Tbet-positive regulatory T cells accumulate in oropharyngeal cancers with ongoing tumor-specific type 1 T cell responses
Regulatory T cells (Tregs) may comprise different subsets allowing them to efficiently suppress different types of effector T cells. In this study, we show that high numbers of both conventional and Tbet co-expressing Foxp3hi Tregs accumulate in human papilloma virus (HPV)-driven oropharyngeal squamous cell carcinoma (OPSCC). The infiltration of Tbet+ Foxp3+ Tregs was strongly correlated with a concomitant tumor-specific and conventional type 1-oriented intratumoral T cell infiltrate. Both conventional CD4+CD25+CD127–Foxp3hi Tregs and their Tbethi counterparts exhibited an activated phenotype, co-expressed high levels of CTLA4 and Helios and exhibited a maximally demethylated Foxp3 gene locus TSDR, indicating their full capacity to impede a type 1 effector T cell response. Interestingly, while the prognostic value of conventional Tregs was neutral, a high intratumoral frequency of Tbet+ Tregs was associated with prolonged disease-specific survival, most likely because their presence reflected high numbers of effector T cells. The presence of these Tbet+ Tregs may in part explain why a dense type 1-oriented immune infiltrate in OPSCC is not enough to fully control tumor growth.
KeywordsTumor microenvironment Type 1 immunity HPV Head and neck cancer Foxp3
Foxp3+ regulatory T cells (Tregs) are pivotal in suppressing pathological and physiological immune responses . In cancer, Tregs are generally considered to suppress the local tumor immune response. However, different types of impact have been reported for Tregs in patients suffering from head and neck cancer, including oropharyngeal squamous cell carcinoma (OPSCC) . Some studies suggested a negative prognostic value for high frequencies of tumor-infiltrating CD4+CD25+Foxp3+ Tregs in head and neck cancer [3, 4, 5], whereas others demonstrated a non-predictive  or even positive prognostic role [7, 8, 9, 10] for these cells.
Recent studies have shown that Tregs can adopt different transcriptional profiles allowing them to regulate specific types of effector T cells . In mice, Tregs express the transcription factor STAT3 to restrain Th17 responses , IRF4 to suppress Th2 responses  and Tbet to suppress Th1 responses in persistent infections or autoimmunity [14, 15]. Foxp3+Tbet+ Tregs are phenotypically stable  and regulate Th1 inflammatory sites in vivo, thereby counterbalancing unwanted tissue destruction and immunopathology .
Since many head and neck cancers can be infiltrated with type 1-oriented T cells , we hypothesized that this would also attract type 1-oriented Tregs. Therefore, the presence and potential impact of conventional and Tbet-expressing Foxp3+ Tregs was analyzed in the context of a prospective study on the immune response in HPV-induced OPSCC  by three-color immunofluorescence staining and 12-parameter flow cytometry analysis. We show that high numbers of activated bona fide Foxp3+Tbet+ Tregs accumulate in OPSCC when a concomitant HPV-specific and a type 1-oriented T cell infiltrate is present.
Materials and methods
The authors acknowledge the reporting of Minimal Information About T-cell Assays (MIATA).
The 50 OPSCC patients were part of an observational study (P07–112) investigating immunity in patients with head and neck cancer . The 19 patients with cervical carcinoma (CxCa) that were included in this study took part in the CIRCLE 2 study (P08–197) investigating cellular immunity against anogenital lesions . All patients were included after signing informed consent. The study was conducted in accordance with the Declaration of Helsinki and approved by the local medical ethical committee of the Leiden University Medical Center (LUMC) and in agreement with the Dutch law. Human papillomavirus (HPV) typing and p16ink4a IHC staining was performed on formalin-fixed paraffin-embedded (FFPE) tumor sections as described . An overview of patient characteristics is given in Additional file 1: Table S1.
Characterization of lymphocytic infiltration was carried out with triple immunofluorescent staining in 41 OPSCC as described previously  using anti-CD8 (mouse IgG2b, clone 4B11; Novocastra, 1:400), anti-Tbet (rabbit polyclonal, clone H210; Santa Cruz 1:400) and anti-Foxp3 (mouse IgG1, clone 236A/E7; Abcam, 1:200), goat-anti-mouse IgG2b Alexa 647, goat-anti-rabbit Alexa 546 and goat-anti-mouse IgG1 Alexa 488 (all from Molecular probes; 1:200). Based on the morphology of cancer cell nests and autofluorescence of keratinocytes the immune cells per mm2 were manually counted as intraepithelial or stromal using the LSM 5 Image Examiner software (average of five images at a 250× magnification).
Blood and tumor cell isolation and culturing
Venous blood samples were drawn prior to surgery, and peripheral blood mononuclear cells (PBMC) were isolated as described previously . OPSCC tumors were obtained and handled as described previously [16, 17]. For two OPSCC tumors (H202, H206), the approach to prepare single cell suspensions was adjusted. The DNAse I/collagenase D enzymes were replaced by 0.38 mg/ml of the commercially available Liberase enzymes (Liberase TL, research grade, Roche), the incubation period was reduced to 15 min, and the GentleMACS dissociator was no longer used.
Tumor-specific T cell reactivity analysis
Cultured tumor infiltrating lymphocytes (TILs) were analyzed for the presence of HPV16-specific T cells using a 5-day [3H]-thymidine-based proliferation assay and antigen-specific cytokine production assay as described previously [16, 18]. To this end, T cell responses against autologous HPV16 E6/E7 synthetic long peptide (SLP; 22-mers with 14 amino acids overlap) loaded monocytes were tested in triplicate. PHA (0.5 μg/ml; HA16 Remel; ThermoFischer Scientific) was taken along as positive control, while unloaded monocytes served as negative control. At day 4 supernatant (50 μl/well) was harvested to determine cytokine production. During the last 16 h of culture, 0.5 μCi/well of [3H]thymidine was added to measure proliferation. The stimulation index was calculated as the average of test wells divided by the average of the medium control wells. A positive response was defined as a stimulation index of at least 3.
Antigen-specific cytokine production was determined by cytometric bead array (CBA, Th1/Th2 kit, BD Bioscience, Breda, the Netherlands) according to the manufacturer’s instructions. The cutoff value for cytokine production was 20 pg/ml, except for IFNγ for which it was 100 pg/ml. Positive cytokine production was defined as at least twice above that of the unstimulated cells. An example of such an HPV16-specific T cell reactivity test is depicted in Additional file 2: Figure S1.
Flow cytometry and cell sorting
Cryopreserved PBMC and/or single cell tumor samples were thawed and Treg subsets were assessed by flow cytometry as described before . Antibodies and kits used were V500-labeled CD3 (clone UCHT1, BD Biosciences), AF700- or PE-Cy7-labeled CD4 (clone RPA-T4 and SK3 respectively, BD Biosciences), PE-CY7- or BV421-labeled CD25 (clone 2A3, BD Biosciences), BV650-labeled CD127 (clone A019D5, Biolegend), APC-H7-labeled CD45RA (clone HI100, BD Biosciences), PerCP-Cy5.5-labeled CD8 (clone SK1, BD Biosciences), PE-CF594- or Alexa Fluor 647-labeled Foxp3 (clone 259D/C7, BD Biosciences), BV421-labeled CTLA-4 (clone BNI3, BD Biosciences), FITC-labeled Ki67 (clone 20Raj1; eBiosciences), APC-labeled Helios (clone 22F6, Biolegend), PE-labeled Tbet (clone ebio4B10; eBiosciences), FITC-labeled CD14 (clone M5E2; BD Biosciences), LIVE-DEAD® Fixable yellow or near-IR dead cell stain kit (ThermoFisher Scientific), and the BD Transcription Factor Buffer set. Acquisition of cells was performed on a BD LSR Fortessa and flow cytometric sorting was done using a BD FACS Aria II. Data was analysed using DIVA software (version 8.02; BD Biosciences).
DNA isolation, bisulfite conversion and FOXP3 TSDR methylation assay
Sorted Foxp3– Tconv, Foxp3+Tbet+ Treg and Foxp3+Tbet– Treg cells (ranging between 240 and 35,000 cells) were digested with Proteinase K for 4 h at 50 °C to obtain genomic DNA, followed by Bisulfite Conversion using the EZ DNA Methylation-Direct Kit (Zymo Research) according to the manufacturer. The Foxp3 Treg specific demethylation region (TSDR) was amplified (Foxp3 TSDR Forward 5′-TGG GGG TAG AGG ATT TAG AGG-3′, Foxp3 TSDR Reverse 5’-TCA CCC CAC CTA AAC CAA A C- 3′ (adapted from ) using platinum high-fidelity Taq (Invitrogen, Thermo Fisher Scientific). PCR products were cloned using a TOPO-TA cloning kit (Invitrogen, Thermo Fisher Scientific) . Plasmid DNA was purified using the NucleoSpin Plasmid kit (Macherey-Nagel). Seventeen to twenty-four clones per population per patient were sequenced using M13 primers (M13 Forward 5′-GTA AAA CGA CGG CCA G-3′, M13 Reverse 5’-CAG GAA ACA GCT ATG AC-3′) and aligned to an imputed TSDR sequence  using SnapGene software (version 4.1.9, GSL Biotech LLC, snapgene.com). Methylated CpGs were identified by the presence of a C nucleotide at the CpG position whereas demethylated CpGs were identified by a T nucleotide. The percentage of clones showing methylation at each individual CpG as well as the total average of all CpGs per population per patient was determined.
Non-parametric Wilcoxon signed-rank or Mann–Whitney tests and parametric paired or unpaired t tests were performed when appropriate. Correlation analysis were done the using Pearson’s correlation test. For survival analysis, patients were grouped into two groups according to the median (i.e., grouped into below or above the median of the total group for each parameter), after which survival was tested using Kaplan–Meier method, and statistical significance of the survival distribution was analyzed by log-rank testing. All statistical tests were performed at the 0.05 significance level, and differences were considered significant when p < 0.05, as indicated with an asterisk (*p < 0.05, **p < 0.01, *** p < 0.001 and **** p < 0.0001). Statistical analyses were performed using GraphPad Prism 7.1 (San Diego, USA).
High numbers of Foxp3+Tbet+ Treg cells are found in the OPSCC microenvironment
The numbers of tumor-infiltrating CD8–Foxp3+Tbet+ Treg cells correlate directly to a strong infiltration with type 1-oriented CD4+ and CD8+ effector T cells
The presence of intratumoral Tbet+ T cells suggests, but does not demonstrate, the presence of tumor-specific IFNγ-producing T cells that can stimulate the accumulation of CD8–Foxp3+Tbet+ Tregs. Hence, we analyzed their relation with the detection of HPV16 E6 and E7 oncoprotein-specific IFNγ and TNFα producing CD4+ and CD8+ T cells in these tumors . The OPSCC patients were divided into two groups based on the presence (+) or absence (−) of an HPV16-specific T cell immune response (IR) among the TIL cultured from these tumors (Additional file 1: Table S1; Additional file 2: Figure S1). This revealed that OPSCC containing HPV-specific T cells also displayed higher numbers of all the T cell populations, and in particular also higher numbers of CD8–Foxp3+Tbet+ Tregs (Fig. 2b). Notably, the correlation between the number of CD8–Foxp3+Tbet+ Tregs and CD8–Foxp3–Tbet+ (CD4) T cells or CD8+Foxp3–Tbet+ T cells is retained in both the IR- and IR+ groups, however, in IR- patients the number of these 3 cell types is lower than in the IR+ group (Additional file 2: Figure S3).
Tbet– And Tbet+ CD4+CD25+CD127–Foxp3hi Tregs are bonafide activated Tregs
To assess if the detected Foxp3hiTbethi T cells are truly Tregs, we sorted Foxp3hiTbethi and Foxp3hiTbet– Treg populations by flowcytometry from HPV16+ OPSCC, of which we knew that they contained Tbethi T cells and HPV16-specific T cells as well as for which freshly digested tumor material was still available (Additional file 2: Figure S7). Previously it has been shown that the TSDR, a conserved region in intron 1 of FOXP3, is strictly hypomethylated in Tregs whereas this region is hypermethylated in conventional T cells and T cell clones, even after activation [20, 23] and, therefore, the TSDR methylation status functions as a true Treg specific parameter . Indeed, epigenetic analysis revealed that Foxp3– conventional T cells (Tconv) exhibited maximal TDSR methylation whereas this region was strongly hypomethylated not only in the conventional Foxp3hiTbet– cells (> 95% for male and 50% for female donor due to Foxp3 methylation on the inactive X-chromosome for all 15 different CpG islands in this region) but also in the Foxp3hiTbethi cells (Fig. 3d), indicating that Foxp3hiTbethi and Foxp3hiTbet– Tregs present in these tumors indeed are bona fide Tregs.
Thus, Foxp3hiTbethi cells accumulate specifically in tumors with an ongoing type 1 immune response and based on the hypomethylation of all 15 CpG islands in the TDSR region of the Foxp3 intron 1 in these T cells sorted from 3 different patients, are true regulatory T cells.
Last but not least, we analyzed the expression of Tbet in CD4+Foxp3hi regulatory T cell clone 148.31, isolated from an HPV16+ cervical cancer patient with an ongoing local type 1 immune response, and for which we have shown its suppressive capacity in several different in vitro assays . This clone not only was suppressive but in addition to Foxp3 also expressed the transcription factor Tbet (Fig. 4c, d).
High numbers of tumor-infiltrating Foxp3hiTbethi Tregs, but not conventional Tregs, correlate with improved survival
In mice, Foxp3+Tbet+ Tregs were shown to play a role in suppressing Th1-like effector T cell responses [14, 15]. We previously showed that a dense type 1-oriented immune infiltrate is present in a large majority of patients with HPV16-driven oropharyngeal and cervical cancers [16, 25] and wondered if these tumors are infiltrated with a similar type of Treg. Not only did we detect high numbers of CD4+ T cells co-expressing both the transcription factors Foxp3 and Tbet in these HPV16-driven tumors but we also found that their numbers are directly related to the number of tumor-infiltrating type 1-oriented (Foxp3-) effector T cells. Due to the inability to isolate viable Foxp3+Tbet+ and Foxp3+Tbet– Tregs by flowcytometry using antibodies against Foxp3 and Tbet, these subtypes of Tregs could not be tested in a functional T cell suppression assay. Therefore, we used the well accepted surrogate for functionality by assessing the methylation status of the FoxP3 gene locus TSDR. Hypomethylation of this locus is generally considered the gold standard for identifying functional Tregs [20, 23]. This method is widely used to distinguish true Tregs from other T cell populations which transiently upregulate FOXP3 after their activation [24, 27] and it showed that the CD4+Foxp3hiTbethi cells found in HPV-driven cancers represent bona fide Tregs. The fact that the in vitro proven suppressive CD4+Foxp3hi Treg clone 148.31, isolated from a patient with an ongoing HPV-specific type 1 T cell response , also co-expresses Tbet corroborates our findings.
The detection of conventional and Tbet+ Treg populations in OPSCC and CxCa confirmed studies reporting the presence of CD4+Foxp3+ T cells co-expressing Tbet in chronic hepatitis C driven liver cancer, oral squamous cell carcinoma and ovarian cancer [28, 29, 30]. Importantly, we found that Foxp3hiTbethi Tregs specifically dominated in those cancers where an ongoing HPV-specific type 1 T cell response could be detected. In mice, it seems that Tbet+ Tregs differentiate de novo from their Tbet– counterparts under situations that drive type 1 immunity. Elimination of these Tbet+ Tregs results in severe Th1 autoimmunity  indicating that their purpose is to keep a type 1-oriented immune response at bay to prevent immune pathology. In human tumors, they are expected to do the same. Our results on the prognostic value of the two different populations for patient survival, therefore, are a bit counterintuitive. The observations that conventional Tregs more often were found in the tumor stroma and correlated less with total effector cell infiltration may explain why they had no impact on survival. In contrast, the Tbethi Tregs do infiltrate the tumor cell nests and their number strongly correlated with that of tumor-infiltrating Tbet+ CD4+ and CD8+ effector T cells as well as with the detection of ongoing HPV-specific type 1 T cell responses. Yet unexpectedly, their infiltration has a strong favorable prognostic value for survival. It should, however, be realized that the number of Tbet+ non-regulatory T cells in these tumors is much higher and the presence of these type 1 effector T cells is associated with improved survival in OPSCC (Fig. 5 and reference ). Hence, it is more likely that the association between high numbers of true regulatory Tbet+ T cells, based on the hypomethylated TSDR and improved survival is a reflection of the presence and clinical impact of a strong ongoing and protective type 1 tumor-specific immune response, than that these Tbet+ Tregs would have a positive contribution to the antitumor response. This view fits with previous studies showing no [6, 7] or a positive [8, 10] prognostic effect of Foxp3+ Tregs in head and neck cancer, as here the number of co-infiltrating (activated) CD4+ and CD8+ T cells is also much higher than that of Tregs. In those situations that a negative impact of Tregs was found, the tumors were scarcely infiltrated with T cells [3, 4]. Thus, Foxp3+Tbet+ Treg cells accumulate at the OPSCC tumor sites where they probably exert a suppressive effect that is strong enough to impede full type 1 effector T cell-mediated control of the tumor (i.e. spontaneous tumor regression). However, given the observed survival benefit for patients with an HPV-specific and type 1-oriented intratumoral T cell infiltrate, their numbers or activity may be insufficient to fully suppress the ongoing clinically beneficial anti-tumor response after standard therapy.
This study was financially supported by grants from the Dutch Cancer Society 2014–6696 to SHvdB, LAvdV and MJPW and 2016–10726 to SHvdB, MJPW and SJAMS.
Availability of data and materials
All data generated or analyzed during this study are included in this published article [and its supplementary information files].
Conception and design: SJS, MJPW, SHvdB. Development of methodology: SJS, ESJ, MJPW, SHvdB. Acquisition of data: SJS, CLD, ESJ, VJvH, IE, MJPW. Analysis and interpretation of data: SJS, CLD, ESJ, VJvH, IE, SLvE, MJPW, SHvdB. Writing, review, and/or revision of the manuscript: SJS, CLD, ESJ, MJPW, SHvdB. Patient management: SLvE. Study supervision: MJPW, SHvdB. All authors read and approved the final manuscript.
Ethics approval and consent to participate
Approved under Institutional Review Board protocol P07–112 and P08–197. These studies were conducted in accordance with the Declaration of Helsinki and approved by the local medical ethical committee of the Leiden University Medical Center (LUMC) and in agreement with the Dutch law.
Consent for publication
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
- 5.Watanabe Y, Katou F, Ohtani H, Nakayama T, Yoshie O, Hashimoto K. Tumor-infiltrating lymphocytes, particularly the balance between CD8(+) T cells and CCR4(+) regulatory T cells, affect the survival of patients with oral squamous cell carcinoma. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2010;109(5):744–52.CrossRefGoogle Scholar
- 9.Lukesova E, Boucek J, Rotnaglova E, Salakova M, Koslabova E, Grega M, et al. High level of Tregs is a positive prognostic marker in patients with HPV-positive oral and oropharyngeal squamous cell carcinomas. Biomed Res Int. 2014;2014(303929).Google Scholar
- 17.Piersma SJ, Welters MJ, van der Hulst JM, Kloth JN, Kwappenberg KM, Trimbos BJ, et al. Human papilloma virus specific T cells infiltrating cervical cancer and draining lymph nodes show remarkably frequent use of HLA-DQ and -DP as a restriction element. Int J Cancer. 2008;122(3):486–94.CrossRefGoogle Scholar
- 18.Santegoets SJ, van Ham VJ, Ehsan I, Charoentong P, Duurland CL, van Unen V, et al. The anatomical location shapes the immune infiltrate in tumors of same etiology and affects survival. Clin Cancer Res. 2018.Google Scholar
- 19.Santegoets SJ, Dijkgraaf EM, Battaglia A, Beckhove P, Britten CM, Gallimore A, et al. Monitoring regulatory T cells in clinical samples: consensus on an essential marker set and gating strategy for regulatory T cell analysis by flow cytometry. Cancer Immunol Immunother. 2015;64(10):1271–86.CrossRefGoogle Scholar
- 22.Gobert M, Treilleux I, Bendriss-Vermare N, Bachelot T, Goddard-Leon S, Arfi V, et al. Regulatory T cells recruited through CCL22/CCR4 are selectively activated in lymphoid infiltrates surrounding primary breast tumors and lead to an adverse clinical outcome. Cancer Res. 2009;69(5):2000–9.CrossRefGoogle Scholar
- 25.Gorter A, Prins F, van Diepen M, Punt S, van der Burg SH. The tumor area occupied by Tbet+ cells in deeply invading cervical cancer predicts clinical outcome. J Transl Med. 2015;13(295).Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.