acute graft-versus-host disease Transplantation 1990 50: 49–54

  19. 19

    Blaise D, Olive D, Hirn M et al. Prevention of acute GVHD by in vivo use of anti-interleukin-2 receptor monoclonal antibody (33B3.1): a feasibility trial in 15 patients Bone Marrow Transplant 1991 8: 105–111

  20. 20

    Anasetti C, Hansen JA, Waldmann TA et al. Treatment of acute graft-versus-host disease with humanized anti-Tac: an antibody that binds to the interleukin-2 receptor Blood 1994 84: 1320–1327

  21. 21

    Tanaka J, Imamura M, Kasai M et al. Cytokine gene expression in peripheral blood mononuclear cells during graft-versus-host disease after allogeneic bone marrow transplantation Br J Haematol 1993 85: 558–565

  22. 22

    Carayol G, Bourhis JH, Guillard M et al. Quantitative analysis of T helper 1, T helper 2, and inflammatory cytokine expression in patients after allogeneic bone marrow transplantation: relationship with the occurrence of acute graft-versus-host disease Transplantation 1997 63: 1307–1313

  23. 23

    Roy J, Blazar BR, Ochs L, Weisdorf DJ . The tissue expression of cytokines in human acute cutaneous graft-versus-host disease Transplantation 1995 60: 343–348

  24. 24

    Mowat AM . Antibodies to IFN-gamma prevent immunologically mediated intestinal damage in murine graft-versus-host reaction Immunology 1989 68: 18–23

  25. 25

    Brok HP, Heidt PJ, van der Meide PH et al. Interferon-gamma prevents graft-versus-host disease after allogeneic bone marrow transplantation in mice J Immunol 1993 151: 6451–6459

  26. 26

    Murphy WJ, Welniak LA, Taub DD et al. Differential effects of the absence of interferon-gamma and IL-4 in acute graft-versus-host disease after allogeneic bone marrow transplantation in mice J Clin Invest 1998 102: 1742–1748

  27. 27

    Fowler DH, Kurasawa K, Smith R et al. Donor CD4-enriched cells of Th2 cytokine phenotype regulate graft-versus-host disease without impairing allogeneic engraftment in sublethally irradiated mice Blood 1994 84: 3540–3549

  28. 28

    Tanaka J, Imamura M, Kasai M et al. The important balance between cytokines derived from type 1 and type 2 helper T cells in the control of graft-versus-host disease Bone Marrow Transplant 1997 19: 571–576

  29. 29

    Hempel L, Korholz D, Nussbaum P et al. High interleukin-10 serum levels are associated with fatal outcome in patients after bone marrow transplantation Bone Marrow Transplant 1997 20: 365–368

  30. 30

    Takatsuka H, Takemoto Y, Okamoto T et al. Predicting the severity of graft-versus-host disease from interleukin-10 levels after bone marrow transplantation Bone Marrow Transplant 1999 24: 1005–1007

  31. 31

    Xun CQ, Thompson JS, Jennings CD et al. 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: 2360–2367

  32. 32

    Remberger M, Ringden O, Markling L . TNF alpha levels are increased during bone marrow transplantation conditioning in patients who develop acute GVHD Bone Marrow Transplant 1995 15: 99–104

  33. 33

    Remberger M, Ringden O . Serum levels of cytokines after bone marrow transplantation: increased IL-8 levels during severe veno-occlusive disease of the liver Eur J Haematol 1997 59: 254–262

  34. 34

    Gibson UE, Heid CA, Williams PM . A novel method for real time quantitative RT-PCR Genome Res 1996 6: 995–1001

Download references


The authors are grateful to Ms I Ushio and Ms K Miki for their excellent technical assistance.

Author information

Correspondence to T Koizumi.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Das, H., Imoto, S., Murayama, T. et al. Kinetic analysis of cytokine gene expression in patients with GVHD after donor lymphocyte infusion. Bone Marrow Transplant 27, 373–380 (2001) doi:10.1038/sj.bmt.1702799

Download citation


  • cytokine
  • GVHD
  • BMT
  • donor lymphocyte infusion

Further reading