Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2009 Aug 15.
Published in final edited form as: J Immunol. 2008 Aug 15;181(4):2506–2512. doi: 10.4049/jimmunol.181.4.2506

γc-signaling cytokines induce a regulatory T cell (Treg) phenotype in malignant CD4+ T lymphocytes

M Kasprzycka ≠, Q Zhang ≠, A Witkiewicz ø, M Marzec ≠, M Potoczek ø, X Liu ≠, HY Wang ≠, M Milone ≠, S Basu ≠, J Mauger ≠, JK Choi ≠, T Abrams ε, S Hou ≈, AH Rook ¢, E Vonderheid £, A Woetmann §, N Odum §, MA Wasik ≠,*
PMCID: PMC2586884  NIHMSID: NIHMS54944  PMID: 18684941

Abstract

Here we demonstrate that malignant mature CD4+ T lymphocytes derived from cutaneous T-cell lymphomas (CTCL) variably display some aspects of the Tregulatory (Treg) phenotype. Whereas seven cell lines representing a spectrum of primary cutaneous T-cell lymphoproliferative disorders expressed CD25 and TGF-β, the expression of FOXP3 and, to a lesser degree, IL-10 was restricted to two CTCL cell lines that are dependent on exogeneous IL-2. IL-2, IL-15 and IL-21, all of which signals through receptors containing the common γ (γc) chain, induced expression of IL-10 in the IL-2-dependent cell lines as well as primary leukemic CTCL cells. However, only IL-2 and IL-15, but not IL-21, induced expression of FOXP3. The IL-2-triggered induction of IL-10 and FOXP3 expression occurred by signaling through STAT3 and STAT5, respectively. Immunohistochemical analysis of the CTCL tissues revealed that FOXP3-expressing cells were common among the CD7-negative enlarged atypical and small lymphocytes at the early, skin patch and plaque stages. Their frequency was profoundly diminished at the tumor stage and in the CTCL lymph node lesions with or without large cell transformation. These results indicate that the Treg cell features are induced in CTCL T cells by γc signaling cytokines such as IL-2 and do not represent a fully pre-determined, constitutive phenotype independent of the local environmental stimuli to which these malignant mature CD4+ T cells become exposed.

Introduction

T-cell lymphomas represent a heterogeneous group of lymphoproliferative disorders with most derived from the CD4+ helper/inducer T-cell subset (1, 2). Primary T-cell lymphoproliferative disorders of skin, in particular the cutaneous T-cell lymphomas (CTCL), represent the most frequently occurring subset; they display a tendency to progress over time to the more malignant forms. Accordingly, the early lesions of CTCL typically present as limited skin patches or plaques, whereas the more advanced lesions form intracutaneous tumors. Sezary Syndrome (SS) represents a leukemic form of CTCL in which the malignant (Sezary) T cells sometimes comprise a vast majority of the peripheral blood lymphocytes. As a result of disease progression, CTCL may involve lymph nodes and, less frequently, bone marrow and internal organs. Finally, CTCL can undergo a large cell transformation, which typically results in a highly aggressive clinical course.

T regulatory (Treg) cells represent a subset of CD4+ T lymphocytes capable of inhibiting immune responses against a large spectrum of antigens including the ones expressed by malignant cells (3, 4). The experimental evidence indicates that while the lymphocytes designated “natural” Treg cells develop in the thymus, those labeled “induced” Treg cells, acquire the Treg phenotype as mature post-thymic cells in response to an antigenic stimulation (3, 4). Whereas all Treg cells typically express the α chain of the IL-2 receptor (IL-2Rα, CD25), expression of the transcription factor FOXP3 is believed to represent the hallmark of the “natural” Treg cells and secretion of IL-10 of the “induced” Treg lymphocytes (5). In addition to secreting IL-10, the latter are also capable of producing another immunosuppressive cytokine TGF-β (6).

IL-2, IL-15, and IL-21 belong to the cytokine family that signals through receptors sharing the common γ chain (γc). In addition to the γc chain, the IL-2R contains a second signaling chain β and, in the case of a high affinity IL-2R, an IL-2 specific, signal non-transducing α chain. Similar to IL-2, IL-15 binds to the entire γc/β chain receptor unit but also utilizes an IL-15 specific, non-transducing α chain. Not surprisingly IL-2 and IL-15 share a number of functional properties including the fostering of T-, NK-, and B-cell proliferation and maturation although certain activities unique to each of the cytokines have been described (7, 8).

IL-21 also displays pleiotropic effects on immune cells (9) with its ability to boost cytotoxicity of NK (10) and CD8+ T cells (11) being the best characterized. Beside the γc, IL-21R contains its own distinct signal transducing α chain. In response to ligand stimulation, IL-2R, IL-15R, and IL-21R activate Jak1 and Jak3. These kinases phosphorylate the receptors as well as several signal-transducing proteins. Among the signaling proteins activated by the receptor/Jak complex STAT5 and STAT3 are particularly important since they are involved in key cellular functions including proliferation, differentiation, and survival (12, 13). Persistent activation of STAT3 and STAT5 has been identified in a large array of malignant cell types (14) including cell lines derived from CTCL and related lymphoproliferative disorders of the skin (15, 16).

Here we report that IL-2 dependent CTCL-derived cell lines as well as the primary leukemic CTCL cells can be induced by IL-2 and IL-15 to acquire key phenotypic features of Treg cells, i.e. expression of IL-2Rα (CD25) and FOXP3 as well as the ability to secrete IL-10. IL-21 was also able to induce IL-10 secretion but not FOXP3 expression. The frequency of FOXP3+ cells in CTCL tissues inversely correlated with advancement of the disease. These findings indicate that CTCL cells display the Treg phenotype in a γc-signaling cytokine dependent fashion and that acquisition of this phenotype may protect the malignant cells from elimination by an immune response, in particular in the early stages of CTCL.

Materials and Methods

Cell lines, CTCL cells and tissues

The cell lines were described previously (15, 16). In brief, IL-2-dependent CTCL cell lines Sez-4 and SeAx were derived from the leukemic cells and skin lesions, respectively. MyLa2056 and 3676 were derived from the advanced CTCL. PB-1, 2A, and 2B cell lines were established from a patient with a primary cutaneous CD30+ T-cell lymphoproliferative disorder. The leukemic cells used in the study were from CTCL patients with a high lymphocytosis and were >90% pure as determined by the CD4:CD8 ratio and CD7 and/or CD26 loss by the CD4+ T cells. Cell lines and primary malignant cells were cultured at 37°C with 5% CO2 in RPMI 1640/10% FBS medium. Twenty two skin biopsies and 9 lymph node samples were obtained from CTCL patients for diagnostic purposes. Skin biopsies included 7 patch, 9 plaque, and 6 tumor stage lesions. Six of the lymph nodes showed standard small to intermediate nuclear size morphology and 3 displayed large cell transformation of CTCL.

Jak kinase inhibitors

Pan-Jak (Jak I; Calbiochem) inhibits all four members of the Jak family with the IC50 of 15 nM for Jak1, 1 nM for Jak2, 5 nM for Jak3 and 1 nM for Tyk2 in in vitro kinase inhibition assay (17). Jak3 inhibitor was synthesized according to the published structure (18) and displays the in vitro IC50 kinase inhibition of Jak3 at 2nM for Jak3, 20nM for Jak2, and 100nM for Jak1.

Immunofluorescence Staining

Cell surface protein expression analysis was performed by using the conjugated antibodies CD25-FITC and CD4-PE (BD PharMingen) and flow cytometry (FACSCalibur; BD Biosciences) analysis that employed the CELLQUEST PRO software.

Human sCD25 EIA

Concentration of soluble CD25 in culture supernatants was evaluated using Quantikine Human sCD25 kit (R&D Systems). In brief, 50μl/well of sample or standard were added to a CD25 antibody pre-coated plate and mixed with 100μl of CD25 antibody-horseradish peroxidase conjugate. The concentration of soluble CD25 was determined by optical density (OD) determination using an EIA plate reader.

Human IL-10 EIA

Concentration of IL-10 in culture supernatants was evaluated using Quantikine Human IL-10 kit (R&D Systems). In brief, 200μl/well of sample or standard were added to an IL-10 antibody pre-coated plate followed by replacement with 200μl of the IL-10 antibody-peroxidase conjugate and OD determination.

Human TGF-β EIA

Cells were cultured in an RPMI 1640/2% BSA medium for 4h. Concentration of TGF-β in the supernatants was evaluated using Quantikine human TGF-β kit (R&D Systems). In brief, the latent TGF-β was activated to the immunoreactive form by incubating 0.5 ml of supernatant with 0.1 ml of 1 M HCl. After 10″, the samples were neutralized with 0.1 ml of 1.2 M NaOH/0.5 M Hepes. Next, 200μl/well of the samples or standard were added to a TGF-β antibody pre-coated plate flowed by addition of 200μl TGF-β-peroxidase conjugate and OD determination.

Western Blot

This procedure was performed as described (15). In brief, the cells were lysed in a buffer supplemented with phenylmethylsulfonyl fluoride, phosphatase inhibitor cocktails I and II (Sigma), and protease inhibitor mixture (Roche). For normalization of the gel loading, the protein extracts were assayed with Lowry method (Bio-Rad Dc protein assay). Cell lysates were separated on a 10% polyacrylamide/SDS gel, and transferred to polyvinylidene difluoride membranes (Amersham Pharmacia). Proteins were detected with antibodies against β-actin, STAT3, STAT5 (Santa Cruz Biotechnology), FOXP3 (rat PCH101, eBioscience), phospho (Y705)-STAT3 and phospho (Y624)-STAT5 (Cell Signaling), and the secondary, peroxidase-conjugated antibodies (Santa Cruz Biotechnology). Blots were developed by using the SuperSignal West Dura nylon membranes (Pierce) and exposed to x-ray film (Kodak).

Induction of FOXP3 expression in normal CD4+ T lymphocytes

Peripheral blood mononuclear cells were obtained from healthy donors by apheresis. Peripheral blood lymphocytes were isolated by centrifugal elutriation, and CD4+ T cells were purified by negative selection using the Rosette-Sep method for purification of CD4+ T cells (Stem Cell Technologies) as described by the manufacturer with the typical outcome of >90% CD3+CD4+ T cells as determined by flow cytometry. T cells were stimulated for 24 hrs in complete RPMI1640/FBS medium with anti-CD3 (clone OKT3) and anti-CD28 (clone 9.3) monoclonal antibodies immobilized on tosyl-activated paramagnetic beads (Invitrogen, Carlsbad, CA) at a T-cell to bead ratio of 3:1.

Cell proliferation suppression assay

To test the suppressive capacity of the CTCL cells, Sez-4 and SeAx cell lines were pre-cultured in the presence or absence of IL-2 for 24–48 hrs and subsequently co-cultured in duplicates with the CFSE-labeled normal peripheral blood mononuclear responder cells (100,000 cells per culture condition) at the ratios of 1:1 (CTCL cell line to responder cells), 1:2, 1:4 and 1:8 in a 96 flat bottom well plate in the presence of beads coated with anti-CD3 Ab (Dynal Biotech) at a ratio of 1:1 (beads to responder cells) in a volume of 200 μl of RPMI 1640/10% FBS medium additionally supplemented with 1% penicillin and streptomycin (GIBCO) and 0.01 M HEPES (GIBCO). After 4 days of the co-culture, the cells were stained with an anti-CD8 antibody and analyzed by FACS for the CFSE labeling pattern of the responder cells.

siRNA

Sez-4 cell line was transfected with 100 nM siRNA specific for STAT3, STAT5, and non-targeting control twice at 24-h intervals (all siRNA were from Dharmacon) using lipofectamine (DMRIE-C; Invitrogen). Efficiency of the siRNA-promoted protein depletion was assessed by Western blotting.

Chromatin Immunoprecipitation (ChIP) Assay

The cells were treated with formaldehyde to cross-link DNA and attached proteins, washed, exposed to lysis buffer, and the chromatin-containing cell lysates sonicated. The lysates were pre-cleared and later incubated with the anti-Stat5 or control IgG antibodies (Santa Cruz). DNA–protein immunocomplexes were precipitated with protein A/G-agarose beads and treated with RNase A and proteinase K; the DNA samples were extracted with phenol chloroform, precipitated with ethanol and PCR-amplified by using primers (5′-TCACCTACCACATCCACCAG-3′ and 5′-GACACCACGGAGGAAGAGAA-3′) specific for the enhancer region located in intron 1 of the FOXP3 gene (19).

Immunohistochemical staining

After the heat antigen retrieval step, the tissues sections were incubated for 60 min at room temperature with the anti-FOXP3 antibody (eBioscience clone 14-4777) diluted at 1:100 and stained using the ABC detection system (Vector Laboratories). Scoring of the FOXP3-expressing cells was performed by counting positive cells per 100 of the atypical and all the lymphoid cells in three representative areas of the lesions. In the double staining experiment, immunoperoxidase staining of CD7 (Leica Microsystems) was done on the BondMax at a dilution of 1:100. Bond polymer refine detection with DAB as the chromogen was used with epitope retrieval 2 for 20 minutes (Leica). Alkaline phosphatase staining of FoxP3 (eBioscience) was done on the bench top at a dilution of 1:20. Ultravision LP detection system (Labvision Corp.) was used with fast red chromogen as the second label. Slides were lightly counterstained with hematoxylin.

Results

Expression of Treg-related proteins in cell lines derived from T-cell lymphomas involving skin

To determine if malignant CD4+ T cells display a Treg phenotype, we examined seven cell lines derived from a spectrum of primary cutaneous T-cell lymphoproliferative disorders ranging from the Sezary Syndrome to the overt (anaplastic) transformed large T-cell lymphoma. All cell lines expressed IL-2Rα/CD25 as determined by flow cytometry analysis of the cells (data not presented) or by EIA of the cell culture supernatants where the soluble form of the receptor (sCD25) could be detected, typically in high concentration (Fig. 1A). Similarly, the cells secreted variable but generally high amount of TGF-β. In contrast, IL-10 was produced by only three cell lines (Fig. 1C). The cytokine could not be detected in supernatants of the remaining four cell lines derived from the advanced, clinically highly aggressive lymphomas (Fig. 1C). Expression of FOXP3 was even more restricted since the identified FOXP3 doublet was detectable in only two cell lines but not in the other five lines (Fig. 1D; upper panel). This FOXP3 doublet looked essentially the same as the FOXP3 doublet expressed by normal CD4+ T cells upon their stimulation with anti-CD3/CD28 antibody-coated beads (Fig. 1D; lower panel).

FIGURE 1.

FIGURE 1

Open in a new tab

Expression of Treg-related proteins in cell lines derived from the T-cell lymphomas involving skin. Cell lines derived from CTCL, both IL-2-dependent (Sez-4 and SeAx) and IL-2-independent (MyLa 2059 and MyLa 3676) as well as cell lines from a progressive lymphoproliferative disorder similar to CTCL (PB-1, 2A, and 2B) were examined for expression of the secreted soluble IL-2Ra (sCD25: panel A), TGF-β (B), IL-10 (C), and intracellular FOXP3 with expression of β-actin serving as a positive control (D; upper panel). Normal CD4+ T cells stimulated with beads coated with anti-CD3 and anti-CD28 antibodies to induce FOXP3 expression, served as an additional control (D; lower panel).

Role of γc-signaling cytokines in the expression of Treg phenotype by the T-cell lymphoma cells

Because the FOXP3-expressing cell lines: Sez-4 and SeAx, were derived from Sezary Syndrome, the moderately aggressive stage of the disease and, accordingly, both require IL-2 to support their growth, in contrast to all the other cell lines which grow without addition of any exogeneous cytokines, we examined next if IL-2 contributes to the ability of the CTCL cells to express the Treg phenotype. In addition to IL-2, we employed IL-15 which signals through the same IL-2Rβ/γc heterodimer as IL-2 as well as IL-21, a cytokine which signals through the γc chain-containing receptor similar to IL-2R and IL-15R. IL-2, IL-15 and, to a lesser degree, IL-21 were capable of inducing CD25 expression in the Sez-4 and SeAx cells in the cell membrane (not shown) and the soluble (Fig. 2A) form. All three cytokines also promoted secretion of TGF-β (Fig. 2B) and IL-10 (Fig. 2C) with the much larger amount of both, in particular IL-10, being produced by SeAx than Sez-4. Of note, IL-2 as well as IL-15 induced expression of FOXP3 (Fig. 2D). In contrast, IL-21 failed to induce FOXP3 expression. To demonstrate that the IL-2-triggered induction of FOXP3 expression in these cells impacts on their immunosuppressive capacity, we co-cultured Sez-4 and SeAx cells with normal peripheral blood mononuclear cells stimulated with a bead-immobilized CD3 antibody. Whereas the cell lines were pre-cultured either with or without IL-2 for 24 or 48 hrs (IL-2 deprivation for these lengths of time affects their proliferation but not viability; M. Marzec and M. Wasik, unpublished data), the mononuclear cells were pre-labeled with CSFE to track their cell division rate. As shown in Fig. 2E depicting representative result, the CTCL cell lines were able to markedly inhibit proliferation of the CD3-stimulated lymphocytes. Whereas the suppression was exerted by both IL-2 pre-stimulated and IL-2 deprived cells, the former were significantly much more potent in a highly reproducible manner. Although the nature of suppressive effect of the IL-2-deprived cells in this assay is uncertain and likely multi-factorial, these results strongly support the notion that the IL-2/STAT5-induced expression of FOXP3 fosters immunosuppressive properties of the CTCL cells.

FIGURE 2.

FIGURE 2

Open in a new tab

Role of the γc-signaling cytokines in the expression of Treg phenotype by the IL-2-dependent CTCL-derived cell lines. Sez-4 and SeAx cell lines were starved of IL-2 for up to 48 hrs and exposed to 100U of IL-2, 20 ng/mL of IL-15, and 100 ng/mL of IL-21, and evaluated for the expression of sCD25 (A), TGF-β (B), IL-10 (C), and FOXP3 with expression of phospho (Y705)-STAT3, phospho (Y694)-STAT5, and β-actin serving as controls (D). E. The IL-2-depleted or pre-exposed Sez-4 and SeAx cell line were co-cultured in duplicates for 4 days with CFSE-labeled normal peripheral blood mononuclear cells stimulated with CD3 antibody-coated beads at the depicted cell line to mononuclear cells ratios. Proliferative rate of the CD8+ T-cell subset was determined by FACS analysis.

Previous studies have demonstrated that STATs are involved in the induction of FOXP3 and IL-10 expression in other types of lymphoid cells (19–23). To determine if this is also the case in the CTCL cells, we examined the effect of siRNA-mediated STAT3 and STAT5 depletion on expression of these two proteins. As shown in Fig. 3A, loss of STAT5 but not STAT3 completely inhibited FOXP3 expression. In contrast, depletion of STAT3 profoundly suppressed the IL-10 expression, whereas STAT5 loss had a more limited effect (Fig. 3B). To provide additional evidence that STAT5 plays a role in induction of the FOXP3 expression, we have performed a ChIP assay using a STAT5-specific antibody and a primer set specific for the regulatory region of the FOXP3 gene (19). As shown in Fig. 3C, STAT5 displayed a strong binding to the FOXP3 gene.

FIGURE 3.

FIGURE 3

Open in a new tab

Role of the Jak/STAT signaling in the IL-2-mediated induction of the Treg phenotype. Sez-4 cell line was treated with STAT3, STAT5, or control (non-sense) siRNA and examined upon re-stimulation with IL-2 for expression of IL-10 (A) or FOXP3 (B). C. Sez-4 cell line stimulated with IL-2 was examined for STAT5 binding in vivo to the FOXP3 gene enhancer region. D. Sez-4 cell line was stimulated with IL-2 in the presence of the pan-Jak or Jak3 inhibitor (300 nM each) or medium alone and evaluated for the synthesis of IL-10. E. Sez-4 and SeAx cell lines were exposed to medium or IL-2 in the presence or absence of the pan-Jak inhibitor and evaluated for expression of FOXP3. ND; not done.

Given the critical role of the Jak1/Jak3 tyrosine kinase complex in IL-2 signaling and STAT activation (14), it was not surprising that inhibition of Jak enzymatic activity by a pan-Jak small molecule inhibitor (22) suppressed FOXP3 expression (Fig. 3D). Inhibition of Jak activity using the pan-Jak as well as a second Jak inhibitor that inactivates preferentially Jak3 (23) profoundly suppressed expression of IL-10 (Fig. 3E).

Role of the γc-signaling cytokines in expression of IL-10 and FOXP3 by primary CTCL cells

To confirm that the key observations in regard to the cytokine-induced expression of a Treg phenotype apply also to native, uncultured cells, we have isolated circulating Sezary cells from several highly leukemic CTCL patients. As shown in Fig. 4A, both IL-2 and IL-21 induced IL-10 synthesis in CTCL cells from all four patients examined, similarly to Sez-4 and SeAz cell lines. Some variability was, however, observed with three patients responding more vigorously to IL-21 than to IL-2. IL-2 also induced FOXP3 expression in cells from all five patients (Fig. 4B). As in the cell lines, IL-21 did not induce FOXP3 expression in the native CTCL cells.

FIGURE 4.

FIGURE 4

Open in a new tab

IL-2 induced expression of Treg-related proteins in the primary CTCL cells. Leukemic (Sezary) CTCL cells were exposed to medium, 100 U of IL-2 or 100 ng/mL of IL-21 and examined for secretion of IL-10 (A) and expression of FOXP3 (B). As depicted, FOXP3 expression in the patient #2 sample was determined in the nuclear rather than the whole cell protein lysate sample with cytoplasmic lysate serving as a control.

Stage-dependent expression of FOXP3 in skin and lymph nodes involved by CTCL

Finally, to determine if FOXP3 can be detected in CTCL tissues, we examined by immunohistochemistry biopsy samples encompassing various stages of the disorder ranging from skin patch lesions to lymph nodes with a large cell transformation. As depicted by the representative images in Fig. 5, cells that expressed FOXP3 were the most frequent in the patch (Fig. 5A) and plaque (Fig. 5B) lesions since on average approximately 40% of the atypical Sezary-type cells as well as all lymphocytes expressed the protein at these early stages of CTCL. Frequency of the FOXP3-expressing cells was markedly decreased in the skin tumor lesions with approximately 5% of the atypical and 10% of all lymphocytes displaying positive staining (Fig. 5C). Similarly, a rather small number of FOXP3 positive cells was present in the lymph nodes containing CTCL lesions with only <10% of the atypical and <20% of all lymphocytes showing the staining (data not shown). The percentage of staining cells was even lower (<5%) among the vastly prevailing atypical cells in the CTCL cases that had undergone large cell transformation (not shown). To provide additional evidence that at least some of the FOXP3-expressing cells represent the malignant Sezary cells, we have performed in selected cases a double staining for FOXP3 and CD7, a surface receptor typically lost in CTCL cells. As shown in Fig. 5D many of the lymphocytes in the dermis and the intra-epithelial lesions (Pautrier abscesses) displayed loss of CD7. Many of these FOXP3+/CD7− cells had enlarged, hyperchromatic, and irregular nuclei characteristic for Sezary cells (Fig. 5D, inset).

FIGURE 5.

FIGURE 5

Open in a new tab

Stage dependent FoxP3 expression in tissues involved by CTCL. Representative immunohistochemical stains for FOXP3 protein of a skin patch (A), plaque (B) and tumor (C) stage. (D) Double staining of a plaque stage for FOXP3 and CD7 cell membrane protein. The main images represent 200X and the insets 400X magnification.

Discussion

Whereas numerous studies that have focused on Treg lymphocytes have undoubtedly fostered our understanding of these cells, they have also left some important questions unanswered as well as created some controversies. Although the earlier findings suggested that the “natural” FOXP3-expressing Treg cells comprise a distinct subset of the CD4+ T cells pre-programed in the thymus, the growing body of evidence indicates that at least some aspects of the Treg phenotype can be modulated in the fully mature T cells. Accordingly, stimulation through the antigen T-cell receptor (TCR) induces FOXP3 in a substantial proportion of the peripheral, “naive” CD4+ T cells in a TGF-β-dependent manner (24). Interestingly, FOXP3 expression alone seems insufficient to confer the Treg function (24, 25) and additional steps such as post-transcriptional modifications of FOXP3 (26) and its interaction with other transcription factors such as NF-AT (27) and Runx-1 (28) are required. In addition to TCR and TGF-β, IL-2 has been shown to induce FOXP3 expression in CD4+CD25+ T cells (19) with circumstantial evidence strongly favoring the key role of STAT5 in the process (19–22). Our study demonstrates that IL-2 alone seems sufficient to potently induce FOXP3 expression in the CTCL-derived malignant CD4+ T cells and that both the Jak1/3 complex and STAT5 but not STAT3 are required to transduce the FOXP3 inducing signal. This observation not only supports the notion that FOXP3 expression can be regulated in post-thymic T lymphocytes but also suggests that CTCL cells have been pre-primed in vivo to bypass the need for in vitro stimulation by TCR to express FOXP3 in response to IL-2, possibly by the TCR engagement in vivo. They also may had been exposed to TGF-β in vivo and may not require it in vitro. However, by virtue of being produced by CTCL cells, albeit sometimes in quite small amount as is the case with the Sez-4 cell line (Fig. 1B), TGF-β may contribute to the FOXP3 induction through an autocrine mechanism.

Our observation that IL-15 is also able to induce FOXP3 expression in the CTCL cells is not surprising given that IL-2 and IL-15 share the same IL-2/IL-15Rβ/γc signaling receptor complex. It is also in agreement with the recent findings that the IL-2/IL-15Rβ deficient mice display markedly diminished development and activity of Treg cells (20, 29).

It also seems that the separation of the “natural” FOXP3-expressing Treg cells from the immunosuppressive cytokine producing “induced” Treg may not be as strict as originally suggested. Our findings indicate that in response to IL-2 the same clonal T-cell populations as represented by both CTCL cell lines and primary cells are capable of simultaneously expressing FOXP3, IL-10 and TGF-β (Fig. 2 and Fig. 4). This observation suggests that the constelation of stimuli to which functionally mature CD4+ T cells have been exposed in vivo and in vitro may be as much, if not more, important than the thymic pre-programing, in regard to the type and degree of their Treg differentiation. Hence, a substantial heterogeneity likely occurs even within the same clonal CD4+ T-cell populations depending on the external stimuli received in the local cellular millieu. Because the IL-2-induced expression of FOXP3 is mediated chiefly by STAT5 and IL-10 expression primarily by STAT3 (Fig. 3), one could argue that the pattern of STAT activation induced by cytokines and other stimuli may be critical in directing the differentiation. Our observation that IL-21, which activates STAT3 but not STAT5, induced expression of IL-10 and not of FOXP3 (Fig. 3), supports this conclusion. Furthermore, activation of STATs most certainly is only one of the variables impacting on functional phenotype of the cell. Although Sez-4 and SeAx, both CTCL-derived cell lines, displayed similar pattern of STAT activation in response to IL-2, IL-15, and IL-21, there markedly varied in the amount of IL-10 and TGF-β produced in response to these cytokines (Fig. 2). Similarly, circumstantial evidence indicates that STAT5 alone is insufficient to induce transcription of the FOXP3 gene. Accordingly, the IL-2 independent T-cell lines depicted in Fig. 1 express persistently activated STAT5 (17, 37) yet, as we show in the figure, do not express FOXP3. The identity and relative contribution of other proteins, which cooperate with STAT5 or, possibly, inhibit its ability to induce FOXP3 gene remains to be elucidated.

Several recent reports indicate that the expression of at least some aspects of the Treg phenotype may be quite common among lymphomas derived from CD4+ T cells. Accordingly, malignant cells from the HTLV1-related adult-type lymphoma/leukemia (ATLL) universally express CD25 and seemingly less frequently FOXP3 (30, 31). HTLV1-encoded proteins apparently foster acquisition of the Treg phenotype since the virus is able to induce in the CD4+ T cells expression of CD25, FOXP3, and TGF-β (32). Although the primary leukemic CTCL cells have also been shown to express a Treg phenotype, this phenotype was demonstrated after in vitro stimulation by the autologous dendritic cells loaded with apoptotic CTCL cells in the presence of four different cytokines GM-CSF, IL-2, IL-7, and IL-4 (33) in that experimental system. In light of our findings, it is likely that IL-2 and, possibly, dendritic cell-derived IL-15 played in this system an important role in acquisition of the Treg phenotype, in general, and induction of the FOXP3 expression, in particular. Finally, we have demonstrated that T-cell lymphomas expressing the chimeric NPM/ALK tyrosine kinase also display the Treg phenotype, characterized mainly by secretion of IL-10 and TGF-β (21). Of note, the phenotype is induced by the NPM/ALK kinase itself through activation of STAT3. We have also reported that the NPM/ALK-transformed cells express FOXP3. In retrospect, the identified protein does not correspond to the FOXP3 doublet described in this study. Not only was the band single, but its molecular weight was higher than that of the doublet. It corresponds to the weak band noted also in the CTCL and related cell lines (Fig. 1D and data not shown). Of note, the issue of cross-reactivity of the anti-FOXP3 PCH101 antibody used by us in Western blots (but not in the immunohistochemical studies) was raised recently by others (24) in regard to the flow cytometry evaluations, in which the molecular weight of the detected protein(s) can not be determined.

It is somewhat unexpected that FOXP3+ cells are prevalent in the early stages of CTCL such as patch and plaque but disappear during progression of the disease to tumors, lymph node involvement and large cell transformation. A similar observation was made in two other recent studies (34, 35) with an interpretation that the FOXP3+ cells are exclusively non-neoplastic. However, as we have shown many of the staining cells are clearly atypical (Sezary) CTCL cells, often located in the intra-epithelial lesions (Pautrier abscesses (Fig. 5B, inset) and display loss of CD7 expression (Fig. 5D, inset). Furthermore, as discussed above, the malignant CTCL cells are capable of expressing FOXP3 upon in vitro exposure to IL-2, IL-15 (Fig. 2D) and possibly other stimuli which they encounter in vivo. Therefore, the local environmental stimuli most likely confer the Treg features on a subset of the clonal, malignant CD4+ T cells. Noteworthy, the disease stage dependence of the FOXP3 expression loss seemingly correlates with the pattern of expression of FOXP3 in the lymphoma cell lines because the cell lines corresponding to the less aggressive stage of the disease as determined by their IL-2 dependence were able to express FOXP3 in response to the cytokine, whereas the other cell lines which do not require exogeneous cytokines for their growth do not (Fig. 1D) even when stimulated with IL-2 (data not shown). These findings also suggest that FOXP3 expression may play a role in inhibiting immune responses against the CTCL cells, in particular at the early stages of the disease. Whether gaining independence of the external stimuli such as antigenic stimulation, TGF-β, or IL-2-type cytokines causes the loss of FOXP3 expression in the malignant CD4+ T cells in the advanced stages of CTCL remains to be determined.

The key role of the IL-2/-15Rβ/γc-associated Jak1/Jak3 kinase complex, and the STATs in induction of the FOXP3 and IL-10 expression suggests new therapeutic approaches to CTCL aimed at reversal of the Treg phenotype. Accordingly, the small molecule inhibitors that target the Jak kinases profoundly inhibits expression of IL-10 (Fig. 3C) and FOXP3 (Fig. 3D) in the CTCL cells. Given the relatively high specificity of the Jak3 inhibitor (18) and the restriction of the γc expression to the immune cells, targeting Jak3 therapeutically in CTCL and, possibly, other CD4+ T-cell lymphomas may prove effective and relatively devoid of side effects. One could argue that STAT5 also represents an attractive therapeutic target, given its role in induction of the FOXP3 expression (Fig. 4B). However, according to our knowledge no suitable small molecule inhibitor of the STAT5 activity has been developed so far. Finally, treatment of CTCL patients with IL-21 may also be considered since, in contrast to IL-2 and IL-15, this cytokine does not activate STAT5 and, consequently, does not induce FOXP3 expression (Fig. 2D), although it variably promotes synthesis of TGF-β (Fig. 2B) and IL-10 (Fig. 2C). Of note, we have recently shown that IL-21 exerts in the CTCL cells a very limited effect on the gene expression, cell signaling, and proliferation although it provides some degree of protection from apoptosis (36, 37). Importantly, IL-21 strongly activates normal immune CD8+ T cells and NK cells derived from both mice and humans including CTCL patients (38–41). These observations indicate that IL-21 may be highly efficacious in CTCL therapy. It is also possible that the appropriately timed application of the Jak3 inhibitor to suppress growth and survival of the CTCL cells alternated with the IL-21 administration to boost the anti-CTCL immune response may prove particularly beneficial in this and similar lymphoproliferative disorders derived from the mature CD4+ T lymphocytes.

Acknowledgments

Supported in part by grants from the National Cancer Institute and Danish Cancer Society.

References

  • 1.Willemze R, Jaffe ES, Burg G, Cerroni L, Berti E, Swerdlow SH, Ralfkiaer E, Chimenti S, Diaz-Perez JL, Duncan LM, Grange F, Harris NL, Kempf W, Kerl H, Kurrer M, Knobler R, Pimpinelli N, Sander C, Santucci M, Sterry W, Vermeer MH, Wechsler J, Whittaker S, Meijer CJ. WHO-EORTC classification for cutaneous lymphomas. Blood. 2005;105:3768–3785. doi: 10.1182/blood-2004-09-3502. [DOI] [PubMed] [Google Scholar]
  • 2.Dummer R. Future perspectives in the treatment of cutaneous T-cell lymphoma (CTCL) Semin Oncol. 2006;33:S33–36. doi: 10.1053/j.seminoncol.2005.12.021. [DOI] [PubMed] [Google Scholar]
  • 3.O’Garra A, Vieira P. Regulatory T cells and mechanisms of immune system control. Nat Med. 2004;10:801–805. doi: 10.1038/nm0804-801. [DOI] [PubMed] [Google Scholar]
  • 4.Chattopadhyay S, Chakraborty NG, Mukherji B. Regulatory T cells and tumor immunity. Cancer Immunol Immunother. 2005;54:1153–1161. doi: 10.1007/s00262-005-0699-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Vieira PL, Christensen JR, Minaee S, O’Neill EJ, Barrat FJ, Boonstra A, Barthlott T, Stockinger B, Wraith DC, O’Garra A. IL-10-secreting regulatory T cells do not express Foxp3 but have comparable regulatory function to naturally occurring CD4+CD25+ regulatory T cells. J Immunol. 2004;172:5986–5993. doi: 10.4049/jimmunol.172.10.5986. [DOI] [PubMed] [Google Scholar]
  • 6.Randolph DA, Fathman CG. Cd4+Cd25+ regulatory T cells and their therapeutic potential. Annu Rev Med. 2006;57:381–402. doi: 10.1146/annurev.med.57.121304.131337. [DOI] [PubMed] [Google Scholar]
  • 7.Waldmann TA. The biology of interleukin-2 and interleukin-15: implications for cancer therapy and vaccine design. Nat Rev Immunol. 2006;6:595–601. doi: 10.1038/nri1901. [DOI] [PubMed] [Google Scholar]
  • 8.Cornish GH, Sinclair LV, Cantrell DA. Differential regulation of T-cell growth by IL-2 and IL-15. Blood. 2006;108:600–608. doi: 10.1182/blood-2005-12-4827. [DOI] [PubMed] [Google Scholar]
  • 9.Leonard WJ, Spolski R. Interleukin-21: A modulator of lymphoid proliferation, apoptosis and differentiation. Nat Rev Immunol. 2005;5:688–698. doi: 10.1038/nri1688. [DOI] [PubMed] [Google Scholar]
  • 10.Kasaian MT, Whitters MJ, Carter LL, Lowe LD, Jussif JM, Deng B, Johnson KA, Witek JS, Senices M, Konz RF, Wurster AL, Donaldson DD, Collins M, Young DA, Grusby MJ. IL-21 limits NK cell responses and promotes antigen-specific T cell activation: a mediator of the transition from innate to adaptive immunity. Immunity. 2002;16:559–569. doi: 10.1016/s1074-7613(02)00295-9. [DOI] [PubMed] [Google Scholar]
  • 11.Zeng R, Spolski R, Finkelstein SE, Oh S, Kovanen PE, Hinrichs CS, Pise-Masison CA, Radonovich MF, Brady JN, Restifo NP, Berzofsky JA, Leonard WJ. Synergy of IL-21 and IL-15 in regulating CD8+ T cell expansion and function. J Exp Med. 2005;201:139–148. doi: 10.1084/jem.20041057. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Moriggl R, Topham DJ, Teglund S, Sexl V, McKay C, Wang D, Hoffmeyer A, Van Deursen J, Sangster MY, Bunting KD, Grosveld GC, Ihle JN. Stat5 is required for IL-2-induced cell cycle progression of peripheral T cells. Immunity. 1999;10:249–259. doi: 10.1016/s1074-7613(00)80025-4. [DOI] [PubMed] [Google Scholar]
  • 13.Buitenhuis M, Coffer PJ, Koenderman L. Signal transducer and activator of transcription 5 (STAT5) Int J Biochem Cell Biol. 2004;36:2120–2124. doi: 10.1016/j.biocel.2003.11.008. [DOI] [PubMed] [Google Scholar]
  • 14.Yu H, Jove R. The Stats of cancer--new molecular targets come of age. Nat Rev Cancer. 2004;4:97–105. doi: 10.1038/nrc1275. [DOI] [PubMed] [Google Scholar]
  • 15.Zhang Q, Nowak I, Vonderheid EC, Rook AH, Kadin ME, Nowell PC, Shaw LM, Wasik MA. Activation of Jak/STAT proteins involved in signal transduction pathway mediated by receptor for interleukin 2 in malignant T lymphocytes derived from cutaneous anaplastic large T-cell lymphoma and Sezary syndrome. Proc Natl Acad Sci U S A. 1996;93:9148–9153. doi: 10.1073/pnas.93.17.9148. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Nielsen M, Kaltoft K, Nordahl M, Röpke C, Geisler C, Mustelin T, Dobson P, Svejgaard A, Odum N. Constitutive activation of a slowly migrating isoform of Stat3 in mycosis fungoides: Tyrphostin AG490 inhibits Stat3 activation and growth of mycosis fungoides tumor cell lines. Proc Natl Acad Sci U S A. 1997;94:6764–6769. doi: 10.1073/pnas.94.13.6764. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Thompson JE, Cubbon RM, Cummings RT, Wicker LS, Frankshun R, Cunningham BR, Cameron PM, Meinke PT, Liverton N, Weng Y, DeMartino JA. Photochemical preparation of a pyridone containing tetracycle: a Jak protein kinase inhibitor. Bioorg Med Chem Lett. 2002;12:1219–1223. doi: 10.1016/s0960-894x(02)00106-3. [DOI] [PubMed] [Google Scholar]
  • 18.Changelian PS, Flanagan ME, Ball DJ, Kent CR, Magnuson KS, Martin WH, Rizzuti BJ, Sawyer PS, Perry BD, Brissette WH, McCurdy SP, Kudlacz EM, Conklyn MJ, Elliott EA, Koslov ER, Fisher MB, Strelevitz TJ, Yoon K, Whipple DA, Sun J, Munchhof MJ, Doty JL, Casavant JM, Blumenkopf TA, Hines M, Brown MF, Lillie BM, Subramanyam C, Shang-Poa C, Milici AJ, Beckius GE, Moyer JD, Su C, Woodworth TG, Gaweco AS, Beals CR, Littman BH, Fisher DA, Smith JF, Zagouras P, Magna HA, Saltarelli MJ, Johnson KS, Nelms LF, Des Etages SG, Hayes LS, Kawabata TT, Finco-Kent D, Baker DL, Larson M, Si MS, Paniagua R, Higgins J, Holm B, Reitz B, Zhou YJ, Morris RE, O’Shea JJ, Borie DC. Prevention of organ allograft rejection by a specific Janus kinase 3 inhibitor. Science. 2003;302:875–878. doi: 10.1126/science.1087061. [DOI] [PubMed] [Google Scholar]
  • 19.Zorn E, Nelson EA, Mohseni M, Porcheray F, Kim H, Litsa D, Bellucci R, Raderschall E, Canning C, Soiffer RJ, Frank DA, Ritz J. IL-2 regulates FOXP3 expression in human CD4+CD25+ regulatory T cells through a STAT-dependent mechanism and induces the expansion of these cells in vivo. Blood. 2006;108:1571–1579. doi: 10.1182/blood-2006-02-004747. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Yao Z, Kanno Y, Kerenyi M, Stephens G, Durant L, Watford WT, Laurence A, Robinson GW, Shevach EM, Moriggl R, Hennighausen L, Wu C, O’Shea JJ. Nonredundant roles for Stat5a/b in directly regulating Foxp3. Blood. 2007;109:4368–4375. doi: 10.1182/blood-2006-11-055756. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Cohen AC, Nadeau KC, Tu W, Hwa V, Dionis K, Bezrodnik L, Teper A, Gaillard M, Heinrich J, Krensky AM, Rosenfeld RG, Lewis DB. Decreased accumulation and regulatory function of CD4+ CD25(high) T cells in human STAT5b deficiency. J Immunol. 2006;177:2770–2774. doi: 10.4049/jimmunol.177.5.2770. [DOI] [PubMed] [Google Scholar]
  • 22.Burchill MA, Yang J, Vogtenhuber C, Blazar BR, Farrar MA. IL-2 receptor beta-dependent STAT5 activation is required for the development of Foxp3+ regulatory T cells. J Immunol. 2007;178:280–90. doi: 10.4049/jimmunol.178.1.280. [DOI] [PubMed] [Google Scholar]
  • 23.Kasprzycka M, Marzec M, Liu X, Zhang Q, Wasik MA. NPM/ALK oncoprotein induces Tregulatory cell phenotype by activating STAT3. Proc Natl Acad Sci U S A. 2006;103:9964–9969. doi: 10.1073/pnas.0603507103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Tran DQ, Ramsey H, Shevach EM. Induction of FOXP3 expression in naive human CD4+FOXP3+ T cells by T-cell receptor stimulation is transforming growth factor-β–dependent but does not confer a regulatory phenotype. Blood. 2007;110:2983–2990. doi: 10.1182/blood-2007-06-094656. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Gavin MA, Torgerson TR, Houston E, DeRoos P, Ho WY, Stray-Pedersen A, Ocheltree EL, Greenberg PD, Ochs HD, Rudensky AY. Single-cell analysis of normal and FOXP3-mutant human T cells: FOXP3 expression without regulatory T cell development. Proc Natl Acad Sci U S A. 2006;103:6659–6664. doi: 10.1073/pnas.0509484103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Tao R, de Zoeten EF, Ozkaynak E, Chen C, Wang L, Porrett PM, Li B, Turka LA, Olson EN, Greene MI, Wells AD, Hancock WW. Deacetylase inhibition promotes the generation and function of regulatory T cells. Nat Med. 2007;13:1299–1307. doi: 10.1038/nm1652. [DOI] [PubMed] [Google Scholar]
  • 27.Wu Y, Borde M, Heissmeyer V, Feuerer M, Lapan AD, Stroud JC, Bates DL, Guo L, Han A, Ziegler SF, Mathis D, Benoist C, Chen L, Rao A. FOXP3 controls regulatory T cell function through cooperation with NFAT. Cell. 2006;126:375–387. doi: 10.1016/j.cell.2006.05.042. [DOI] [PubMed] [Google Scholar]
  • 28.Ono M, Yaguchi H, Ohkura N, Kitabayashi I, Nagamura Y, Nomura T, Miyachi Y, Tsukada T, Sakaguchi S. Foxp3 controls regulatory T-cell function by interacting with AML1/Runx1. Nature. 2007;446:685–589. doi: 10.1038/nature05673. [DOI] [PubMed] [Google Scholar]
  • 29.Soper DM, Kasprowicz DJ, Ziegler SF. L-2Rbeta links IL-2R signaling with Foxp3 expression. Eur J Immunol. 2007;37:1817–1826. doi: 10.1002/eji.200737101. [DOI] [PubMed] [Google Scholar]
  • 30.Matsubara Y, Hori T, Morita R, Sakaguchi S, Uchiyama T. Phenotypic and functional relationship between adult T-cell leukemia cells and regulatory T cells. Leukemia. 2005;19:482–483. doi: 10.1038/sj.leu.2403628. [DOI] [PubMed] [Google Scholar]
  • 31.Kohno T, Yamada Y, Akamatsu N, Kamihira S, Imaizumi Y, Tomonaga M, Matsuyama T. Possible origin of adult T-cell leukemia/lymphoma cells from human T lymphotropic virus type-1-infected regulatory T cells. Cancer Sci. 2005;96:527–533. doi: 10.1111/j.1349-7006.2005.00080.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Walsh PT, Benoit BM, Wysocka M, Dalton NM, Turka LA, Rook AH. A role for regulatory T cells in cutaneous T-Cell lymphoma; induction of a CD4 + CD25 + FOXP3+ T-cell phenotype associated with HTLV-1 infection. J Invest Dermatol. 2006;126:690–692. doi: 10.1038/sj.jid.5700121. [DOI] [PubMed] [Google Scholar]
  • 33.Berger CL, Tigelaar R, Cohen J, Mariwalla K, Trinh J, Wang N, Edelson RL. Cutaneous T-cell lymphoma: malignant proliferation of T-regulatory cells. Blood. 2005;105:1640–1647. doi: 10.1182/blood-2004-06-2181. [DOI] [PubMed] [Google Scholar]
  • 34.Tiemessen MM, Mitchell TJ, Hendry L, Whittaker SJ, Taams LS, John S. Lack of suppressive CD4+CD25+FOXP3+ T cells in advanced stages of primary cutaneous T-cell lymphoma. J Invest Dermatol. 2006;126:2217–2223. doi: 10.1038/sj.jid.5700371. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Gjerdrum LM, Woetmann A, Odum N, Burton CM, Rossen K, Skovgaard GL, Ryder LP, Ralfkiaer E. FOXP3+ regulatory T cells in cutaneous T-cell lymphomas: association with disease stage and survival. Leukemia. 2007;21:2512–2518. doi: 10.1038/sj.leu.2404913. [DOI] [PubMed] [Google Scholar]
  • 36.Marzec M, Halasa K, Kasprzycka M, Wysocka M, Liu X, Tobias JW, Baldwin D, Zhang Q, Odum N, Rook AH, Wasik MA. Differential effects of IL-2 and IL-15 vs. IL-21 on malignant CD4+ T lymphocytes. Cancer Res. 2008 doi: 10.1158/0008-5472.CAN-07-2403. in press. [DOI] [PubMed] [Google Scholar]
  • 37.Marzec M, Liu X, Kasprzycka M, Witkiewicz A, Raghunath PN, El-Salem M, Robertson E, Odum N, Wasik MA. IL-2- and IL-15-induced activation of the rapamycin-sensitive mTORC1 pathway in malignant CD4+ T lymphocytes. Blood. 2007;111:2181–2189. doi: 10.1182/blood-2007-06-095182. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Parrish-Novak J, Dillon SR, Nelson A, Hammond A, Sprecher C, Gross JA, Johnston J, Madden K, Xu W, West J, Schrader S, Burkhead S, Heipel M, Brandt C, Kuijper JL, Kramer J, Conklin D, Presnell SR, Berry J, Shiota F, Bort S, Hambly K, Mudri S, Clegg C, Moore M, Grant FJ, Lofton-Day C, Gilbert T, Rayond F, Ching A, Yao L, Smith D, Webster P, Whitmore Y, Maurer M, Kaushansky K, Holly RD, Foster D. Interleukin 21 and its receptor are involved in NK cell expansion and regulation of lymphocyte function. Nature. 2000;408:57. doi: 10.1038/35040504. [DOI] [PubMed] [Google Scholar]
  • 39.Moroz A, Eppolito C, Li Q, Tao J, Clegg CH, Shrikant PA. IL-21 enhances and sustains CD8+ T cell responses to achieve durable tumor immunity: comparative evaluation of IL-2, IL-15, and IL-21. J Immunol. 2004;173:900. doi: 10.4049/jimmunol.173.2.900. [DOI] [PubMed] [Google Scholar]
  • 40.Zeng R, Spolski R, Finkelstein SE, Oh S, Kovanen PE, Hinrichs CS, Pise-Masison CA, Radonovich MF, Brady JN, Restifo NP, Berzofsky JA, Leonard WJ. Synergy of IL-21 and IL-15 in regulating CD8+ T cell expansion and function. J Exp Med. 2005;201:139. doi: 10.1084/jem.20041057. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Yoon JS, Newton SM, Wysocka M, Troxel AB, Hess SD, Richardson SK, Lin JH, Benoit BM, Kasprzycka M, Wasik MA, Rook AH. IL-21 Enhances Antitumor Responses Without Stimulating Proliferation of Malignant T-cells of Patients With Cutaneous T-cell Lymphoma. J Invest Derm. 2008;128:473–480. doi: 10.1038/sj.jid.5701027. [DOI] [PubMed] [Google Scholar]

RESOURCES