SEARCH

SEARCH BY CITATION

Keywords:

  • regulatory T cells;
  • melanoma;
  • FOXP3;
  • human Treg marker;
  • cancer vaccine

Abstract

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

The human suppressive T cells that stably express transcription factor FOXP3, or regulatory T cells (Tregs), are thought to suppress antitumor immune responses. The most specific marker for human Tregs is the demethylation of CpG dinucleotides located in the first intron of FOXP3 (FOXP3i1). FOXP3i1 is completely methylated in other hematopoietic cells, including nonsuppressive T cells that transiently express FOXP3 after activation. Previously, we and others reported estimations of the frequency of Tregs in the blood of melanoma patients using a FOXP3i1 methylation-specific qPCR assay. Here, we attempted to quantify Tregs inside tumor samples using this assay. However, we found demethylated FOXP3i1 sequences in the melanoma cells themselves. This demethylation was not associated with substantial FOXP3 mRNA or protein expression, even though the demethylation extended to the promoter and terminal regions of the gene in some melanoma cells. Our results imply that analyzing Treg frequencies by quantification of demethylated FOXP3i1 will require that tumor-infiltrating T cells be separated from melanoma cells.

Regulatory T cells (Tregs) are a subset of CD4+CD25+ T lymphocytes that inhibit immune responses. Their differentiation and function depends on transcription factor FOXP3; individuals carrying a FOXP3 mutation have a Treg deficiency and severe autoimmune syndrome.1, 2 The physiological role of Tregs is to prevent autoimmune pathology through suppression of self-reactive T cells. In cancer, excessive Treg function or numbers may decrease spontaneous or vaccine-induced tumor-specific T cell responses. While there is ample evidence of such a deleterious role for Tregs in murine tumors,3 it is much less clear in cancer patients, because of the lack of a good marker for human Tregs. High proportions of CD4+ T lymphocytes expressing CD25 or FOXP3 were observed in the blood or in tumors of patients with various types of cancer, sometimes associated with a poor prognosis.4 However, in humans, neither CD25 nor FOXP3 are specific Treg markers: CD25 is present on all activated T cells, and FOXP3 is transiently expressed in nonsuppressive CD4+ T cells, notably after T cell receptor (TCR) or transforming growth factor-β (TGF-β) stimulation.5–9 This lack of specificity is particularly problematic inside tumors for two reasons: First, many tumor-infiltrating T cells recognize tumor antigens10; hence, they may have been recently stimulated through their TCRs. Second, TGF-β secretion and activity are important in many tumors, in line with the tumor-promoting role of TGF-β.11

Recently, it was shown that a conserved region in the first intron of the FOXP3 gene is demethylated in human Treg cells and clones, while it is completely methylated in other human blood cells, including the nonsuppressive T cells that express FOXP3.6, 7, 9, 12 This region, which we call FOXP3i1, but which is also referred to as treg-specific demethylated region (TSDR)12, 13 or CNS2,14 is required to maintain stable Foxp3 expression in murine Tregs.6, 7, 12–15 We and others developed a methylation-specific qPCR assay (MS-qPCR) to quantify the proportion of demethylated FOXP3i1 sequences in blood samples, as a surrogate for Treg frequency.9, 15 Using this assay, we previously showed that none of the three medications that we tested, which were expected to deplete blood Tregs, actually did so in a majority of the treated cancer patients.9 Here, we wished to quantify Tregs directly in melanoma samples, to examine whether a correlation exists between elevated proportions of tumor-infiltrating Tregs and prognosis or clinical response to anticancer vaccines.

Material and Methods

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

MS-qPCR for FOXP3i1

Genomic DNA (gDNA) was prepared from frozen cell pellets or from pools of two or three sections (15–25 μm thick) of frozen melanoma biopsies with the PureLink Genomic DNA Mini Kit (Invitrogen, UK). One to two micrograms of EcoRI-digested gDNA was treated with sodium bisulfite using the MethylCode Bisulfite Conversion kit (Invitrogen, UK). Real-time PCR amplification of methylated and demethylated FOXP3i1 sequences were performed on 125–350 ng of bisulfite-converted DNA as previously described (primers are listed in Supporting Information Table S1). The proportion of cells with demethylated FOXP3i1 was calculated as follows: (number of demethylated FOXP3i1 sequences/(number of demethylated FOXPi1 sequences + number of methylated FOXP3i1 sequences)) × number of X chromosomes per cell. The calculated proportions are approximate, because the MS-qPCR assay detects only sequences that are either completely demethylated or completely methylated in the FOXP3i1 region.

Immunohistochemistry

Cryosections (7-μm thick) adjacent to those used for FOXP3i1 MS-qPCR were fixed in 4% formalin. Staining with mouse anti-human CD3 antibody (BD Biosciences, USA; clone UCHT1, 1/50) was performed after inhibition of endogenous peroxidases, and detected with the EnVision+ System-horseradish peroxidase (HRP) (Dako, Denmark) and 3-amino-9-ethylcarbazole (AEC). Staining with mouse anti-human FOXP3 antibody (eBioscience, USA; clone 236A/E7, 1/200) was performed after inhibition of endogenous peroxidases and biotins, and followed by biotinylated goat anti-mouse antibody (6 μg/ml, Vector Laboratories, UK), and streptavidin-HRP (R&D Systems, UK). Color detection was performed with AEC. All sections were counterstained with hematoxylin (Sigma-Aldrich, USA). Images were acquired with a Zeiss Mirax Midi digital microscope and analyzed with Mirax Viewer software (Carl Zeiss, Germany). Proportions of FOXP3+ cells were estimated by manually counting FOXP3+ cells and total nucleated cells, in 8 enlarged views per sample. A total of 7,902 to 23,928 nucleated cells were counted per sample.

RT-qPCR for FOXP3 expression

Total RNA prepared with the Invisorb Spin Cell RNA Mini Kit (STRATEC Molecular GmbH, Germany) was reverse transcribed with MMLV reverse transcriptase (Invitrogen, UK). Real-time PCR amplifications were performed in a final volume of 25 μl with 0.6 U of HotGoldStar DNA polymerase (Eurogentec, Belgium), 300 nM each primer, 100 nM probe, 200 μM dNTP and 5 mM MgCl2 in an ABI Prism 7700 Sequence Detector (Applied Biosystems, USA) under standard conditions: 94°C for 10 min, 45 cycles of 94°C for 15 sec and 60°C for 1 min. Sequences of primers for FOXP3 and EF-1, a housekeeping gene, are indicated in Supporting Information Table S1.

Bisulfite sequencing

Bisulfite-treated genomic DNA (30–100 ng) prepared as described above underwent two rounds of 40 PCR cycles with nested sets of primers (Supporting Information Table S1). Products from the second round of PCR amplification were cloned using the TOPO TA Cloning kit (Invitrogen, UK) before sequencing of individual plasmids.

Treatment of melanoma lines with TGF-β

Cells were plated in complete medium and left to adhere for 24–72 hr. Medium was replaced with serum-free X-VIVO-10 (Lonza) containing recombinant human TGF-β1 or TGF-β2 (R&D Systems, UK) at a final concentration of 4 ng/ml. Cells were incubated for an additional 72 or 24 hr, respectively, then collected and analyzed by reverse transcription qPCR (RT-qPCR) for FOXP3 expression and by Western blot with antibodies against phosphoSMAD2, total SMAD2 (Cell Signaling Technologies, USA) and β actin (Sigma-Aldrich, USA).

Results

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Human melanoma samples contain high proportions of cells with demethylated FOXP3i1 but not a high proportion of FOXP3+ cells

Genomic DNA was purified from two or three adjacent sections of frozen melanoma skin metastases, then treated with sodium bisulfite, which converts demethylated cytosines into uracils but leaves methylated cytosines unchanged. Bisulfite-treated DNA was amplified by real-time PCR with primers and probes specific to the methylated and demethylated forms of FOXP3i1. In 13 melanoma samples, the proportions of cells with demethylated FOXP3i1 ranged from 0.6 to 39.3%; the latter value occurred in melanoma LB2763 and was more than 10-fold higher than the value measured in a healthy lymph node (Table 1). Immunohistochemical analyses indicated that in melanoma LB2763, FOXP3+ cells were not more abundant than in the control lymph node, and represented <3% of all cells in the section (Fig. 1). Overall, the proportion of FOXP3+ cells was inferior to that of cells with demethylated FOXP3i1 in the six melanoma metastases tested, in contrast to what was observed in the control lymph node. We concluded that in melanomas, cells other than Tregs carried demethylated FOXP3i1 sequences but did not contain detectable FOXP3 protein.

thumbnail image

Figure 1. Immunohistochemical detection of CD3+ and FOXP3+ cells. Serial cryosections of a subcutaneous metastasis from patient LB2763 and a normal lymph node were stained with anti-CD3 and anti-FOXP3 antibodies (in red). Enlarged views correspond to the black squares in the tissue sections.

Download figure to PowerPoint

Table 1. Demethylation of FOXP3i1 and expression of FOXP3 mRNA or protein in human tissue samples and cell lines
inline image

Human melanoma cell lines carry demethylated FOXP3i1 but do not express high levels of FOXP3 mRNA

To test whether demethylated FOXP3i1 sequences are present in melanoma cells, we performed MS-qPCR on 15 melanoma cell lines (Table 1). Six lines were derived from patients from whom we had analyzed a tumor sample. In three of the 15 cell lines, more than 35% of the cells contained demethylated FOXP3i1, including the melanoma line derived from patient LB2763, in which 77% of the cells contained demethylated FOXP3i1. This strongly suggested that the high proportion of cells with demethylated FOXP3i1 in melanoma fragment sample LB2763 was due to the presence of melanoma cells with demethylated FOXP3i1.

In Tregs, demethylation of FOXP3i1 is associated with high and stable FOXP3 mRNA and protein expression. We therefore tested whether melanoma cells with demethylated FOXP3i1 expressed FOXP3 (Table 1). In contrast to a Treg clone, all of the melanoma lines expressed very low levels of FOXP3 mRNA, similar to those detected in T helper (Th) or cytotoxic T cell (CTL) clones, which were used as negative controls. Therefore, demethylation of FOXP3i1 in melanoma cells was not associated with high FOXP3 expression.

FOXP3i1 is not demethylated in a few colorectal or pulmonary carcinoma cell lines tested

Quantification of demethylated FOXP3i1 sequences has been used by others to measure Treg frequency in colorectal or lung carcinoma fragments, but the possibility that colorectal or lung cancer cells could themselves carry demethylated FOXP3i1 was not examined previously.15 We analyzed six lung and seven colorectal carcinoma cell lines with FOXP3i1 MS-qPCR (Table 1). None contained substantial proportions of cells with demethylated FOXP3i1.

In some melanoma cells, FOXP3 demethylation extends to the promoter and terminal regions of the gene

We performed bisulfite sequencing to confirm the FOXP3i1 demethylation observed in melanoma cells and to extend our analysis to other regions of the gene. Briefly, regions of the promoter, first intron, and terminal part of the FOXP3 gene were amplified by PCR from bisulfite-treated gDNA, then cloned and sequenced to determine the methylation status of individual CpG dinucleotides, as previously described.6, 12

In the FOXP3i1 region analyzed by MS-qPCR, melanoma lines LB2763 and LB2259 showed 45% and 36% demethylated CpGs, respectively (Fig. 2). Because both cell lines were derived from female patients, these proportions corresponded to ± 90% and 72% of cells with a demethylated FOXP3i1 allele, respectively; this is because FOXP3 is located on chromosome X, a copy of which is inactivated by methylation in female cells. Bisulfite sequencing therefore confirmed the MS-qPCR results for melanoma lines LB2763-MEL and LB2259-MEL (Table 1 and Supporting Information Table S2). Sequencing and MS-qPCR results were also concordant for two other melanoma lines (KUL68-MEL and LB2273-MEL), with low percentages of FOXP3i1 demethylation. Finally, the proportion of demethylated FOXP3i1 CpGs amounted to 12% in a noncancerous melanocyte cell line, 7% in a control Th clone, and 79% in a control Treg clone.

thumbnail image

Figure 2. Bisulfite sequencing of three regions of the FOXP3 gene in various human cell lines. Schematic representation of FOXP3 with introns represented as lines and exons as boxes (empty boxes: untranslated regions). Regions amplified by PCR from bisulfite-treated gDNA are magnified to show individual CpG dinucleotides as vertical bars. Numbers indicate positions relative to the first nucleotide of exon 1. CpGs analyzed in the FOXP3i1 MS-qPCR assay are indicated by brackets. PCR products obtained from the indicated cell lines were cloned and several bacterial clones were sequenced. Each line of squares represents a single bacterial clone (i.e., a single DNA molecule). Methylated CpGs are represented by filled squares and demethylated CpGs by open squares. The overall percentage of demethylated CpGs within the indicated regions is shown for each cell line.

Download figure to PowerPoint

In the promoter region of FOXP3, proportions of demethylated CpGs ranged from 4% to 42% in the melanoma lines, with the highest proportion found in LB2763, 26% and 61% in the control Th and Treg clones, respectively, and 11% in the noncancerous melanocyte cell line (Fig. 2).

In the terminal region of FOXP3, a high proportion (50%) of demethylated CpGs was observed in melanoma LB2763. Proportions were 29% in the control Treg clone, but only 9–12% in all of the other cell lines (Fig. 2).

In conclusion, the methylation status of gene FOXP3 in melanoma LB2763 was similar to that found in Tregs. Most cells carried a FOXP3 allele that was completely demethylated in the first intronic and terminal regions, and highly demethylated in the promoter region (Fig. 2).

Activation of SMAD transcription factors is not sufficient to induce FOXP3 expression in melanoma cells carrying a highly demethylated FOXP3 allele

Melanoma cells LB2763 and LB2259 carried a highly demethylated FOXP3 gene but did not express this gene, which could be due to the lack of appropriate transcription factors in their nuclei. To test this hypothesis with candidate transcription factors, we treated melanoma lines LB2763 and LB2259 with TGF-β1 or TGF-β2, cytokines known to induce FOXP3 expression in human TCR-stimulated T cells or pancreatic carcinoma lines, respectively.8, 16, 17 TGF-β1 and TGF-β2 transduce their signal through phosphorylation and activation of SMAD transcription factors. Activating phosphorylation of SMAD2 was readily observed in TGF-β1- or TGF-β2-treated melanoma cells (Supporting Information Fig. S1), but gene FOXP3 remained silent (Table 1).

Discussion

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Of the blood cells, only Tregs carry demethylated FOXP3i1 sequences.6, 9, 12, 15 The quantification of this stable epigenetic mark was used previously to measure Tregs in the blood of cancer or transplanted patients.9, 15 We show here that in contrast to blood, demethylated FOXP3i1 is not specific to Tregs in the melanoma tumor environment. In 2 out of 15 melanoma lines, >50% of the cells carried demethylated FOXP3i1, as did 15% and 40% of the cells in the respective corresponding melanoma sections. This observation precludes the use of FOXP3i1 MS-qPCR to quantify Tregs in unmanipulated melanoma samples. Nevertheless, separation of the tumor-infiltrating T cells from the melanoma cells may be followed by FOXP3i1 MS-qPCR for accurate quantification; i.e., quantification that is not confounded by the detection of activated CD4+ T cells, as is the case with anti-FOXP3 or anti-CD25 staining.

This caveat to the quantification of tumor Tregs by FOXP3i1 MS-qPCR may apply only to melanoma. A comparable assay was previously used by others to measure Tregs in samples of lung and colorectal carcinoma,15, 18, 19 and we could not detect substantial proportions of cells with demethylated FOXP3i1 in six lung and seven colorectal carcinoma lines. However, the number of tested lines should be increased before the Treg specificity of the FOXP3i1 MS-qPCR assay can be considered fully validated for nonmelanoma tumors.

In contrast to what is observed in T cells, FOXP3 demethylation does not correlate with expression in melanoma cells. In Tregs, FOXP3i1 demethylation is thought to contribute to the stability of FOXP3 expression and to the maintenance of the suppressive phenotype.14 In melanoma, however, FOXP3 demethylation may have no functional role. Alterations in patterns of genomic methylation are seen in virtually all human tumors. Regions of DNA hypomethylation and hypermethylation coexist in tumor cells, but loss of DNA methylation is predominant.20 Genomic hypomethylation in tumors is associated with the transcriptional activation of a very limited number of genes, including cancer-germline (CG) genes such as those of the MAGE family, that rely on DNA methylation for silencing in normal somatic tissues.20, 21 Unlike CG genes, demethylation of FOXP3 in melanoma cells was not sufficient to activate its expression. Activation of SMAD transcription factors in melanoma cells with a highly demethylated FOXP3 gene was not sufficient either, suggesting that other transcription factors that are present in Tregs but absent in melanoma cells are required for FOXP3 expression.

In conclusion, our study reveals that prior isolation of tumor-infiltrating T cells will be required to measure Tregs in melanoma samples through quantification of FOXP3i1 demethylated sequences. Notwithstanding this limitation, demethylated FOXP3i1 thus far remains the most specific marker of human Tregs.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

The authors thank Amandine Nizet and Marjorie Mercier for excellent technical assistance.

References

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information
  • 1
    Zheng Y, Rudensky AY. Foxp3 in control of the regulatory T cell lineage. Nat Immunol 2007; 8: 45762.
  • 2
    Sakaguchi S, Miyara M, Costantino CM, Hafler DA. FOXP3+ regulatory T cells in the human immune system. Nat Rev Immunol 2010; 10: 490500.
  • 3
    Colombo MP, Piconese S. Regulatory-T-cell inhibition versus depletion: the right choice in cancer immunotherapy. Nat Rev Cancer 2007; 7: 8807.
  • 4
    Wang HY, Wang RF. Regulatory T cells and cancer. Curr Opin Immunol 2007; 19: 21723.
  • 5
    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 USA 2006; 103: 665964.
  • 6
    Stockis J, Fink W, Francois V, Connerotte T, de Smet C, Knoops L, van der Bruggen P, Boon T, Coulie PG, Lucas S. Comparison of stable human Treg and Th clones by transcriptional profiling. Eur J Immunol 2009; 39: 86982.
  • 7
    Francois V, Ottaviani S, Renkvist N, Stockis J, Schuler G, Thielemans K, Colau D, Marchand M, Boon T, Lucas S, van der Bruggen P. The CD4(+) T-cell response of melanoma patients to a MAGE-A3 peptide vaccine involves potential regulatory T cells. Cancer Res 2009; 69: 433545.
  • 8
    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-beta dependent but does not confer a regulatory phenotype. Blood 2007; 110: 298390.
  • 9
    de Vries IJ, Castelli C, Huygens C, Jacobs JF, Stockis J, Schuler-Thurner B, Adema GJ, Punt CJ, Rivoltini L, Schuler G, Coulie PG, Lucas S. Frequency of circulating Tregs with demethylated FOXP3 intron 1 in melanoma patients receiving tumor vaccines and potentially Treg-depleting agents. Clin Cancer Res 2011; 17: 8418.
  • 10
    Lurquin C, Lethé B, Corbière V, Théate I, van Baren N, Coulie PG, Boon T. Contrasting frequencies of anti-tumor and anti-vaccine T cells in metastases of a melanoma patient vaccinated with a MAGE tumor antigen. J Exp Med 2005; 201: 24957.
  • 11
    Levy L, Hill CS. Alterations in components of the TGF-beta superfamily signaling pathways in human cancer. Cytokine Growth Factor Rev 2006; 17: 4158.
  • 12
    Baron U, Floess S, Wieczorek G, Baumann K, Grutzkau A, Dong J, Thiel A, Boeld TJ, Hoffmann P, Edinger M, Turbachova I, Hamann A, et al. DNA demethylation in the human FOXP3 locus discriminates regulatory T cells from activated FOXP3(+) conventional T cells. Eur J Immunol 2007; 37: 237889.
  • 13
    Floess S, Freyer J, Siewert C, Baron U, Olek S, Polansky J, Schlawe K, Chang HD, Bopp T, Schmitt E, Klein-Hessling S, Serfling E, et al. Epigenetic control of the foxp3 locus in regulatory T cells. PLoS Biol 2007; 5: e38.
  • 14
    Zheng Y, Josefowicz S, Chaudhry A, Peng XP, Forbush K, Rudensky AY. Role of conserved non-coding DNA elements in the Foxp3 gene in regulatory T-cell fate. Nature 2010; 463: 80812.
  • 15
    Wieczorek G, Asemissen A, Model F, Turbachova I, Floess S, Liebenberg V, Baron U, Stauch D, Kotsch K, Pratschke J, Hamann A, Loddenkemper C, et al. Quantitative DNA methylation analysis of FOXP3 as a new method for counting regulatory T cells in peripheral blood and solid tissue. Cancer Res 2009; 69: 599608.
  • 16
    Fantini MC, Becker C, Monteleone G, Pallone F, Galle PR, Neurath MF. Cutting edge: TGF-beta induces a regulatory phenotype in CD4+CD25-T cells through Foxp3 induction and down-regulation of Smad7. J Immunol 2004; 172: 514953.
  • 17
    Hinz S, Pagerols-Raluy L, Oberg HH, Ammerpohl O, Grussel S, Sipos B, Grutzmann R, Pilarsky C, Ungefroren H, Saeger HD, Kloppel G, Kabelitz D, et al. Foxp3 expression in pancreatic carcinoma cells as a novel mechanism of immune evasion in cancer. Cancer Res 2007; 67: 834450.
  • 18
    Loddenkemper C, Hoffmann C, Stanke J, Nagorsen D, Baron U, Olek S, Huehn J, Ritz JP, Stein H, Kaufmann AM, Schneider A, Cichon G. Regulatory (FOXP3+) T cells as target for immune therapy of cervical intraepithelial neoplasia and cervical cancer. Cancer Sci 2009; 100: 11127.
  • 19
    Sehouli J, Loddenkemper C, Cornu T, Schwachula T, Hoffmuller U, Grutzkau A, Lohneis P, Dickhaus T, Grone J, Kruschewski M, Mustea A, Turbachova I, et al. Epigenetic quantification of tumor-infiltrating T-lymphocytes. Epigenetics 2011; 6: 23646.
  • 20
    De Smet C, Loriot A. DNA hypomethylation in cancer: epigenetic scars of a neoplastic journey. Epigenetics 2010; 5: 20613.
  • 21
    De Smet C, Lurquin C, Lethé B, Martelange V, Boon T. DNA methylation is the primary silencing mechanism for a set of germ line- and tumor-specific genes with a CpG-rich promoter. Mol Cell Biol 1999; 19: 732735.

Supporting Information

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
IJC_26198_sm_suppfig1.eps8898KSupporting Figure 1.
IJC_26198_sm_supptable1.doc73KSupporting Table 1.
IJC_26198_sm_supptable2.doc34KSupporting Table 2.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.