Published online before print February 12, 2008
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Center for Immunology and Department of Medicine and
Department of Surgery, University of Minnesota, Minneapolis, Minnesota, USA
1Correspondence: University of Minnesota, Department of Medicine, 6-134 Niels Hasselmo Hall, 312 Church St. S.E., Minneapolis, MN 55455, USA. E-mail: khoru001{at}umn.edu
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: tolerance/suppression/anergy • cell differentiation
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
βTCR interactions with self-antigens [1
2
3
4
5
6
], we and others have shown that extrathymic de novo generation of Tregs is also possible following stimulation with foreign peptides [7
8
9
]. The fraction of the peripheral Treg pool in the normal animal generated outside the thymus is unknown. However, the ability to produce Tregs with desired antigen specificity is an important goal in the development of therapeutic protocols for treatment of allergic and autoimmune diseases. Previously, using the DO11.10 TCR transgenic (Tg) adoptive transfer system, we have shown that low doses of systemic antigen can drive differentiation of naïve CD4 T cells into CD4+CD25+ T cells with suppressive properties [7 ], and later, it was shown that induction of this phenotype also correlates with increased Foxp3 expression [10 ]. Interestingly, we and others [7 8 9 ] have also noted that emergence of Tregs correlates with low cell-cycle activity and is favored by low antigen-dose exposure. In contrast, increased cell-cycle activity associated with high antigen dose, autocrine IL-2 production, and activation of dendritic cells (DC) by LPS or anti-CD40 antibodies interferes with Treg differentiation [7 8 9 ]. A number of immunosuppressive strategies are already used in the clinics that limit T cell activation and proliferation. It is possible that some of these treatment regiments also encourage de novo Treg induction. Obviously, the ideal therapeutic approach to chronic inflammatory diseases would block activity of pathogenic T cells and induce Tregs specific for the same antigens. Costimulatory signals, such as B7, enhance proliferative responses of T cells. In fact, blockade of B7 costimulation is already being introduced in the clinics [11 , 12 ]. However, a negative impact on the Treg compartment is a potential drawback of this therapeutic approach. B7 signals have been shown to play important roles in the thymic development and peripheral maintenance of Tregs [13 14 15 16 17 ]. However, the role of B7:CD28 costimulation on de novo generation of Tregs in the periphery is not entirely clear. In fact, absence of CD28 benefited induction of Foxp3 mRNA expression following tolerizing systemic antigen administration, and greater degrees of B7 expression inhibited Foxp3 mRNA levels [10 ]. Here, we extend these findings by following the fate of antigen-specific Tregs induced in the absence of CD28 signaling. We found that although CD28 costimulation is not required to initiate the Treg differentiation, absence of CD28 costimulation prohibits their accumulation.
One of the targets of B7:CD28 costimulation is the mammalian target of rapamycin (mTOR) signaling pathway, which plays a critical role in the control of cap-dependent mRNA translation, cell growth, and proliferation [18 ]. Rapamycin (sirolimus) inhibits protein kinase activity of the mTOR/raptor complex 1 and is currently used as an immunosuppressive drug to prevent allograft rejection as well as an antineoplastic agent, because of its antiproliferative property. Interestingly, Tregs appear to be relatively resistant to effects of rapamycin, and the drug may promote generation of Tregs in vitro or allow their selective outgrowth in culture [19 20 21 22 ]. Here, we tested the effects of rapamycin on de novo generation of antigen-specific Tregs in vivo. As expected, we found the drug to have potent, antiproliferative effects on antigen-stimulated CD4 T cells. Furthermore, we observed that rapamycin promotes conversion of naïve CD4 T cells into Tregs and allows their persistence. Importantly, induction of Tregs following antigen stimulation in the presence of mTOR blockade by rapamycin is independent of specific antigen dose, which is a practical issue in most clinical situations.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Adoptive transfer and cell preparations
Donor DO11.10 RAG–/– T cells were collected from secondary lymphoid tissues (axillary, brachial, cervical, mesenteric, and inguinal lymph nodes and spleen) and adoptively transferred into normal, unirradiated BALB/c mice, as described previously, 2.5–5.0 x 106 cells per recipient animal [7
]. In adoptive transfers using non-RAG–/– donors, such as DO11.10 CD28–/– mice, Tregs were depleted using magnetic microbeads (Miltenyi Biotec, Auburn, CA, USA) and antibodies against CD25 and glucocorticoid-induced TNFR family-related gene (GITR). In most experiments, the donor cells were labeled with CFSE (Molecular Probes, Eugene, OR, USA) before transfer using a modification of the technique described previously [26
]. Briefly, lymph node and spleen cells were suspended at a concentration of 1 x 107 cells/ml in HBSS containing 0.5 µM CFSE and incubated for 10 min at 37°C. The reaction was stopped with the addition of a 1:1 mixture of Eagle’s Hanks’ amino acids medium containing 10% FCS, 2 mM L-glutamine, 100 U/ml penicillin, 100 U/ml streptomycin, and 5 x 10–5 M 2-ME.
Antigen treatment
Chicken OVA peptide 323–339 (OVAp) was synthesized by Invitrogen (Carlsbad, CA, USA). Varying doses of the peptide (5–100 µg) were administered i.v. as a single bolus. The 5-µg dose was established previously as the minimum dose necessary to ensure that most responder DO11.10 T cells encounter the antigen and the optimal dose for induction of constitutive expression of CD25 [7
].
Skin transplantation
Wild-type BALB/c recipient mice adoptively transferred with DO11.10 RAG–/– Thy1.1 T cells were administered 100 µg OVAp 8 days prior to skin transplantation. During this period of time, half of the recipients received daily i.p. injection of rapamycin (1 µg/g body weight) suspended in 0.2% carboxymethylcellulose as described previously [27
, 28
], and the other half received daily injection of the vehicle alone. On the day of transplantation, the recipient mice were anesthetized by i.p. injection of sodium pentobarbitone (0.1 mg/g body weight). Two 25-mm2 incisions were made on the back of the recipients. One of these sites was patched with 25 mm2-size skin grafts taken from tails of mOVA mice, and the contralateral side was patched with wild-type BALB/c control tail skin. The recipient mice were bandaged for 10 days following completion of the skin transplants. In addition, in a separate group of experiments, the bandages were removed from the recipients 7 days after surgery, and the skin grafts were peeled from the recipients for extraction of T cells, which was accomplished using 0.1 M EDTA.
Flow cytometry
Lymph nodes (generally axial, brachial, and inguinal, unless specified otherwise) were harvested and mashed in Petri dishes using 3 ml syringes. The suspensions were washed with staining buffer containing 2% FCS and 0.2% sodium azide and stained with combinations of antibodies for 15 min in the cold. All antibodies were purchased from eBioscience (San Diego, CA, USA). Cytokine staining was performed on cells fixed for 20 min in 2% formaldehyde as described previously [7
, 29
]. Staining for Foxp3 was done using the eBioscience kit according to the manufacturer’s instructions.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
As expected, we found few DO11.10 RAG–/– T cells expressing Foxp3 in the donor mice (Fig. 1 ). The DO11.10 T cells were adoptively transferred into normal BALB/c mice, which were then injected i.v. with 5 µg chicken OVAp 323–339. As we had originally demonstrated [7 ], virtually all Tg responders up-regulated CD25 expression within 12 h of challenge with this antigen dose (Fig. 1A) . However, the early expression of CD25 was transient and not accompanied by expression of Foxp3. On Day 3, a small, distinct population of Foxp3+ DO11.10 T cells emerged (Fig. 1A) . Over the next week, the population of Foxp3+ DO11.10 T cells increased in relative (Fig. 1B) and absolute cell numbers (Fig. 2B , left bar graph). Virtually all Foxp3+ cells expressed high levels of CD25, and few Foxp3– cells expressed CD25 (Fig. 1C) . Interestingly, the level of Foxp3 expression within individual Foxp3+ cells increased over the 1st week (Fig. 2B , right bar graph). As expected, conversion to Foxp3+ status correlated inversely with cell-cycle activity (Fig. 2A) . In fact, the greatest fraction of Foxp3+ DO11.10 T cells was seen among cells that did not undergo any cell division at all. Thus, the accumulation of Foxp3+ DO11.10 T cells cannot possibly be explained by expansion of the exceedingly small number of Foxp3+ cells within the pool of donor cells.
 View larger version (18K): [in a new window] |
Figure 1. Emergence of Foxp3+ antigen-specific CD4 T cells in the lymphoid periphery following administration of low-dose i.v. antigen. (A) The histograms show CD25 and Foxp3 expression by adoptively transferred DO11.10 T cells at indicated times after administration of 5 µg OVAp. (B) The graphs show the fraction of DO11.10 T cells expressing Foxp3 over the first 8 days (left graph) and 3 weeks (right graph) following exposure to the antigen. Each point on the graph represents the mean of four animals ± SD. The experiment is representative of three individual experiments. (C) The histograms show expression of CD25 by naïve (shaded), Foxp3– (dashed line), and Foxp3+ (bold line) DO11.10 T cells on Day 8 following injection with 5 µg OVAp.
|
 View larger version (31K): [in a new window] |
Figure 2. Peripheral acquisition of Foxp3 expression correlates inversely with cell-cycle history. (A) CFSE profiles and Foxp3 expression by DO11.10 T cells on Days 3 and 8 following antigen administration are shown in the representative histograms and density plots on the left. The individual cell-cycle bins are indicated within the density plots. The graph on the right side shows the relationship between cell-cycle history and fraction of DO11.10 T cells expressing Foxp3. ( ) DO11.10 T cells 3 days after antigen administration (5 µg OVAp); (•) DO11.10 T cells 8 days after antigen. At least 2000 DO11.10 events were collected from each individual animal. (B) The left bar graph shows absolute numbers of DO11.10 T cells within the axial, brachial, and inguinal lymph nodes on Days 3 and 8 after antigen administration. The middle bar graph shows absolute numbers of Foxp3+ DO11.10 T cells. The right bar graph shows the flow cytometric measurement of relative Foxp3 expression in terms of mean fluorescence intensity (MFI) of Foxp3+ cells. The bar graphs are derived from the experiment shown in A. There were five animals per group in this experiment, which is representative of two individual experiments. Comparisons between groups were done using the two-tailed Student’s t-test.
|
B7 costimulatory signals are not required for induction of Foxp3+ T cells but are necessary for their survival
CD28 signaling is well established to be an important signal for induction of Foxp3 expression in the developing thymocytes. However, as lack of CD28 costimulation limits antigen activation and cell-cycle activity, we hypothesized that it may favor peripheral induction of Foxp3 expression. First, we compared the general experience of CD28-sufficient and CD28–/– DO11.10 T cells following administration of various doses of the OVAp. The CD28-sufficient and CD28–/– DO11.10 T cells were cotransferred in these experiments and distinguished by differential expression of the Thy1.1 and Thy1.2 markers, respectively. The survey included CD69 expression at 1 h (Fig. 3A
, top panel), CD25 expression at 12 h (Fig. 3A
, middle panel), blastogenesis at 18 h (Fig. 3A
, bottom panel), and cell-cycle activity at 72 h (Fig. 3B)
. The results showed that CD28–/– DO11.10 T cells require approximately a fivefold higher antigen dose to experience a comparable degree of activation in this system. Similar results were seen when CD28-sufficient and CD28–/– DO11.10 responders were transferred separately into different animals (data not shown). The rather modest effect of removing CD28 costimulation may appear surprising compared with other studies [34
, 35
]. However, it is noteworthy that there is no adjuvant in our system; therefore, the response is measured under conditions of relatively low B7 expression.
 View larger version (30K): [in a new window] |
Figure 3. Early activation and commitment to cell division by antigen-specific CD4 T cells are modestly decreased in the absence of CD28. Thy1.2 CD28-deficient and Thy1.1 CD28-sufficient DO11.10 T cells were cotransferred into BALB/c recipients, which were injected with varying doses of the OVAp i.v. Mice were killed at indicated times after antigen injection. (A) Up-regulation of CD69 and CD25 as well as size of the DO11.10 T cells [forward-scatter (FSC)] was measured at indicated times by flow cytometry. Empty bars represent CD28-deficient responders, and filled bars represent CD28-sufficient responders. Each bar shows the mean of three animals. The experiment is representative of two independent experiments. (B) The upper panel shows CFSE content of CD28-deficient and CD28-sufficient DO11.10 T cells on Day 3 following injection of various doses of the OVAp. The lower panel shows the mean number of cell divisions calculated according to the following formula: M =  ·(Nnxn)/T, where M is the mean number of cell divisions, N is the number of cells within a given cell division bin, n is the number of cell divisions, and T is the total number of responder DO11.10 T cells. Each bar represents the mean of three animals, and the experiment is representative of two independent experiments.
|
 View larger version (29K): [in a new window] |
Figure 4. The absence of CD28 costimulation prevents accumulation of induced Foxp3+ antigen-specific CD4 T cells. CFSE-labeled, CD28-deficient DO11.10 T cells were adoptively transferred into BALB/c recipients, which were injected with various doses of the OVAp i.v. (A) Foxp3 expression by CD28–/– DO11.10 T cells is shown in the density plots. The percentage represents the fraction of all DO11.10 responders. The graphs next to the density plots show the summary of the data indicating Foxp3 expression by CD28-deficient DO11.10 T cells as a function of their cell division history. Each point in the graphs represents the mean of three animals. The experiment is representative of two independent experiments. (B) This bar graph summarizes Foxp3 expression by CD28-deficient DO11.10 responders on Day 3 after antigen stimulation as a function of antigen dose. Each value represents the mean of three animals. This experiment is representative of two independent experiments. (C) These bar graphs show the absolute number of all (left) and Foxp3+ (right) CD28-deficient Foxp3+ DO11.10 T cells on Days 3 and 8 following i.v. injection of 5 µg OVAp within axillary, brachial, and inguinal lymph nodes. Each bar represents the mean of five animals. The experiment is representative of two independent experiments.
|
BALB/c mice, adoptively transferred with CFSE-labeled, naïve DO11.10 RAG–/– T cells, were injected with 5 or 100 µg OVAp i.v. and received rapamycin or vehicle control by i.p. injection daily for 1 week. Expression of Foxp3 in DO11.10 T cells was measured on Day 8 after antigen administration. As we had expected, DO11.10 T cells proliferated considerably less in rapamycin-treated mice compared with animals that received the vehicle-control injections. Furthermore, we saw increased accumulation of Foxp3+ DO11.10 T cells in all rapamycin-treated animals, and their absolute numbers were comparable regardless of the antigen dose (Fig. 5 ). To probe the question of optimal timing of mTOR blockade for induction of Foxp3+ T cells in relation to antigen encounter, we administered rapamycin for the first 3, 5, and 7 days after injection of 100 µg OVAp. Longer administration of the drug correlated with better inhibition of cell divisions and greater induction of Foxp3+ DO11.10 T cells (Fig. 6 ). Interestingly, the percentage of Foxp3+ DO11.10 T cells increased within individual cell division bins, suggesting that inhibition of cell-cycle activity alone may be insufficient to explain the beneficial effects of rapamycin on Treg induction. Of course, interpretation of these experiments is limited by poor knowledge of pharmacokinetics of rapamycin in vivo and the likely lingering presence of the drug following initial i.p. injections.
 View larger version (39K): [in a new window] |
Figure 5. Treatment with rapamycin (RAPA) favors de novo induction of Foxp3+ antigen-specific CD4 T cells, irrespective of the stimulating antigen dose. CFSE-labeled DO11.10 T cells were adoptively transferred into BALB/c recipients, which were then injected with 5 or 100 µg OVAp i.v. The animals were then injected daily with rapamycin or only the vehicle control i.p. Mice were killed 8 days after antigen injection. The density plots on the left side of the figure show Foxp3 expression and CFSE content of the DO11.10 responders. The graphs on the right side show the percentage (upper) and absolute numbers (lower) of Foxp3+ DO11.10 T cells in the indicated experimental groups of mice. Each bar represents the mean of three animals. The experiment is representative of three independent experiments.
|
 View larger version (34K): [in a new window] |
Figure 6. Continuous administration of rapamycin favors conversion of antigen-specific responders to Foxp3+ phenotype. CFSE-labeled DO11.10 T cells were adoptively transferred into BALB/c mice, which were then injected with 100 µg OVAp. The mice were then injected daily with rapamycin i.p. for the first 3, 5, or 7 days before being killed on Day 8. (A) Density plots show Foxp3 expression and CFSE content of DO11.10 T cells within different experimental groups. (B) The absolute number of induced Foxp3+ DO11.10 T cells within the axial, brachial, and inguinal lymph nodes is shown as a function of length of rapamycin treatment. (C) The fraction of Foxp3+ DO11.10 T cells was determined within an individual cell division bin. Each bar in the figure represents the mean of three animals. The experiment is representative of two independent experiments.
|
Induction of tolerance toward the OVA 323–339 epitope by high-dose antigen alone (100 µg OVAp) was not sufficient to offer any graft protection (Fig. 7A ). This antigen dose leads to induction of "anergy" or nonresponsiveness in DO11.10 T cells but little induction of the Treg phenotype associated with expression of CD25 or Foxp3 [7 ]. In this model, rejection presence is not enhanced by DO11.10 T cell presence at all (data not shown). Therefore, mere induction of anergy would not be expected to offer any graft protection. However, the same antigen dose accompanied by rapamycin treatment does lead to induction of Foxp3+ DO11.10 T cells (Fig. 6) , and this is the only group of mice where we did observe significant, although incomplete, graft protection (Fig. 7A) . Importantly, rapamycin treatment alone did not offer any graft protection (Fig. 7A , right panel). Therefore, these data strongly suggest that Foxp3+ DO11.10 T cells are directly responsible for inhibition of the rejection response. In fact, we did observe graft infiltration by these cells in rapamycin-treated mice (Fig. 7B) .
 View larger version (24K): [in a new window] |
Figure 7. Induction of single, epitope-specific Foxp3+ T cells with assistance of rapamycin dampens rejection of skin grafts expressing a known minor histocompatibility antigen. DO11.10 T cells were adoptively transferred into BALB/c recipients and received indicated treatments for 8 days prior to skin grafting; all injections were stopped on Day 8. Tail skin from mOVA mice and BALB/c mice was grafted on opposite sides on the back. The skin bandages were removed on Day 10 after transplantation, and grafts were visually monitored after that. (A) The Kaplan-Meier plots show survival of the skin grafts. ( , Left graph) mOVA grafts in animals injected with 100 µg OVAp and treated with rapamycin. ( ) mOVA graft animals injected with 100 µg OVAp but treated with the vehicle control. The graph on the right shows survival of mOVA grafts in animals, which were not injected with the OVAp and were treated with rapamycin () or vehicle control (•). (B) Contour plots show Foxp3 expression in DO11.10 T cells from skin grafts (left) or axillary, brachial, and inguinal lymph nodes (right) taken from mice 14 days after grafting. Skin grafts were pooled within individual experiments. The plots are representative of three independent experiments. SSC, Side-scatter.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our results fit well with the recent literature exploring effects of rapamycin on Treg homeostasis and induction. Thus, the overall size of the Treg compartment is stable in mice or human patients treated with rapamycin, which contrasts with negative effects of cyclosporine [38 , 39 ]. Rapamycin has been shown to allow thymic generation of Tregs as well as their homeostatic proliferation in the periphery [40 ]. Furthermore, Tregs have been found to be relatively resistant to antiproliferative effects of rapamycin in comparison with naïve CD4 T cells, and the drug may be a useful tool for expansion of Tregs in vitro [19 20 21 22 ]. Recently, Gao et al. [41 ] found that rapamycin acts synergistically with TGF-β1 in promoting induction of murine Tregs in vitro. Furthermore, these authors have shown that in vivo treatment of mice with rapamycin promotes de novo generation of alloantigen-specific Foxp3+ T cells capable of mediating skin-graft protection. Our results confirm and extend these findings. By tracking a monoclonal population of antigen-specific CD4 T cells, we were able to address the issue of antigen dose during rapamycin-assisted Treg induction. We anticipate that this syngeneic, adoptive transfer model can be used in future mechanistic studies, as it allows tracking the responder T cells throughout the immune response.
Rapamycin, through FK506-binding protein, inhibits mTOR, a serine-threonine kinase, which forms a complex along with raptor, survivin, aurora B, p70S6k, and 4E-BP1. This complex controls cell-cycle progression at the G1-S transition point. One major effect of rapamycin is inhibition of 4E-BP1 phosphorylation. This results in blockade of 5' cap-dependent mRNA translation initiation, which is likely critical for production of many proteins involved in cell cycle [42
, 43
]. However, translation of some mRNAs may be initiated at the internal ribosome entry sites (IRES). It is tempting to speculate that mRNAs encoding proteins involved in the Treg induction, and perhaps Foxp3 itself, contain these IRES sequences, which would explain the beneficial effects of rapamycin seen here. Such a mechanism would be analogous to up-regulation of anergy factors in CD8 T cells stimulated by antigen in the presence of high IDO activity [44
]. In this case, there is activation of eukaryotic initiation factor 2- (eIF2) kinase, general control nonderepressible-2. Activation of eIF2 kinases abrogates assembly of the initiator tRNA-eIF2 ternary complex and leads to general inhibition of mRNA translation. However, this is accompanied by concurrent activation of the integrated stress-response pathway, which may actually increase expression of specific target genes [44
]. In fact, high IDO activity has been shown to promote induction of Foxp3+ T cells in vitro [45
, 46
]. It is also reasonable to consider that other eIF2 kinases, such as RNA-dependent protein kinase, which is associated with viral infections, may also promote Treg induction. In fact, the Treg subpopulation of CD4 T cells appears to be relatively spared during SIV and HIV infections [47
48
49
].
It is interesting to reflect that the two known activators of the mTOR pathway that accompany antigen stimulation are CD28 and IL-2 [18 , 35 , 50 ]. These same molecules also play critical roles in Treg generation in the thymus and turnover in the periphery [13 ]. In addition, expression of IL-2 and IL-2R benefits from CD28 costimulation. Our results suggest existence of some nuances for induced Tregs. First, CD28 signaling is not required for initial induction of Foxp3 expression in the periphery. In fact, absence of CD28 costimulation appears to be beneficial for Treg differentiation in the first days following antigen encounter. However, these newly generated Tregs fail to accumulate after Day 3 following antigen encounter and decrease in numbers. These results are consistent with findings of Lyddane et al. [10 ], who noted that basal CD28 costimulation was required for Akt phosphorylation and up-regulation of the antiapoptotic factor Bcl-xL. They have also found that absence of CD28 costimulation favored high initial levels of Foxp3 mRNA expression, and progressive increase in CD28 costimulation resulted in progressive loss in Foxp3 mRNA levels. Therefore, it was argued that augmented levels of CD28 costimulation associated with adjuvants such as LPS abrogate Treg generation [7 , 10 ]. Interestingly, recently, Zhong et al. [51 ] showed that treatment with antibody against CD86, but not CD80, may favor Treg induction. These results suggest that it may be yet possible to target the CD80/CD86:CD28 interaction therapeutically in a way that would promote Treg development. Alternatively, CD80/CD86-specific antibodies may trigger signaling in the APCs that can result in production of factors, such as IDO, which could benefit Treg induction [52 53 54 ]. Thus, effects of CTLA-4-Ig treatment on the Treg compartment are likely to be more complex than those of mere abrogation of CD28 costimulation.
Engagement of CD28 and IL-2R leads to activation of the PI-3K/Akt pathway, which is connected to mTOR through the tuberous sclerosis proteins TSC1 and TSC2. Removal of CD28 costimulation is expected to decrease mTOR activity, and the early rise in Treg induction in the absence of CD28 is consistent with results obtained using rapamycin to block mTOR. However, antiproliferative effects of rapamycin were more potent, which indicates that mTOR receives stimulation that is independent of CD28 and IL-2. In fact, proliferation of CD28–/– DO11.10 T cells was virtually, completely inhibited by rapamycin (data not shown). Conversely, it may be expected that some antiapoptotic signals resulting from CD28 costimulation and IL-2 are independent of mTOR, as rapamycin treatment didn’t compromise survival of induced Tregs. One plausible mechanism could operate through IL-2R-mediated activation of the STAT5/Bcl-xL pathway [55 , 56 ], which can be activated independently of mTOR [57 , 58 ]. Indeed, STAT5 binds to the promoter of the Foxp3 gene and plays a critical role in the development and homeostasis of Tregs [59 ]. Therefore, it may be possible to design rapamycin-assisted tolerance protocols, where the survival of Tregs can be enhanced further by increasing STAT5 signaling.
In summary, we showed here that antigen stimulation of naïve CD4 T cells in the presence of mTOR blockade by rapamycin favors their differentiation into the Foxp3+ phenotype, and these cells can participate in mediation of antigen-specific tolerance. This may be useful clinically in design of tolerance protocols to prevent graft rejection, treat autoimmunity, or enhance allergen-specific immunotherapy. Some autoimmune diseases may actually be driven by hypersignaling of the mTOR pathway in lymphocytes, and mTOR may also be an excellent target for dampening the immune response in these situations [60 ]. However, it is likely that pathogenic effector T cells exposed to a chronic source of antigen will also be able to develop resistance to rapamycin. Thus, effector Th1 cells express increased levels of Pim-1 [61 ], one of the kinases that overlaps in function with mTOR. In fact, deficiency in Pim kinases greatly increases sensitivity of T cells to effects of rapamycin on cell growth, survival, and cell-cycle activity [62 ]. Additional mechanisms of resistance to rapamycin have also been defined in cancer studies, as the drug and its analogs are being tested for their antineoplastic potential [63 ]. Some of these include dysregulation of protein phosphatase and tensin homologue expression, decreased levels of 4E-BP1, loss of the cdc inhibitor p27Kip1, shift of mTOR into a relatively rapamycin-insensitive complex with rictor, and others [63 ]. Our results suggest that rapamycin may also dampen the immune response to malignancies by contributing to induction of cancer-specific Tregs. Clearly, considerable work remains in devising optimal approaches to modify the immune responses in different clinical situations.
Received December 23, 2007; revised January 17, 2008; accepted January 22, 2008.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  K. Bocian, J. Borysowski, P. Wierzbicki, J. Wyzgal, D. Klosowska, A. Bialoszewska, L. Paczek, A. Gorski, and G. Korczak-Kowalska Rapamycin, unlike cyclosporine A, enhances suppressive functions of in vitro-induced CD4+CD25+ Tregs Nephrol. Dial. Transplant., March 1, 2010; 25(3): 710 - 717. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. E. von Allmen, N. Schmitz, M. Bauer, H. J. Hinton, M. O. Kurrer, R. B. Buser, M. Gwerder, S. Muntwiler, T. Sparwasser, R. R. Beerli, et al. Secretory phospholipase A2-IID is an effector molecule of CD4+CD25+ regulatory T cells PNAS, July 14, 2009; 106(28): 11673 - 11678. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Y. Wuest, J. Willette-Brown, S. K. Durum, and A. A. Hurwitz The influence of IL-2 family cytokines on activation and function of naturally occurring regulatory T cells J. Leukoc. Biol., October 1, 2008; 84(4): 973 - 980. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
