Graft-versus-host Disease

Novel mechanism of rapamycin in GVHD: increase in interstitial regulatory T cells

Article metrics


Rapamycin (RAPA) is an immunosuppressive drug that prevents and treats graft-versus-host disease (GVHD) after allogeneic hematopoietic cell transplant (HCT). One possible mechanism for its efficacy is induction of tolerance, through increased number or enhanced survival of regulatory T cells. In our experiments, B10.D2 BM and splenocytes were injected into lethally irradiated BALB/cJ recipients. The mice received i.p. injections of either RAPA or vehicle control on days 1–28. There was a significant survival advantage in RAPA-treated mice. Evaluation of the skin biopsies showed a dense cellular infiltrate in RAPA-treated mice. Further characterization of these cells revealed a higher percentage of regulatory T cells characterized by FoxP3-positive cells in high-dose RAPA-treated mice as compared with controls on day 30. This effect appears to be dose dependent. When peripheral blood analysis for FoxP3-positive cells was performed, there was no significant difference observed in the RAPA-treated mice as compared with control mice. These data show a novel mechanism of rapamycin in GVHD, accumulation of regulatory T cells in the GVHD target tissue: the skin.


Rapamycin (RAPA) is a macrolide antibiotic produced by streptomyces hygroscopicus and is also a potent immunosuppressant. RAPA is used extensively in solid organ transplant, and has an emerging role in GVHD.1, 2, 3, 4, 5, 6, 7 RAPA prevents GVHD in several murine models, although the mechanism is not clear. It is well established that RAPA suppresses proliferation of conventional CD4+ T cells (Tconv) through blocking the mammalian target of RAPA. Further studies show that T cells activated in the presence of RAPA may have a more tolerogenic phenotype.8, 9 There are emerging data suggesting that RAPA helps induce tolerance by conditioning DC to preferentially activate suppressive T-cell subsets.10 Finally, RAPA appears to cause an increase in regulatory T cells (Tregs), or at least provide selective survival advantage11, 12, 13, 14

In studies performed in our lab and others, RAPA prevented mortality from GVHD in murine models1, 4 On histopathology, it was observed that there was a dense infiltrate of inflammatory cells, which in later analysis indicated that a majority of these cells were CD4+4. When the splenocytes were analyzed at a later date, they were found to have suppressive properties in MLCs.

Tregs are a subset of T cells characterized by CD4+CD25hi cells that also express forkhead transcription factor FoxP3. They suppress Tconv and help promote tolerance. Several laboratories have shown that giving donor Tregs can suppress GVHD in murine models.15, 16, 17 Some studies show in humans with acute GVHD that there is a decrease in the number of Tregs in the peripheral blood and the affected tissue.18, 19, 20 It is unclear where in the immunologic reaction Tregs have a role, in the peripheral blood, lymph nodes or affected tissues.

Our aim in the current study was to further explore the nature of the cellular infiltrate in mice that clinically do not have significant GVHD. Both ear biopsies and peripheral blood were analyzed for Tregs. The data show a novel mechanism of rapamycin in GVHD, accumulation of regulatory T cells in the GVHD target tissue: the skin.

Materials and methods


B10.D2/nSnJ (H2d, Mls-2b, Mls-3b) and BALB/cJ (H2d, Mls-2a, Mls-3a) were purchased from the Jackson Laboratory (Bar Harbor, ME, USA). For the in vivo studies, only female mice were used. The mice were kept in a specific pathogen-free facility throughout the study. All animal protocols were approved by our IACUC.

GVHD Model

BM cells (10 × 106 B10/D2) and splenocytes (100 × 106 B10.D2) were injected in 0.5 ml plain RPMI-1640 through the tail vein into lethally irradiated (8.5 Gy) BALB/cJ recipients. Both recipient and donor mice were 12–14 weeks at the time of the experiments. The mice were monitored daily for wt and mortality. In experiment 1, mice were killed at days 14, 28 and 42 and their organs harvested for histology. In experiment 2, mice were monitored for 60 days and had ear biopsies and blood draws on days 14, 28 and 42. In addition, paraffinized ear biopsies from experiments performed previously in the laboratory4 were also used.

Drug and treatment

RAPA was purchased from LC Laboratories (Woburn, MA, USA) in pure powder form. It was prepared fresh daily in carboxymethylcellulose 0.2% and thoroughly homogenized before injection. The mice were given 3–5 mg/kg of RAPA or carboxymethylcellulose vehicle control daily through i.p. injection on days 1–28 after transplantation. The volume given was 0.01 ml/g. Doses of greater than 5 mg/kg resulted in increased toxicity and mortality.

Histology and immunohistochemistry

We obtained ear biopsies from the mice given the ease of procurement, and the findings are consistent with the skin biopsies (data not shown).

Tissue was obtained, placed in 10% formalin and paraffin embedded. Tissues from previous experiments were in paraffin and processed. Samples were then stained with hematoxylin–eosin (H&E) or immunohistochemistry (IHC) for intracellular FoxP3.

IHC was performed on 5-μm slices obtained from paraffin blocks. The MOM Immunodetection Kit from Vector Laboratories (Burlingame, CA, USA) (cat. no. PK-2200) was used. The methods were fully described by Nakmura et al.21 Briefly, the slides were placed in a steam bath for 40 min for Ag retrieval. Then, after several blocking steps, they were incubated overnight with 1:10 mouse anti-mouse/human/rat—FoxP3 IgG. The slides were washed and then incubated with anti-mouse IgG conjugated to biotin substrate. The slides were then stained with fluorescein avidin DCS, and subsequently developed.

The cells were counted using a Leica microscope. For the H&E stained slides to assess cellularity of the tissues, a total of 10 h.p.f. were counted or the maximum possible depending on the quality of the slide. For the IHC slides, 1000 nucleated cells were assessed, or as many as possible given the quality of staining.

Images of skin and liver sections were acquired with an AxioCam MRc digital camera mounted on an Axiovert 200 inverted microscope (Carl Zeiss Microimaging, Thornwood, NY, USA). A-Plan × 10/0.25 and LD-Plan NEOFLUAR × 20/0.4 objectives were used. Images were recorded using AxioVision Rel. 4.5 software (Carl Zeiss Microimaging).

FACS analysis

Cells were stained with a mouse Treg staining kit (Biolegend, San Diego, CA, USA cat no. 320018) as described by Bamias et al.22 Briefly, peripheral blood (50 μl) was incubated with anti-CD4 APC (clone RM4-5), anti-CD25 PE (clone PC61) at room temperature for 15 min. The cells were then rinsed with FACS buffer, followed by treatment with 1 × FACS lysing solution (Becton Dickinson, San Jose, CA, USA). Cells were then permeabilized with Biolegend fix/perm solution × 30 min. The cells were then incubated with anti-FoxP3 conjugated to alexa-fluor 488 (Clone 150D) or isotype control for 20 min at room temperature. The cells were washed with FACS buffer, and analyzed on an FACS Canto machine, using FACS Diva software.

Statistical analysis

Group comparisons were made with Student's t-test.


Rapamycin prevents GVHD

Irradiated BALB/c recipients of B10.D2 BM and splenocytes were treated with RAPA at a dose of 3–5 mg/kg for the first 4 weeks after transplantation. Consistent with previous experiments,4 mice treated with RAPA (n=7) did not develop GVHD and all RAPA-treated mice survived more than 90 days (Figures 1 and 2). By contrast, all mice treated with carboxymethylcellulose (n=10) developed severe GVHD and all of them died within 70 days after transplantation with a median survival period of 55 days (Figures 1 and 2). Similar to our published data,4 histological analysis showed that the mice treated with RAPA were free of acute and chronic GVHD in the skin, liver and intestine (data not shown). These mice were used for subsequent mechanistic studies.

Figure 1

Survival curve of mice treated with RAPA or vehicle control. BALB/cJ mice were lethally irradiated, and given BM and splenocytes from B10.D2 mice. The mice were monitored over a 3-month period for mortality.

Figure 2

Wt of mice in grams after transplant. After transplant, mice were weighed two times weekly.

Histologic changes

As seen in previous experiments,4 there was a significant cellular infiltrate noted in the ear skin of the RAPA-treated mice as compared with control mice. Analysis of H&E slides showed significantly more cells/h.p.f. on mice treated with RAPA as compared with control on days 28 and 42 (Figure 3). On days 28 and 42, control mice had significantly fewer cells per high-powered field, as compared with RAPA mice, 46 vs 89 (P=0.028) and 36 vs 68 (P=0.016), respectively. There was no significant difference on day 14. Evaluation of IHC staining performed on the ear biopsies showed an increase in the percentage of FoxP3-positive cells per infiltrating nucleated cell. This was shown over three separate experiments (Figure 4 and 5), and appears to be dose dependent. In mice treated with control vehicle and low-dose RAPA (1.5 mg/kg), there was a low percentage, however, in a high dose of RAPA (3–5 mg/kg) and the percentage increased to 14% of FoxP3-positive cells per total nucleated cell, which was statistically significant (P<0.01).

Figure 3

Infiltrating nucleated cells per high power field. H&E stained biopsies were evaluated for cellular infiltrate. Shown are infiltrating cells per high-powered field. 10 h.p.f. or the maximum possible was counted and the average calculated. In the control group, the following number of slides were analyzed: day 14, n=5; day 28, n=7; day 42, n=6. In the RAPA group, the following number of slides were analyzed: day 14, n=6; day 28, n=6; day 42, n=6.

Figure 4

Histopathology of ear biopsies and liver biopsies. (a) IHC for FoxP3 vehicle control × 10. (b) IHC for FoxP3 RAPA 1.5 mg/kg (c) IHC for FoxP3. RAPA 3–5 mg/kg × 10 (d) IHC for FoxP3 RAPA 3–5 mg/kg × 20. H&E liver biopsy: (e) Vehicle control-treated mouse × 10. (f) RAPA-treated mouse × 10. These are representative samples.

Figure 5

Percent FoxP3-positive cells per total infiltrating nucleated cell. Immunohistochemistry staining was performed on the ear biopsies from day 30 for FoxP3. A percentage of FoxP3-positive cells per total nucleated infiltrating cell were analyzed. One thousand cells, or the maximum possible cells, were counted to determine the percentage. The above graph represents the slides obtained in three separate experiments, in the 0 mg/kg group, n=10; in the 1.5 mg/kg group, n=4; in the 3–5 mg/kg group, n=11.

To determine whether Tregs were present in other target organs of GVHD, we further analyzed the liver and gut. In the liver, a cellular infiltrate was noted in control mice, in addition to destruction of the structures in the portal triad. There was minimal cellular infiltration noted in the liver of RAPA-treated mice, and no compromise of the structures was noted on day 30 (Figure 4). There was no evidence of Tregs in any of the liver specimens (data not shown). In the gut, there was more inflammation in the control mice as compared with the RAPA-treated mice; however, this was difficult to quantify and Tregs were present in both treated and untreated mice (data not shown).

Peripheral blood

To determine whether RAPA expanded Tregs systemically, FoxP3+ cells were quantified in peripheral blood. As shown in Figure 6, analysis of peripheral blood did not show any significant difference in FoxP3-positive cells on days 14, 28 or 42 in the RAPA-treated vs control mice, suggesting that RAPA does not affect Treg numbers systemically.

Figure 6

Percent FoxP3-positive cells per CD4+ cell in peripheral blood. Peripheral blood was obtained on days 14, 28 and 42 and analyzed by FACS analysis. The above represents FoxP3-positive cells as a percentage of total CD4+ cells. No significant difference was observed between the groups. In the control group, the following number of samples were analyzed: Day 14, n=5; day 28, n=7; day 42, n=6. In the RAPA group, the following number of samples were analyzed: day 14, n=6; day 28, n=6; day 42, n=6.


These experiments confirm that RAPA can prevent mortality from GVHD, as observed in previous experiments.1, 4 We also show in mice treated with GVHD that a densely cellular infiltrate that contains a high percentage of FoxP3-positive cells is present in the clinically unaffected tissues. These findings suggest that abrogation of clinical disease is mediated at least in part by Tregs. It is noted that an increase in Tregs is not appreciated in the peripheral blood at the time points measured.

The immunomodulatory effects of RAPA have been an area of active interest for a number of years. Initial studies by Blazar et al.1 10 years ago showed that RAPA improved survival in mice undergoing transplantation; however, the mechanism remains a subject of debate. One possible mechanism is the suppression of Tconv proliferation, mediated by RAPA binding to FK-binding protein and inhibition of the mammalian target of RAPA. This is mediated through an IL-2 independent pathway.

Another potential mechanism is through the induction of Tregs. Several investigators have shown the ability to expand Tregs by culturing them in the presence of RAPA.11, 23 Zeiser et al. further explored this phenomenon and showed that RAPA promoted Treg expansion and proliferation in vivo with luciferase imaging.24 Tregs appear to induce the STAT5 pathway in the presence of RAPA, which promotes proliferation.13 Furthermore, RAPA may exert immodulatory effects through DC. Turnquist et al.10 showed that RAPA-conditioned DC were poor stimulators of effector T cells, but enriched Tregs. This may be secondary to effects on maturation,25 Ag uptake26 or survival.27

It is intriguing that Tregs were present in the skin, but were not increased in the peripheral blood, liver or gut. Luciferase imaging data suggest that Tregs are primed in the lymph nodes and then home to organs to prevent inflammation,28 and this study confirmed the presence of FoxP3+ cells in one affected tissue, the skin. The fraction of Tregs in the peripheral blood in each group is equal, consistent with earlier studies13, 28 This phenomenon may reflect the fact that Tregs proliferate in the lymph nodes before entering the target tissues.28 In our experiments, the presence of Tregs in the ear biopsies was very clear. However, although it appears that there is no increased accumulation of Tregs in the gut or liver, this needs to be confirmed on further analysis as the sample size was small and could be subject to sampling error.

One question remains unanswered: how does RAPA prevent mortality in this model? Although we clearly show the presence of Tregs in the ear biopsies, skin GVHD, though well described in this model, is not the cause of mortality. Presumably, there are other mechanisms that result in improved survival. The most likely cause of mortality in this model is GVHD of the gut. The gut biopsies obtained were unrevealing, though notably there were Tregs in both the RAPA-treated and control mice. We also might have missed areas with increased Tregs because of sampling error. However, we know that even at baseline, Tregs are present in the intestinal tissue.29 In the initial experiments using RAPA to prevent GVHD in this mouse model, the splenocytes were rich in T cells that were suppressive in MLRs, which may in fact be Tregs.4 This would suggest that Tregs present in the lymphatic system may suppress Tconv function before their entrance into the affected tissues.

This study indicates that at least a part of the mechanism of GVHD suppression provided by RAPA is through the induction of Tregs, and that these cells act within the affected tissues, mainly the skin. Our study shows the presence of Tregs in tissues known to be affected by GVHD in untreated mice. The presence of a dense lymphocytic infiltrate without tissue injury indicates that, at least in the skin, their presence prevents the action of Tconv. However, on the basis of previous data showing that cells of a suppressive nature populate the spleen,4 there are indications that they may also exert their effect on effector cells in the lymphatic tissue. Further studies will be necessary to better understand this phenomenon and how this can be exploited in a clinical trial.

Conflict of interest

The authors of this paper have no conflict of interest to declare. These data were presented in abstract form at ASBMT, 17 February 2008.


  1. 1

    Blazar BR, Taylor PA, Panoskaltsis-Mortari A, Vallera DA . Rapamycin Inhibits the Generation of Graft-Versus-Host Disease- and Graft-Versus-Leukemia-Causing T Cells by Interfering with the Production of Th1 or Th1 Cytotoxic Cytokines. J Immunol 1998; 160: 5355–5365.

  2. 2

    Couriel DR, Saliba R, Escalon MP, Hsu Y, Ghosh S, Ippoliti C et al. Sirolimus in combination with tacrolimus and corticosteroids for the treatment of resistant chronic graft-versus-host disease. Br J Haematol 2005; 130: 409–417.

  3. 3

    Antin JH, Cutler C . Sirolimus for GVHD prophylaxis in allogeneic stem cell transplantation. Bone Marrow Transplantation 2004; 34: 471–476.

  4. 4

    Chen BJ, Morris R, Chao NJ . Graft-Versus-Host Disease Prevention by Rapamycin: Cellular Mechanisms. Biol Blood Marrow Transplant 2000; 6: 529–536.

  5. 5

    Noris M, Casiraghi F, Todeschini M, Cravedi P, Cugini D, Monteferrante G et al. Regulatory T cells and T cell depletion: role of immunosuppressive drugs. J Am Soc Nephrol 2007; 18: 1007–1018.

  6. 6

    Antin JH, Kim HT, Cutler C, Ho VT, Lee SJ, Miklos DB et al. Sirolimus, tacrolimus, and low-dose methotrexate for graft-versus-host disease prophylaxis in mismatched related donor or unrelated donor transplantation. Blood 2003; 102: 1601–1605.

  7. 7

    Cutler C, Li S, Ho VT, Koreth J, Alyea E, Soiffer RJ et al. Extended follow-up of methotrexate-free immunosuppression using sirolimus and tacrolimus in related and unrelated donor peripheral blood stem cell transplantation. Blood 2007; 109: 3108–3114.

  8. 8

    Valmori D, Tosello V, Souleimanian NE, Godefroy E, Scotto L, Wang Y et al. Rapamycin-mediated enrichment of T cells with regulatory activity in stimulated CD4+ T cell cultures is not due to the selective expansion of naturally occurring regulatory T cells but to the induction of regulatory functions in conventional CD4+ T cells. J Immunol 2006; 177: 944–949.

  9. 9

    Powell JD, Lerner CG, Schwartz RH . Inhibition of cell cycle progression by rapamycin induces T cell clonal anergy even in the presence of costimulation. J Immunol 1999; 162: 2775–2784.

  10. 10

    Turnquist HR, Raimondi G, Zahorchak AF, Fischer RT, Wang Z, Thomson AW . Rapamycin-conditioned dendritic cells are poor stimulators of allogeneic CD4+ T cells, but enrich for antigen-specific foxp3+ T regulatory cells and promote organ transplant tolerance. J Immunol 2007; 178: 7018–7031.

  11. 11

    Battaglia M, Stabilini A, Roncarolo M-G . Rapamycin selectively expands CD4+CD25+FoxP3+ regulatory T cells. Blood 2005; 105: 4743–4748.

  12. 12

    Battaglia M, Stabilini A, Migliavacca B, Horejs-Hoeck J, Kaupper T, Roncarolo M-G . Rapamycin promotes expansion of functional CD4+CD25+FOXP3+ regulatory T cells of Both Healthy Subjects and Type 1 Diabetic Patients. J Immunol 2006; 177: 8338–8347.

  13. 13

    Zeiser R, Leveson-Gower DB, Zambricki EA, Kambham N, Beilhack A, Loh J et al. Differential impact of mammalian target of rapamycin inhibition on CD4+CD25+Foxp3+ regulatory T cells compared with conventional CD4+ T cells. Blood 2008; 111: 453–462.

  14. 14

    Strauss L, Whiteside TL, Knights A, Bergmann C, Knuth A, Zippelius A . Selective survival of naturally occurring human CD4+CD25+Foxp3+ regulatory T cells cultured with rapamycin. J Immunol 2007; 178: 320–329.

  15. 15

    Ermann J, Hoffmann P, Edinger M, Dutt S, Blankenberg FG, Higgins JP et al. Only the CD62 L+ subpopulation of CD4+CD25+ regulatory T cells protects from lethal acute GVHD. Blood 2005; 105: 2220–2226.

  16. 16

    Wysocki CA, Jiang Q, Panoskaltsis-Mortari A, Taylor PA, McKinnon KP, Su L et al. Critical role for CCR5 in the function of donor CD4+CD25+ regulatory T cells during acute graft-versus-host disease. Blood 2005; 107: 3300–3307.

  17. 17

    Hoffmann P, Ermann J, Edinger M, Fathman CG, Strober S . Donor-type CD4+CD25+ Regulatory T Cells Suppress Lethal Acute Graft-Versus-Host Disease after Allogeneic Bone Marrow Transplantation. J Exp Med 2002; 196: 389–399.

  18. 18

    Rieger K, Loddenkemper C, Maul J, Fietz T, Wolff D, Terpe H et al. Mucosal FOXP3+ regulatory T cells are numerically deficient in acute and chronic GvHD. Blood 2006; 107: 1717–1723.

  19. 19

    Nguyen VH, Zeiser R, Negrin RS . Role of naturally arising regulatory T cells in hematopoietic cell transplantation. Biol Blood Marrow Transplant 2006; 12: 995–1009.

  20. 20

    Schneider M, Munder M, Karakhanova S, Ho AD, Goerner M . The initial phase of graft-versus-host disease is associated with a decrease of CD4+CD25+ regulatory T cells in the peripheral blood of patients after allogeneic stem cell transplantation. Clin Lab Hematol 2006; 28: 382–390.

  21. 21

    Nakamura T, Shima T, Saeki A, Hidaka T, Nakashima A, Takikawa O et al. Expression of indoleamine 2, 3-dioxygenase and the recruitment of Foxp3-expressing regulatory T cells in the development and progression of uterine cervical cancer. Cancer Sci 2007; 98: 874–881.

  22. 22

    Bamias G, Okazawa A, Rivera-Nieves J, Arseneau KO, De La Rue SA, Pizarro TT et al. Commensal bacteria exacerbate intestinal inflammation but are not essential for the development of murine ileitis. J Immunol 2007; 178: 1809–1818.

  23. 23

    Keever-Taylor CA, Browning M, Johnson B, Truitt R, Bredeson C, Behn B et al. Rapamycin enriches for CD4(+) CD25(+) CD27(+) Foxp3(+) regulatory T cells in ex vivo-expanded CD25-enriched products from healthy donors and patients with multiple sclerosis. Cytotherapy 2007; 9: 144–157.

  24. 24

    Zeiser R, Nguyen VH, Beilhack A, Buess M, Schulz S, Baker J et al. Inhibition of CD4+CD25+ regulatory T-cell function by calcineurin-dependent interleukin-2 production. Blood 2006; 108: 390–399.

  25. 25

    Hackstein H, Taner T, Zahorchak AF, Morelli AE, Logar AJ, Gessner A et al. Rapamycin inhibits IL-4—induced dendritic cell maturation in vitro and dendritic cell mobilization and function in vivo. Blood 2003; 101: 4457–4463.

  26. 26

    Hackstein H, Taner T, Logar AJ, Thomson AW . Rapamycin inhibits macropinocytosis and mannose receptor-mediated endocytosis by bone marrow-derived dendritic cells. Blood 2002; 100: 1084–1087.

  27. 27

    Woltman AM, van der Kooij SW, Coffer PJ, Offringa R, Daha MR, van Kooten C . Rapamycin specifically interferes with GM-CSF signaling in human dendritic cells, leading to apoptosis via increased p27KIP1 expression. Blood 2003; 101: 1439–1445.

  28. 28

    Nguyen VH, Zeiser R, daSilva DL, Chang DS, Beilhack A, Contag CH et al. In vivo dynamics of regulatory T-cell trafficking and survival predict effective strategies to control graft-versus-host disease following allogeneic transplantation. Blood 2007; 109: 2649–2656.

  29. 29

    Kang SG, Lim HW, Andrisani OM, Broxmeyer HE, Kim CH . Vitamin a metabolites induce gut-homing FoxP3+ regulatory T cells. J Immunol 2007; 179: 3724–3733.

Download references


Many thanks to Zouwei Su, PhD, for his help in the preparation of the histopathology.

Author information

Correspondence to J M Palmer.

Rights and permissions

Reprints and Permissions

About this article


  • regulatory T cells
  • rapamycin
  • graft-versus-host disease

Further reading