|Year : 2020 | Volume
| Issue : 2 | Page : 309-319
Targeting EZH2 depletes LMP1-induced activated regulatory T cells enhancing antitumor immunity in nasopharyngeal carcinoma
Wei Sun1, Lin Chen1, Jun Tang2, Chengcheng Zhang1, Yihui Wen1, Weiping Wen1
1 Department of Otorhinolaryngology-Head and Neck Surgery; Guangzhou Key Laboratory of Otorhinolaryngology-Head and Neck Surgery, The First Affiliated Hospital of Sun Yat-sen University, Guangzhou, China
2 Department of Otorhinolaryngology Head and Neck Surgery, The First People's Hospital of Foshan, Foshan, China
|Date of Submission||15-Nov-2019|
|Date of Decision||22-Jan-2020|
|Date of Acceptance||24-Mar-2020|
|Date of Web Publication||28-May-2020|
No. 58, Zhongshan 2nd Road, Guangzhou
No. 58, Zhongshan 2nd Road, Guangzhou
Source of Support: None, Conflict of Interest: None
Objective: Regulatory T cells (Tregs) are critical factors that impair antitumor immunity. Epstein–Barr virus (EBV)-encoded latent membrane protein 1 (LMP1) is one of the most pathogenic factors in nasopharyngeal carcinoma (NPC). However, the role of EBV-encoded LMP1 in regulating Treg generation in NPC remains unclear.
Materials and Methods: The in vitro stability of activated Tregs (aTregs) influenced by LMP1 was analyzed by flow cytometry. The inhibitory effects of LMP1-HONE1 antigen-induced aTregs on tumor-associated antigen (TAA)-specific T cells were analyzed in vitro and in vivo. Finally, the expression of LMP1, Foxp3, and enhancer of zeste homolog 2 (EZH2) were analyzed in samples from 86 NPC patients by immunohistochemistry and immunofluorescence.
Results: LMP1 upregulated the expression of EZH2, which increased the stability of aTregs in vitro. EZH2 inhibitor, DZnep, depleted LMP1-HONE1 antigen-induced aTregs in vitro and led to potent TAA-specific T cell antitumor immunity in vivo. In NPC tissues, LMP1 expression level was positively correlated with the number of EZH2+ Tregs, which was positively correlated with clinical stage and overall survival.
Conclusions: EZH2 is essential for maintaining the stability and inhibitory functions of aTregs that are induced by EBV-encoded LMP1 in NPC.
Keywords: Enhancer of zeste homolog 2, immunity, latent membrane protein 1, nasopharyngeal carcinoma, regulatory T cells
|How to cite this article:|
Sun W, Chen L, Tang J, Zhang C, Wen Y, Wen W. Targeting EZH2 depletes LMP1-induced activated regulatory T cells enhancing antitumor immunity in nasopharyngeal carcinoma. J Can Res Ther 2020;16:309-19
|How to cite this URL:|
Sun W, Chen L, Tang J, Zhang C, Wen Y, Wen W. Targeting EZH2 depletes LMP1-induced activated regulatory T cells enhancing antitumor immunity in nasopharyngeal carcinoma. J Can Res Ther [serial online] 2020 [cited 2020 Jul 16];16:309-19. Available from: http://www.cancerjournal.net/text.asp?2020/16/2/309/285205
Authors Wei Sun, Lin Chen, Jun Tang contributed equally to this work as first authors.
Authors Weiping Wen, Yihui Wen contributed equally to this work as corresponding authors.
| > Introduction|| |
Nasopharyngeal carcinoma (NPC) is a common tumor in the South and Southeast Asia, with a peak annual incidence rate of 30/100,000 persons. Approximately 46% of NPC patients have locally advanced disease on diagnosis. The 5-year survival rate of NPC has not changed significantly despite the improvements in treatments in recent years.,
Growing evidence indicates that antitumor immunosuppressive activity is an important determinant of tumor progression. Foxp3+ CD4+ regulatory T cells (Tregs) are crucial factors in host immune tolerance. Several studies have shown that an increased number of tumor-infiltrating Tregs are positively correlated with tumor progression and poor prognosis in NPC patients., We recently reported that Foxp3high CD45RA− CD4+ Treg cells (activated Treg [aTreg] cells), a functionally distinct subset of Tregs, were increased in the peripheral blood of NPC patients and significantly suppressed immune responses. In addition, our most recent study showed that preferential ablation of intratumoral aTregs using a CCR4 antagonist and GdCl3 achieved more effective antitumor immunity in vivo. Therefore, targeting aTregs may provide a powerful means to unleash more potent antitumor immune response in cancer immunotherapy.
Epstein–Barr virus (EBV) infection is identifiable in all undifferentiated NPC cases. EBV encodes a series of functional proteins and non-coding RNAs, including latent membrane protein 1 (LMP1), latent membrane protein 2, EBV nuclear antigen 1 (EBNA1), and EBV-encoded small RNAs (EBNAs)., LMP1 is one of the most pathogenic factors in NPC development. Given the viral-specific pathogenesis of NPC, it is essential to study the role of EBV-encoded LMP1 in regulating aTreg generation.
The development of small molecule antitumor agents designed to directly target the key pathways of tumor cells provides new opportunities for targeting intracellular pathways to control immune plasticity. By understanding how these agents impact immune cells, they may be used to simultaneously alter key immune cell populations to complement cancer immunotherapy. Small-molecule inhibitors of enhancer of zeste homolog 2 (EZH2), a member of the polycomb repressive complex, which represses gene expression through interaction with specific chromatin marker(s) that associate with the promoter region of target genes, are used as direct antitumor agents in clinical trials. However, their potential ability to disrupt aTregs and to promote antitumor immunity remains unexplored. Most recently, EZH2 was shown to regulate methylation of the Foxp3 promoter in Tregs and to affect the stability of Tregs in tissues., Here, we sought to determine whether EBV-encoded LMP1 regulates EZH2 expression in NPC resulting in aTreg generation, which then impairs antitumor immunity. We explored the regulatory role of LMP1 in the differentiation of aTregs in NPC and changes in EZH2 expression in vitro. We also investigated the immunosuppressive effect of LMP1-HONE1 antigen-induced aTregs (with or without EZH2 inhibitor DZnep) on tumor-associated antigen (TAA)-specific T cellsin vivo and in vitro. Furthermore, we correlated LMP1 immunohistochemical (IHC) staining and EZH2+ Treg numbers with clinicopathological factors in NPC patients.
| > Materials and Methods|| |
In total, 93 NPC patients were recruited in this study. Seven patients were subsequently excluded due to incomplete follow-up data. The clinicopathological characteristics of the remaining 86 NPC patients are presented in [Table 1]. Clinical staging and anatomical tumor sites were assessed according to the 6th edition of the Union for International Cancer Control (2008) tumor–node–metastasis classification of malignant tumors. All patients received standard curative radiotherapy with or without chemotherapy at the theFirst People's Hospital of Foshan between January 2008 and July 2013. All clinicopathological data were obtained from the hospital's database with the approval of the Ethics Committee of theFirst People's Hospital of Foshan (Foshan, China).
|Table 1: Clinicopathological features of nasopharyngeal carcinoma patients|
Click here to view
Ethics approval and consent to participate
The study protocol was approved by the Ethics Committee of theFirst Affiliated Hospital of Sun Yat-Sen University. Healthy donor (HD) informed consent was obtained before enrollment. All clinicopathological data were obtained from the hospital's database with approval of the Ethics Committee of theFirst People's Hospital of Foshan (Foshan, China). Animal experimental procedures were performed in accordance with the Declaration of Helsinki and were approved by the Institutional Animal Studies Committee and conducted in accordance with the Institutional Animal Care and Use Committee guidelines.
Cell lines and latent membrane protein 1 transfection
The human NPC cell line HONE1, which was established from a biopsy specimen of a poorly differentiated squamous cell carcinoma of the nasopharynx, was maintained in RPMI 1640 media (Sigma-Aldrich, St. Louis, MO, USA) supplemented with 2 mM L-glutamine and 10% fetal bovine serum. The cells were grown for a maximum of ten passages before thawing fresh cells. pLV[Exp]-EGFP: T2A: Neo-EF1A>ORF_1161 bp (LMP1, accession number: YP_401722) and pLV[Exp]-CMF>EGFP: T2A: Neo were packaged into a lentiviral packaging construct (Cyagen, Guangzhou, China). To generate stable LMP1-expressing cells (LMP1-HONE1), HONE1 cells were transduced with the lentiviral vector multiplicity of infection (MOI = 50:1) followed by selection with 300 μg/ml G418 (Merck, Kenilworth, NJ, USA) for 2 weeks. LMP1 expression was analyzed by immunoblotting using the primary antibodies anti-LMP1 (1:1000; ab78113, CS1-4, Abcam, Cambridge, UK) and anti-β-actin (1:5000; A5441, AC-15, Sigma-Aldrich). The signals were detected using ECL reagent (GE Healthcare, Chicago, IL, USA).
Multicolor flow cytometry was conducted using a 10-color (3-laser: 488 nm blue, 638 nm red, and 405 nm violet) Gallios flow cytometer equipped with Gallios software v1.0 (Beckman Coulter, Hercules, CA, USA). To determine the frequency of the three distinct subsets of Foxp3+ CD4+ T cells, the following antibodies were used: anti-hCD4-FITC (11-0048-42, surface staining, eBioscience, Waltham, MA, USA), anti-hCD45RA-BV570 (304131, surface staining, Biolegend, USA), and anti-hFoxp3-PE (12-4776-42, intracellular staining, eBioscience). To determine the expression of EZH2 and phosphatase and tensin homolog (PTEN) in the Tregs subsets, anti-hEZH2-AF647 and anti-hPTEN-AF647 antibodies were used (563491 and 560003, BD Biosciences, Franklin Lakes, NJ, USA).
Cell preparation and coculture
The study protocol was approved by the Ethics Committee of theFirst Affiliated Hospital of Sun Yat-Sen University. HD informed consent was obtained before enrollment. Peripheral blood mononuclear cells (PBMCs) were obtained from HDs by Ficoll density gradient centrifugation. CD14+ cells were isolated from PBMCs using CD14 microBeads (130-050-201, Miltenyi Biotec, Bergisch Gladbach, Germany). CD14+ cells were then seeded into 6-well plates at a density of 1.5 ×106 cells/well and cultured as previously described. On day 6, tumor antigen (HONE1 cell, LMP1-HONE1 cell, and vector-HONE1 cell antigens), prepared as previously described, was added to the dendritic cell (DC) cultures at a ratio of 3:1 (tumor cells: DCs). On day 7, maturation of the DCs was induced by adding 1 μg/mL lipopolysaccharide (Sigma-Aldrich) and culturing for 2 days. Antigen-loaded DCs were then added to the autologous T cells as stimulators at a ratio of 1:20 in round-bottom 96-well microplates. On days 5 or 7, the CD3 + cells were isolated from the coculture using CD3 microbeads (130-050-101, Miltenyi Biotec). Then, TAA-specific CD3+ CD25− T cells and antigen-induced subsets of Foxp3+ CD4+ T cells were isolated as described previously. Typically, the cells were stained with anti-hCD4-FITC, anti-hCD25-APC, and anti-hCD45RA-eFluor 450 antibodies (eBioscience) and sorted using a BD Influx cell sorter (BD Biosciences). Three subsets of living Foxp3+ CD4+ T cells were prepared as described previously. CD45RA+ Foxp3low CD4+ T cells, which were CD25++ (I: Resting Tregs [rTregs]), CD45RA− Foxp3high CD4+ T cells, which were CD25+++ (II: aTregs), and CD45RA− Foxp3low CD4+ T cells, which were CD25++ (III: Cytokine-secreting CD4+ T cells), were prepared by sorting CD45RA+ CD25++ CD4+, CD45RA− CD25+++ CD4+, and CD45RA − CD25++ CD4 + cells, respectively.
In vitro tumor-associated antigen immunosuppression in the presence of antigen-induced activated Tregs
To analyze TAA-specific T cell immunosuppression in the presence of antigen-induced aTregs, HONE1 and LMP1-HONE1 antigen-induced aTregs were added to Cell Counting Kit-8-labeled TAA-specific CD3+ CD25− T cells at a ratio of 1:20 in the presence of anti-CD3 and anti-CD28. The proliferation of TAA-specific CD3+ CD25− T cells was then assessed by Cell Counting Kit-8 analysis.
Nude mice (BALB/c-nu/nu, male, 8 weeks old) were purchased from the Animal Care Center of the Sun Yat-Sen University (Guangzhou, China) and housed under specific pathogen-free conditions. All experimental procedures were approved by the Institutional Animal Studies Committee and conducted in accordance with the Institutional Animal Care and Use Committee guidelines.
Forty mice were randomly and equally divided into eight groups and ear-tagged prior to treatment. On day 0, 1 × 105 tumor cells were subcutaneously injected into the right flank of each mouse. Cells used for injection included TAA-specific T cells and antigen-induced aTregs (HONE1 and LMP1-HONE1 aTregs). Seven groups of nude mice were treated as follows: (a) control group; (b) infusion of TAA-specific CD3+ CD25− T cells; (c) infusion of TAA-specific CD3+ CD25− T cells and HONE1 aTregs; (d) infusion of TAA-specific CD3+ CD25− T cells and LMP1-HONE1 aTregs; (e) infusion of TAA-specific CD3+ CD25− T cells and vector-HONE1 aTregs; (f) infusion of TAA-specific CD3+ CD25− T cells and DZnep-treated (48-h treatment with the EZH2 inhibitor DZnep) HONE1 aTregs; (g) infusion of TAA-specific CD3+ CD25− T cells and DZnep-treated LMP1-HONE1 aTregs; and (h) infusion of TAA-specific CD3+ CD25− T cells and DZnep-treated vector-HONE1 aTregs.
The immune cells were injected around the tumor twice a week for 2 weeks starting on day 7 following tumor cell injection. A group of five mice that received saline in parallel was established as a control. Tumor sizes were measured every 7 days using fine calipers. Tumor volumes were calculated as: Volume = Length × width2/2. All mice were sacrificed within 2 weeks after the last injection.
Immunohistochemical and immunofluorescent analysis
Paraffin-embedded formalin-fixed 4-mm-thick tissue sections were processed for IHC with anti-LMP1 (1:100; ab78113, Abcam) antibody and for immunofluorescence with anti-EZH2 (1:250; ab186006, Abcam) and anti-Foxp3 (1:100; ab20034, Abcam) antibodies.
IHC expression levels for LMP1 were assessed by a semi-quantitative scoring system according to the intensity of staining and the percentage of stained tumor cells. Staining intensity was scored as 0 = negative, 1 = weak, and 2 = strong. The percentage of stained tumor cells were scored as 0 = no stained tumor cells, 1 = 1%–10% stained tumor cells, 2 = 11%–50% stained tumor cells, and 3 = 51%–100% stained tumor cells. The two individual parameters were added, resulting in an immunoreactivity score (IRS) ranging from 0 to 5. We defined cases with IRS ≥2 as high expression and cases with IRS <2 as low expression.
The densities of EZH2- and Foxp3-positive cells were quantitatively evaluated from the mean counts of four representative fields at ×400 magnification by two independent observers who were blinded to the clinical outcomes.
All statistical analyses were performed with SPSS version 22.0 (IBM, Armonk, NY, USA). For Foxp3 and EZH2, the median value was used to define the cutoff to the subgroups. Survival variables were estimated using the Kaplan–Meier method and compared using log-rank tests. The differences between groups were assessed using the Mann–Whitney U-test, Student's t-test, or Kruskal–Wallis test. P < 0.05 was considered statistically significant.
| > Results|| |
Latent membrane protein 1 enhanced the stability of HONE1 antigen-induced activated Tregs
To investigate effects of LMP1 on Tregs, LMP1 overexpressing lentivirus was transduced into HONE1 cells to generate the LMP1-HONE1 cell line [Figure 1]a. Tumor cell antigens were extracted using a previously reported method (Sun et al., 2016). Then, the DCs were stimulated ith LMP1-HONE1, vector-HONE1, or HONE1 antigens and individually cocultured with CD3+ T cells. The results showed that CD3+ T cell proliferation was observed in the HONE1 and vector-HONE1 groups, but not in the LMP1-HONE1 group [Figure 1]b. Flow cytometry results showed that although the frequencies of aTregs, rTregs, and Foxp3low CD45RA− CD4+ T cells were higher in the LMP1-HONE1, vector-HONE1, and HONE1 groups than in the control group on day 5, there was no significant difference in the frequency of aTregs among the LMP1-HONE1, vector-HONE1, and HONE1 groups. There was also no significant difference in the frequency of rTregs and Foxp3low CD45RA− CD4+ T cells among the LMP1-HONE1, vector-HONE1, and HONE1 groups. However, the frequency of aTregs (1.76 ± 0.16) in the LMP1-HONE1 group remained at a high level on day 7, while the frequency of aTregs decreased in the HONE1 (0.88 ± 0.25) and vector-HONE1 (0.92 ± 0.23) groups. In addition, no significant difference in rTregs was observed among the four groups (0.36 ± 0.03, 0.43 ± 0.02, 0.42 ± 0.03, and 0.43 ± 0.03, respectively). The frequency of Foxp3low CD45RA− CD4+ T cells in the LMP1-HONE1 group (6.96 ± 1.09) was higher than in the other groups (4.36 ± 1.13, 5.01 ± 0.83, and 4.66 ± 0.62, respectively), but the difference was not statistically significant [Figure 2]a and [Figure 2]b.
|Figure 1: LMP1 expression in NPC cells and the proliferation of CD3 + T cells in different conditions. (a) LMP1 expression in the HONE1, vector-HONE1, and LMP1-HONE1 groups. (b) Proliferation of CD3 + T cells in cultures stimulated with HONE1, vector-HONE1, and LMP1-HONE1 antigens. Red arrows show proliferating CD3 + T cells (×100). LMP1 = Latent membrane protein 1, NPC = Nasopharyngeal carcinoma|
Click here to view
|Figure 2: Flow cytometry results of Foxp3+CD4+ T cell subsets in groups with different antigen stimulation. (a) Representative frequencies of Foxp3+CD4+ T cell subsets induced in groups following stimulation with different antigens on days 5 and 7. (b) Statistical analysis showing that the percentage of aTregs, rTregs, and Foxp3lowCD45RA−CD4+ T cells in the HONE1, LMP1-HONE1, and vector-HONE1 groups is higher than in the control group on day 5, and that the percentages of aTregs on day 7 is higher in the LMP1-HONE1 group than in the other groups (n = 4). *P < 0.05; error bars are shown; I: rTregs, II: aTregs, III: Foxp3lowCD45RA−CD4+ T cells. aTregs = Activated Tregs, rTregs = Resting Tregs, LMP1 = Latent membrane protein 1|
Click here to view
Enhancer of zeste homolog 2 but not phosphatase and tensin homolog was upregulated in latent membrane protein 1-HONE1 antigen-induced activated Tregs
PTEN, a major tumor suppressor, was recently shown to enhance the stability of Tregs by inhibiting the phosphatidylinositol-4,5-bisphosphate 3-kinase (PI3K) signaling pathway. Therefore, next, we analyzed the average fluorescence intensity of EZH2 and PTEN in aTregs under different conditions. The results showed that the mean fluorescence intensity (MFI) of EZH2 in aTregs in the LMP1-HONE1 group (10168 ± 1368) was higher than in the control (2528 ± 1084, P = 0.0047), HONE1 (2935 ± 998.7, P = 0.0053), and vector-HONE1 (2786 ± 1030; P = 0.005) groups [Figure 3]a. However, there were no statistically significant differences in EZH2 MFIs among the control, LMP1-HONE1, vector-HONE1, and HONE1 groups for rTregs and Foxp3low CD45RA− CD4+ T cells. In addition, PTEN MFI in the Foxp3+ CD4+ T cell subset was also not significantly different among the groups [Figure 3]b.
|Figure 3: EZH2 and PTEN expression in Foxp3+CD4+T cell subsets from the different groups. (a) Representative EZH2 expression in aTregs, rTregs, and Foxp3lowCD45RA− CD4+ T cells from the different groups (upper panel). Statistical analysis shows that the MFI of EZH2 in aTregs is higher in the LMP1-HONE1 group than in other groups. There were no statistically significant differences in EZH2 MFIs among the control, LMP1-HONE1, vector-HONE1, and HONE1 groups (lower panel) (n = 4) for rTregs and Foxp3lowCD45RA− CD4+ T cells. (b) Representative PTEN expression in aTregs, rTregs, and Foxp3lowCD45RA− CD4+ T cells from the different groups (upper panel). The MFI of PTEN in Foxp3+ CD4+ T cell subsets among the groups showed no significant difference (lower panel) (n = 4). *P < 0.05; error bars are shown. EZH2 = Enhancer of zeste homolog 2, PTEN: Phosphatase and tensin homolog, aTregs = Activated Tregs, rTregs = Resting Tregs, MFI = Mean fluorescence intensity, LMP1 = Latent membrane protein 1|
Click here to view
Inhibiting enhancer of zeste homolog 2 depleted latent membrane protein 1-HONE1 antigen-induced activated Tregs
The pharmacological EZH2 inhibitor, DZnep, was used to further study whether EZH2 contributed to LMP1-induced aTreg stability. Following DZnep treatment, the proportion of aTregs was depleted even with LMP1-HONE1 antigen stimulation [Figure 4]a. Moreover, the proportion of rTregs (0.445 ± 0.02 vs. 0, P < 0.001) and Foxp3low CD45RA− CD4 + T cells (5.10 ± 0.22 vs. 0.575 ± 0.05, P < 0.001) in the DZnep group was decreased compared with controls. These results support the hypothesis that LMP1 increases the stability of Foxp3+ CD4+ T cells, especially aTregs, by upregulating EZH2 expression.
|Figure 4: The EZH2 inhibitor DZnep induced potent anticancer immunity in vitro and in vivo. (a) Dznep depleted the number of LMP1-HONE1 antigen-induced aTregs. The proportion of rTregs and Foxp3lowCD45RA− CD4+ T cells in the DZnep group was decreased compared to the controls (n = 4). (b) Cell Counting Kit-8 was used to quantitate the proliferation of TAA-specific CD3+ CD25− T cells (TAA-T cells) in the presence of different antigen-induced aTregs (n = 4). The inhibitory effect of LMP1-HONE1 antigen-induced aTregs on TAA-T cell proliferation was enhanced compared to HONE1 antigen-induced aTregs. (c) Tumor volumes in the different groups (n = 5). Tumor sizes were measured every 4 days following the first lymphocyte injection. Tumor volumes in the DZnep-treated group was similar to that in the TAA-T group, both of which were smaller than those without DZnep treatment. Tumor volume in the TAA-T cell + LMP1-HONE1 aTreg group was larger than in the TAA-T cell + HONE1 aTreg group. *P < 0.05; error bars are shown. EZH2 = Enhancer of zeste homolog 2, aTregs = Activated Tregs, rTregs = Resting Tregs, LMP1 = Latent membrane protein 1, TAA = Tumor-associated antigen|
Click here to view
Potent anticancer immunity in mice receiving DZnep-treated antigen-induced activated Tregs
First, we investigated whether LMP1 could enhance the inhibitory function of LMP1-HONE1 antigen-induced aTregs. Antigen-induced CD25+++ CD45RA− CD4+T cells were cocultured with Cell Counting Kit-8-labeled TAA-specific CD3 + CD25 − T cells. The results showed that the inhibitory effect of LMP1-HONE1 antigen-induced aTregs on TAA-specific CD3 + CD25 − T cell proliferation was enhanced compared to HONE1 antigen-induced aTregs (P < 0.05) [Figure 4]b, which was in line with the observation that fewer proliferative CD3+ T cells were observed in the LMP1-HONE1-stimulated system.
Animal models were then designed to testin vivo antitumor immunity following targeted pharmacological inhibition of EZH2 in antigen-induced aTregs. TAA-specific CD3 + CD25 − T cells and antigen-induced aTregs treated with/without DZnep were injected into mice on days 7 and 15. Tumor volumes in each group are shown in [Figure 4]c. After 7 days, the tumor volume in the DZnep-treated group was similar to that of the TAA-specific CD3+ CD25 − T cell group, both of which were lower than in the groups without DZnep treatment, indicating that inhibition of EZH2 with DZnep enhanced the antitumor immunity, possibly by attenuating the stability of LMP1-HONE1 antigen-induced aTregs. In addition, tumor volume in the TAA-specific CD3+ CD25 − T cell+ LMP1-HONE1 aTreg group was larger than in the TAA-specific CD3+ CD25 − T cell+ HONE1 aTreg group, indicating that LMP1 enhanced the inhibitory effect of HONE1 antigen-induced aTregs on TAA-specific T cells.
Tumor-infiltrating enhancer of zeste homolog 2+ Tregs were positively associated with latent membrane protein 1 expression and predicted poor survival outcomes
Eighty-six NPC patients with an average age of 48.15 ± 14.15 years were enrolled in this study [Table 1]. IHC results showed that LMP1 staining was primarily localized to the plasma membrane and cytoplasm of NPC cells [Figure 5]a. Forty-one samples were found to have low LMP1 expression, whereas 45 showed high LMP1 expression. High and low LMP1 expressions were observed in 32 (73%) and 12 (27%) Stage IV samples, respectively. High LMP1 expression was observed in nine (35%) and 4 (25%) Stage III and Stages I/II samples, respectively, whereas low LMP1 expression was found in 17 (65%) Stage III and 12 (75%) Stages I/II samples [Figure 5]b.
|Figure 5: Relationships between LMP1 expression, tumor-infiltrating EZH2− /+ Tregs, and clinical stage in NPC. (a) Representative images of LMP1-stained (immunohistochemical) NPC tissues with low and high LMP1 expression. Brown color indicates LMP1 expression (×400). (b) Comparison of the percentage of clinical stages with low and high LMP1 expression. (c) Foxp3 and EZH2 expression in NPC tissues by immunofluorescence staining. Green indicates Foxp3+, red indicates EZH2+, and yellow indicates Foxp3 + and EZH2+ (×400). (d-i) The number of tumor-infiltrating cells in the different groups, *P < 0.05. LMP1 = Latent membrane protein 1, EZH2 = Enhancer of zeste homolog 2, NPC = Nasopharyngeal carcinoma|
Click here to view
Further, EZH2− Foxp3+, EZH2+ Foxp3+, and EZH2+ Foxp3− cells were quantified by immunofluorescence [Figure 5]c. Analysis of the relationships between LMP1 level and tumor-infiltrating EZH2− /+ Tregs showed that there were no statistically significant differences between the number of tumor-infiltrating EZH2− /+ Tregs in cases with low LMP1 expression [Figure 5]d, whereas the number of tumor-infiltrating EZH2+ Tregs was higher than EZH2 − Tregs in the cases with high LMP1 expression [Figure 5]e. Furthermore, the number of tumor-infiltrating EZH2+ Tregs in cases with high LMP1 expression was higher than in cases with low LMP1 expression [Figure 5]f. There was no significant difference in the number of EZH2 − Tregs between cases with high and low LMP1 expression [Figure 5]g. These results suggested that tumor-infiltrating EZH2+ Tregs were positively associated with LMP1 expression.
The number of tumor-infiltrating EZH2+ Tregs in Stage IV (29.7 ± 7.5) was higher than in Stages I/II (13.5 ± 2.7) (P < 0.001) and Stage III (22.6 ± 5.1) (P < 0.001). Moreover, the number of tumor-infiltrating EZH2+ Tregs in Stage III (22.6 ± 5.1) was higher than in Stages I/II (13.5 ± 2.7) (P < 0.001) [Figure 5]h. However, the number of EZH2 − Tregs was not significantly different between Stages I/II (17.6 ± 5.1), III (19.6 ± 5.6), and IV (18.9 ± 5.3) (all P > 0.05) [Figure 5]i.
Patients with high LMP1 expression had worse overall survival (42.6 months) than those with low LMP1 expression (55.9 months) [Figure 6]a. There was no significant difference in the overall survival between patients with high and low number of tumor-infiltrating Tregs (P > 0.05) [Figure 6]b. However, in terms of EZH2 expression, patients with higher number of tumor-infiltrating EZH2+ Tregs had reduced overall survival (42.8 months) compared with those with lower number of tumor-infiltrating EZH2+ Tregs (55.6 months) [Figure 6]c. There was no significant difference in the overall survival between patients with high and low number of EZH2 − Tregs (P > 0.05) [Figure 6]d.
|Figure 6: Kaplan–Meier overall survival curves based on LMP1 levels and number of tumor-infiltrating Tregs, EZH2+ Tregs, and EZH2− Tregs in 86 NPC patients. (a) LMP1 predicts poor survival in NPC patients. (b) The number of tumor-infiltrating Tregs was not correlated with overall survival. (c and d) Numbers of tumor-infiltrating EZH2+ Tregs (c) but not EZH2− Tregs (d) predicted poor survival in NPC patients. LMP1 = Latent membrane protein 1, EZH2 = Enhancer of zeste homolog 2, NPC = Nasopharyngeal carcinoma|
Click here to view
Cox multivariate analysis showed that only LMP1 expression influenced overall survival (P = 0.005, relative risk: 1.65) [Supplementary Table 1]. Differences in treatment modalities and other factors known to correlate with survival were also included in this model and did not change the significance of these variables.
| > Discussion|| |
The role of Tregs in tumor immune escape has been widely recognized. Several studies have confirmed that Tregs are highly expressed in the peripheral blood and tumor microenvironment of patients with different malignant tumors, such as gastric cancer, liver cancer, pancreatic cancer, breast cancer, and NPC. Currently, immunotherapy against Tregs has become an area of intensive research for combined strategies for the treatment of malignant tumors. However, recent studies have shown that functionally distinct subsets of Foxp3 + Tregs exist, and not all subsets of Foxp3 + T cells have immunosuppressive activity., Miyara et al. showed that Foxp3 + T cells could be divided into three functionally distinct subsets: CD45RA+ Foxp3low rTregs and CD45RA − Foxp3high aTregs, both of which are suppressive in vitro, and cytokine-secreting CD45RA − Foxp3low CD4 + non-suppressive T cells. Based on this new classification, our recent study showed that the number of peripheral blood aTregs with significant suppressive ability was significantly increased in NPC patients compared to HDs. Therefore, it is necessary to further study the specific mechanism of aTregs differentiation in NPC. In this study, we found that the EBV-encoded pathogenic protein LMP1 increased the stability of aTregs by upregulating EZH2, which further enhanced the immunosuppressive effects of aTregs on TAA-specific T cell antitumor immunity. Likewise, in tumor specimens, LMP1 expression was positively correlated with the number of EZH2+ Tregs, which was associated with advanced clinical stage and poor survival in NPC patients.
EBV may escape immune clearance through LMP1 mimicking CD40 signals as part of the “normal” human biological function.In vitro studies have confirmed that LMP1 inhibits T cell responses by inducing Treg differentiation. However, the effects of LMP1 on Treg differentiation in NPC, especially on the differentiation of functional Treg subsets, are unknown. In this study, using our previously described tumor antigen preparation methods, the differentiation of functional Treg subsets was compared between the LMP1-HONE1 and vector-HONE1 groups. The results showed that the frequency of aTregs in the HONE1-LMP1 group was significantly higher than in the HONE1-vector group on day 7, although there was no significant difference between the groups on day 5. This suggested that LMP1 may not be involved in the initial stages of NPC tumor antigen-induced aTreg differentiation from CD4 + T cells, rather it played a role subsequently in maintaining the stability of the induced aTregs.
We next examined the specific mechanism of LMP1-HONE1-induced aTreg stability in NPC. Recent studies have shown that EZH2 and PTEN are two key players in maintaining the stability and differentiation of Tregs., For example, Zhang et al. found that the differentiation of Tregs from T cells could be reduced by silencing EZH2. Arvey et al. showed that EZH2 aggregated around the Foxp3 locus when Tregs were activated, suggesting that EZH2 might play a role in regulating Foxp3 expression. Yang et al. confirmed that Treg suppressive ability was impaired in EZH2-knockout mice. In addition, Huynh et al. found that regulation of PI3K signaling by PTEN was essential for maintaining Treg function and stability. Shrestha et al. showed that PTEN regulated Treg stability through metabolic pathways such as mTORC2. Our results showed that the MFI of EZH2 but not PTEN was significantly higher in aTregs from the LMP1-HONE1 group than the vector-HONE1 group, and that the increased number of aTregs in LMP1-HONE1 group could be depleted by treatment with the EZH2 inhibitor. This suggested that EZH2 was involved in maintaining aTreg stability downstream of LMP1 in NPC.
We further studied the immunosuppressive capability of LMP1-HONE1-induced aTregs and found that they suppressed TAA-specific T cell proliferation more effectively than HONE1-induced aTregs, which corroborated thein vitro mixed lymphocyte culture system results, which also showed decreased T cell proliferation in the LMP1-HONE1 group compared to the HONE1 group.In vivo experiments also showed that the tumor volumes in the TAA-specific T cells + LMP1-HONE1 aTregs group were significantly increased compared to the TAA-specific T cells + HONE1 aTregs and TAA-specific T cells + LMP1-HONE1 aTregs group when EZH2 inhibitor pretreatment was included. Taken together, thesein vitro andin vivo results suggest that LMP1/EZH2 signaling plays a crucial role in maintaining the suppressive activity and functional stability of aTregs, which in turn inhibits TAA-specific immune responses.
Different views on the relationships between LMP1, tumor-infiltrating Tregs, clinical stage, and prognosis have been reported for NPC patients., Zhang et al. showed that LMP1 expression (negative or positive in their study) was not associated with the number of Tregs in the prognosis of NPC patients. In addition, their study showed that Foxp3 (high and low expression) was negatively correlated with tumor stage and was positively correlated with overall survival and progression-free survival in NPC patients. In another study, LMP1 expression was determined by staining intensity and the percentage of positively stained tumor cells. This study concluded that LMP1 expression was associated with the prognosis of NPC patients. In addition, a study of 197 NPC cases showed that the number of tumor-infiltrating Tregs were negatively correlated with overall survival and disease-free survival in NPC patients. These inconsistent findings have primarily been attributed to different definitions of positive/high and negative/low expression. In this study, we found that LMP1 expression was positively correlated with numbers of tumor-infiltrating EZH2+ Tregs and clinical stage. In addition, our results also showed that the number of tumor-infiltrating EZH2+ but not EZH2− Tregs was positively correlated with clinical stage and reduced overall survival. These clinical observations suggested that LMP1-induced EZH2+ Tregs are involved in immunosuppressive processes during NPC development.
| > Conclusions|| |
Our results showed that EBV-encoded LMP1 increased tumor-specific immunosuppression by upregulating EZH2 in aTregs, which increased their functional stability. In addition, the number of EZH2+ Tregs in the tumor microenvironment positively correlated with clinical stage and reduced overall survival, suggesting that immunotherapy targeting EZH2+ aTregs could be a potentially new therapeutic strategy for NPC.
We thank James P. Mahaffey, PhD, from Liwen Bianji, Edanz Editing China (www.liwenbianji.cn/ac), for editing the English text of a draft of this manuscript.
Financial support and sponsorship
This study was supported by grants from the Natural Science Foundation of Guangdong Province (2016A030310153, 2014A030313031, and 2016A030313257), the National Natural Science Foundation of China (81602365, 81670902, 81470674 and 81972527), the Guangzhou Science and Technology Program Key Projects (201605030003), and the Natural Science Foundation of Guangdong Province (2017A030310362 and 2018A030313667).
Conflicts of interest
There are no conflicts of interest.
| > References|| |
Tao Q, Chan AT. Nasopharyngeal carcinoma: Molecular pathogenesis and therapeutic developments. Expert Rev Mol Med 2007;9:1-24.
Aksoy A, Elkiran ET, Harputluoglu H, Dagli AF, Isikdogan A, Urakci Z. Is excision repair cross-complementation Group 1 expression a biological marker in nasopharynx carcinoma? J Cancer Res Ther 2019;15:550-5.
Ribas A, Wolchok JD. Cancer immunotherapy using checkpoint blockade. Science 2018;359:1350-5.
Miyara M, Yoshioka Y, Kitoh A, Shima T, Wing K, Niwa A, et al
. Functional delineation and differentiation dynamics of human CD4+T cells expressing the FoxP3 transcription factor. Immunity 2009;30:899-911.
Li J, Qian CN, Zeng YX. Regulatory T cells and EBV associated malignancies. Int Immunopharmacol 2009;9:590-2.
Lu J, Chen XM, Huang HR, Zhao FP, Wang F, Liu X, et al
. Detailed analysis of inflammatory cell infiltration and the prognostic impact on nasopharyngeal carcinoma. Head Neck 2018;40:1245-53.3.
Sun W, Li WJ, Wu CY, Zhong H, Wen WP. CD45RA-Foxp3high but not CD45RA+Foxp3low suppressive T regulatory cells increased in the peripheral circulation of patients with head and neck squamous cell carcinoma and correlated with tumor progression. J Exp Clin Cancer Res 2014;33:35.
Sun W, Wei FQ, Li WJ, Wei JW, Zhong H, Wen YH, et al
. A positive-feedback loop between tumour infiltrating activated Treg cells and type 2-skewed macrophages is essential for progression of laryngeal squamous cell carcinoma. Br J Cancer 2017;117:1631-43.
Yang L, Lu Z, Ma X, Cao Y, Sun LQ. A therapeutic approach to nasopharyngeal carcinomas by DNAzymes targeting EBV LMP-1 gene. Molecules 2010;15:6127-39.
Liu J, Liu Y, Zhang Z, Sun H, Ji X, Li B, et al
. Prognostic value of the Epstein-Barr virus and tumor suppressor gene p53 gene in nasopharyngeal squamous cell carcinoma. J Cancer Res Ther 2019;15:426-36.
Dawson CW, Port RJ, Young LS. The role of the EBV-encoded latent membrane proteins LMP1 and LMP2 in the pathogenesis of nasopharyngeal carcinoma (NPC). Semin Cancer Biol 2012;22:144-53.
DuPage M, Chopra G, Quiros J, Rosenthal WL, Morar MM, Holohan D, et al
. The chromatin-modifying enzyme Ezh2 is critical for the maintenance of regulatory T cell identity after activation. Immunity 2015;42:227-38.
Wu C, Chen Z, Kuchroo VK. Ezh2 lines up the chromatin in T regulatory cells. Immunity 2015;42:201-3.
Sun W, Li WJ, Wei FQ, Wong TS, Lei WB, Zhu XL, et al
. Blockade of MCP-1/CCR4 signaling-induced recruitment of activated regulatory cells evokes an antitumor immune response in head and neck squamous cell carcinoma. Oncotarget 2016;7:37714-27.
Wei FQ, Sun W, Wong TS, Gao W, Wen YH, Wei JW, et al
. Eliciting cytotoxic T lymphocytes against human laryngeal cancer-derived antigens: Evaluation of dendritic cells pulsed with a heat-treated tumor lysate and other antigen-loading strategies for dendritic-cell-based vaccination. J Exp Clin Cancer Res 2016;35:18.
Chen J, Hu CF, Hou JH, Shao Q, Yan LX, Zhu XF, et al
. Epstein-Barr virus encoded latent membrane protein 1 regulates mTOR signaling pathway genes which predict poor prognosis of nasopharyngeal carcinoma. J Transl Med 2010;8:30.
Huynh A, DuPage M, Priyadharshini B, Sage PT, Quiros J, Borges CM, et al
. Control of PI(3) kinase in Treg cells maintains homeostasis and lineage stability. Nat Immunol 2015;16:188-96.
Liu Z, Kim JH, Falo LD Jr., You Z. Tumor regulatory T cells potently abrogate antitumor immunity. J Immunol 2009;182:6160-7.
Morse MA, Hobeika AC, Osada T, Serra D, Niedzwiecki D, Lyerly HK, et al
. Depletion of human regulatory T cells specifically enhances antigen-specific immune responses to cancer vaccines. Blood 2008;112:610-8.
Litzinger MT, Fernando R, Curiel TJ, Grosenbach DW, Schlom J, Palena C. IL-2 immunotoxin denileukin diftitox reduces regulatory T cells and enhances vaccine-mediated T-cell immunity. Blood 2007;110:3192-201.
Nishikawa H, Sakaguchi S. Regulatory T cells in tumor immunity. Int J Cancer 2010;127:759-67.
Wingate PJ, McAulay KA, Anthony IC, Crawford DH. Regulatory T cell activity in primary and persistent Epstein-Barr virus infection. J Med Virol 2009;81:870-7.
Zhang Y, Kinkel S, Maksimovic J, Bandala-Sanchez E, Tanzer MC, Naselli G, et al
. The polycomb repressive complex 2 governs life and death of peripheral T cells. Blood 2014;124:737-49.
Arvey A, van der Veeken J, Samstein RM, Feng Y, Stamatoyannopoulos JA, Rudensky AY. Inflammation-induced repression of chromatin bound by the transcription factor Foxp3 in regulatory T cells. Nat Immunol 2014;15:580-7.
Yang XP, Jiang K, Hirahara K, Vahedi G, Afzali B, Sciume G, et al
. EZH2 is crucial for both differentiation of regulatory T cells and T effector cell expansion. Sci Rep 2015;5:10643.
Shrestha S, Yang K, Guy C, Vogel P, Neale G, Chi H. Treg cells require the phosphatase PTEN to restrain TH1 and TFH cell responses. Nat Immunol 2015;16:178-87.
Zhang YL, Li J, Mo HY, Qiu F, Zheng LM, Qian CN, et al
. Different subsets of tumor infiltrating lymphocytes correlate with NPC progression in different ways. Mol Cancer 2010;9:4.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6]