Article Text
Abstract
Aims As numerous signalling molecules regulate effector functions of peripheral blood lymphocytes (PBLs) that have an important anti-tumour activity, the aim of this study was to analyse their level in patients with metastatic melanoma (MM) compared with healthy controls (HCs).
Methods Peripheral blood mononuclear cells (PBMCs) of 36 MMs and 28 HCs were analysed for the level of perforin, interferon-regulating transcription factor-1 (IRF-1), DAP10 and Src homology 2 domain-containing tyrosine phosphatase-1 by reverse transcriptase PCR, level of phosphorylated signal transducers and activators of transcription (pSTAT)-1, pSTAT-4, pSTAT-5 by western blot and interferon (IFN)-γ production by ELISA. The expression of activating NKG2D and inhibitory killer immunoglobulin-like receptors (KIR), CD158a and CD158b, on PBL, CD3−CD56+ natural killer (NK) cells and CD3+CD8+ cytotoxic T lymphocytes (CTLs), as well as the percentage of CD14+HLA-DR- cells in PBMC were estimated by flow cytometry.
Results Patients with MM, compared with HCs, had significantly lower level of cytotoxic molecule perforin and decreased IFN-γ production, as well as lower level of pSTAT-1, pSTAT-4, pSTAT-5 and IRF-1 signalling molecules in PBMC. Furthermore, MM had decreased expression of activating NKG2D receptor on PBL and NK cells and low level of its DAP10 signalling molecule contrary to no changes in KIR expression on all investigated cells. These results could be associated with increased percentage of immunosuppressive CD14+HLA-DR− myeloid-derived suppressor cells detected in patients with MM.
Conclusions The altered signalling molecules of PBL could represent biomarkers of impaired cytotoxic and immunoregulatory function of these cells, indicating melanoma-associated immunosuppression that facilitates tumour progression.
- CELL CYCLE REGULATION
- IMMUNOPHENOTYPING
- MELANOMA
Statistics from Altmetric.com
Introduction
Natural killer (NK) cells and CD8+ cytotoxic T lymphocytes (CTLs) exert their cytotoxic function against tumour cells by releasing perforin and granzymes from their granules and together with CD4+ T lymphocytes have an immunoregulatory role by producing many cytokines especially interferon (IFN)-γ.1
It is well known that numerous transcription factors such as signal transducers and activators of transcription (STATs) or interferon-regulating transcription factor-1 (IRF-1) play an important role in regulating and maintaining both innate and adaptive immune responses. STAT proteins regulate the proliferation and survival of lymphocytes, Th1 differentiation, as well as cytotoxicity and cytokine production by NK and T cells.2 ,3 Furthermore, IRF-1 is a critical effector molecule in IFN-γ-mediated signalling and in the development and function of NK, NKT cells and CTL.4–7 However, there are only a few data in literature about the level of these molecules in peripheral blood lymphocytes (PBLs) of patients with melanoma and healthy controls (HCs).8
The activating NKG2D, c-lectin-like receptor, is expressed by a variety of immune cells, mostly NK cells and CTL. This receptor upon binding to its ligands, the major histocompatibility complex (MHC) class I chain-related proteins A and B, and the UL-16 binding proteins, upregulated on transformed cells, in association with its DAP10 signalling molecule induces NK cell and CTL-mediated cytotoxicity against cancer.9 In this sense, NKG2D-DAP10 receptor complex is fully sufficient to trigger NKG2D-mediated lymphocyte killing against tumour cells.10
Killer immunoglobulin-like receptors (KIRs) belong to the immunoglobulin superfamily, and they are responsible for the inhibition of NK cell-mediated lysis of normal cells that express MHC-I molecules. In this sense, according to the ‘missing-self’ hypothesis, the activation of NK cells occurs in contact with malignantly transformed cells that have lost MHC-I molecules and have therefore become susceptible to lysis. Other than NK cells, CD4 and CD8 T cells, as well as γδ T cells also express KIRs.11 Depending on the sequence in their intracellular domain, KIRs are divided into inhibitory and activating receptors, although inhibitory KIRs are dominant. These KIRs display immunoreceptor tyrosine-based inhibition motifs, which are tyrosine-phosphorylated upon cross-linking with MHC-I molecules and recruit Src homology two domain-containing tyrosine phosphatases, SHP-1 and SHP-2, that dephosphorylate activating adaptor molecules primarily DAP10 and Vav-1 and stop activation.12
Melanoma is the most aggressive form of skin cancer due to its capacity to form metastases.13 Aside from that, various experimental and clinical data indicate that melanoma cells are susceptible to NK cell and CTL-mediated cytotoxicity.14 However, it is well known that tumour cells and numerous immunosuppressive cells in tumour microenvironment such as myeloid-derived suppressor cells (MDSCs) by producing cytokines, growth factors and enzymes evolve various mechanisms to create immunosuppression.15 Numerous new data show that MDSCs play a key role in multiple steps of cancer progression. These heterogeneous immature myeloid cells are characterised by their ability to suppress both adaptive and innate immune responses mainly through direct inhibition of the cytotoxic function of NK cells and CTL. Additionally, MDSCs can be grouped into two main populations exhibiting either a granulocytic or monocytic phenotype, but in contrast to the murine system, widely accepted markers for human MDSCs are still under evaluation.16 In accordance with literature, CD14+HLA-DR− cells could be relevant for the characterisation of human monocytic MDSCs in malignant melanoma, hepatocellular carcinoma, metastatic prostate cancer and renal cell cancer.17–20
In this study, it was of interest to analyse, for the first time, the level of various signalling molecules that regulate effector functions of PBL in patients with metastatic melanoma (MM) compared with HCs. The results obtained in this study could be of importance as these lymphocyte parameters may be good biomarkers of impaired NK and T cell function that is associated with increased susceptibility to melanoma progression.
Materials and methods
Patients
Peripheral venous blood was obtained from 36 patients with MM (stage IV according to 7th modified American Joint Committee on Cancer (AJCC)/Union for International Cancer Control (UICC) staging system)21 and 28 HCs, age and gender matched, with no evidence of any disease or infection. Blood was drawn at the time of diagnosis prior to chemotherapy. Clinicopathological characteristics of all patients and characteristics of HCs enrolled in this study are listed in table 1. Furthermore, patients with MM are divided into two groups based on the localisation of distant metastases according to AJCC/UICC staging system. Patients that have metastases in distant skin, the subcutaneous layer or in distant lymph nodes and normal lactate dehydrogenase (LDH) serum level (M1a) and patients with metastases in the lungs (M1b) are included in M1a+M1b group, while the patients with metastases in vital organs other than the lungs with normal serum LDH level or the patients that have any distant metastasis with elevated LDH are included in M1c group.
Peripheral blood mononuclear cells isolation
Peripheral blood mononuclear cells (PBMCs) were isolated from heparinised blood obtained from HCs and patients with MM using Lymphoprep (Nypacon, Oslo, Norway) density gradient, centrifuged at 500 g for 40 min and washed three times in Roswell Park Memorial Institute 1640 culture medium (CM) (Sigma, St Louis, USA) supplemented with 10% fetal calf serum (Sigma). After washing, PBMCs were immediately used for various analyses.
Flow cytometry analysis
Surface phenotype of freshly isolated PBMC subsets was identified using the following combinations of monoclonal antibodies CD3PerCP/CD56FITC/CD158bPE, CD3PerCP/CD8FITC/CD158bPE, CD3PerCP/CD56PE/CD158aFITC, CD3PerCP/CD8PE/CD158aFITC, CD14PerCP/HLA-DRFITC (Becton Dickinson, San Jose, USA), and CD3PerCP/CD56FITC/NKG2DPE, CD3PerCP/CD8FITC/NKG2DPE (R&D, Minneapolis, USA). The samples were prepared for extracellular staining as previously described and analysed by flow cytometry (FACS Calibur) (Becton Dickinson).2 ,22 A total of 50 000 gated events, verified as PBLs according to their physical characteristics (forward-scattered (FSC) and side-scattered (SSC)), were collected per sample and analysed using CellQUEST software. NK cells and CD8+ T lymphocytes were defined within the PBL population according to their expression of CD3 and CD56 receptors (CD3−CD56+ NK cells) and CD3 and CD8 receptors (CD3+CD8+ T cells). The expression of NKG2D, CD158a and CD158b receptors on total PBL, as well as on CD3−CD56+ NK cells and CD3+CD8+ T cells, was given in the percentage in PBL. The subpopulation of CD14+HLA-DR− cells was derived from PBMC and was also given in the percentage in these cells.
ELISA analysis of IFN-γ
PBMCs (1×106) from HCs and patients with MM were cultured in CM alone and CM supplemented with interleukin (IL)-12 (10 ng/mL) and IL-18 (100 ng/mL) in combination for 48 h. Cell-free supernatants were collected and frozen at −70°C until cytokine assays are preformed. IFN-γ was analysed in the supernatants using ELISA kit (eBioscience, San Diego, USA) according to the manufacturer's protocol.23 The concentration of IFN-γ was calculated from the colorimetric optical density (model 2550EIA Reader; Bio-Rad, Richmond, USA), which was read at 450 nm. The sensitivity limit of assay was 4 pg/mL, and the standard curve range was from 4 to 500 pg/mL. ELISA analysis was performed in duplicate, and the data were calculated from mean of two tests for each sample. The level of IFN-γ was presented as pg/mL.
Reverse transcriptase-PCR
Total RNA was extracted from PBMC of HCs and patients with MM by Trizol reagent (Invitrogen, Madison, USA). One microgram of total RNA was reverse-transcribed in a 20 µL volume using a reverse transcriptase (RT)-PCR kit (Fermentas Life Science, Waltham, USA), according to the manufacturer's instructions. RT-PCR was performed as described previously.24 Amplified samples (10 µL) were electrophoresed on 1.5% agarose gels containing ethidium bromide. The primer sequences are as follow: perforin-1: sense 5′-AAAGTCAGCTCCACTGAAGCTGTG-3′, antisense 5′-AGTCCTCCACCTCGTTGTCCGTGA-3′; IRF-1: sense 5′-GACCAGAGCAGGAACAAG-3′, antisense 5′-TAACTTCCCTTCCTCATCC-3′; DAP10: sense 5′–CAGACCCCAGTCCACCATG-3′, antisense 5′-GTGCCACCACACACCATC-3′; SHP-1: sense 5′-GTGACCCATATTCGGATCCAG-3′, antisense 5′-CTTGAAATGCTCCACCAGGTC-3′; β-actin, housekeeping gene: sense 5′-TGGGTCAGAAGGATTCCTAT-3′, antisense 5′-AAGGAAGGCTGGAAGAGT-3′. Level of all investigated molecules was calculated and given with respect to β-actin.
Western blot analysis
The protein content from freshly isolated PBMC of HCs and patients with MM was determined using Trizol reagent (Invitrogen). Equal amounts of protein (20 μg/well) were separated on 10% sodium dodecyl sulfate polyacrylamide gel and blotted onto nitrocellulose membranes (Bio-Rad). After blocking for 1 h with 5% skim milk in Tris-buffered saline buffer containing 0.1% tween-20, primary mouse antihuman antibodies against pSTAT-1, pSTAT-4 and pSTAT-5 (Becton Dickinson) were added onto membranes and incubated overnight at 4°C. Membranes were washed and incubated with secondary antibody, horseradish peroxidase-conjugated antimouse immunoglobulin G, for 1 h at room temperature and developed using 3,3'-Diaminobenzidine substrate tablets. The blots were developed using enhanced chemiluminescence detection system (Amersham, Arlington Heights, USA).
Quantification of blots and gels
Blots and agarose gels were scanned by using gel image system (Kodak Image 1D Image 1 3.6.), in a greyscale mode at 169 mm pixel size and 1250–1650 (X–Y) pixel count, using the autodensity feature on a scale ranging from 0 (clear) to 255 (opaque). The pixel density was determined and used to calculate the integrated density of a selected band. Values of integrated density were reported in volume units of pixel intensity per mm2.
Statistical analysis
Significance of differences for obtained results was carried out by non-parametric Mann–Whitney U test.
Results
The level of perforin-1 mRNA presented relative to β-actin transcript in PBMC of patients with MM (0.76±0.08) is significantly lower (p<0.05, Mann–Whitney U test) compared with this cytotoxic molecule level in HCs (0.44±0.09) (figure 1A, B).
In patients with MM, IFN-γ supernatant level (41 680.00±113 000.00 pg/mL) from PBMC 48 h in vitro treated with IL-12 and IL-18 in combination is significantly (p<0.05, Mann–Whitney U test) lower compared with IFN-γ production level in HCs (13 790.00±6387.00 pg/mL) (figure 1C).
The measurements of pSTAT protein levels presented relative to β-actin band in PBMC by western blot analysis show that HC individuals have 1.7-fold higher level of pSTAT-1, 1.6-fold higher level of pSTAT-4 and 3-fold higher level of pSTAT-5 compared with the levels of these signalling molecules in patients with MM (figure 2A, B).
The transcription level of IRF-1 normalised to β-actin transcript in PBMC of patients with MM (0.21±0.04) is significantly decreased (p<0.05, Mann–Whitney U test) compared with this signalling molecule level in HCs (0.36±0.04) (figure 2C, D).
By analysing the expression of DAP10 and SHP-1 signalling molecules, we show a significant decrease (p<0.05, Mann–Whitney U test) in mRNA level of DAP10 in PBMC of patents with MM (0.39±0.03) in comparison to the level of this signalling molecule in HCs (0.67±0.09). Furthermore, there is no significant difference (p>0.05, Mann–Whitney U test) in the transcription level of the KIR signalling molecule, SHP-1, in PBMC between patients with MM and HCs (figure 3A, B).
By analysing the expression of activating NK cell receptor, NKG2D, we show a significant decrease (p<0.05, Mann–Whitney U test) in its percentage on total PBL and highly significant decrease (p<0.01, Mann–Whitney U test) on CD3–CD56+ NK cells in patients with MM (40.19±2.97% on PBL and 9.55±1.09% on NK cells) compared with the expression of this receptor in HCs (49.43±2.35% on PBL and 15.95±1.93% on NK cells). Contrary to this, there is no significant difference (p>0.05, Mann–Whitney U test) in the percentage of NKG2D on CD3+CD8+ T cells between patients with MM and HCs (figure 3C).
Furthermore, there is no significant difference (p>0.05, Mann–Whitney U test) in the percentage of both investigated inhibitory KIRs, CD158a and CD158b, on total PBL, CD3−CD56+ NK cells, as well as on CD3+CD8+ T cells between patients with MM and HCs (figure 3D).
We show that both investigated groups of patients with MM, M1a+M1b and M1c have significantly (p<0.05, Mann–Whitney U test) higher percentage of CD14+HLA-DR− cells in PBMC (0.68±0.06% in M1a+M1b and 1.09±0.04% in M1c) compared with the percentage of these cells in HCs (0.14±0.03%). Contrary to this, there is no significant difference (p>0.05, Mann–Whitney U test) in the percentage of these cells between two groups of patients with MM (figure 4A, B).
Discussion
Although NK and CD8+ cells are the most important cytotoxic lymphocytes and perforin, cytolytic protein in their granules, has an important role in these cells-mediated lysis of tumour cells,2 ,25 in this study we show decreased level of perforin in PBMC of patients with MM. Therefore, our results suggest limited cytotoxic potential of peripheral lymphocytes in patients with MM.
Also in support of suppressed lymphocyte cytotoxicity in patients with MM is our finding of decreased IFN-γ production from these cells in patients with MM compared with HCs, considering that the secretion of this cytokine, which constitutes the other major, immunoregulatory role of lymphocytes, is suppressed in MM.23
It is well known that perforin synthesis is regulated transcriptionally and that STAT-1, STAT-4 and STAT-5 represent important regulators of this molecule transcription.26 Now we show decreased protein level of these signalling molecules in PBMC of patients with MM that could be in association with decreased perforin level, shown in this study. So these STAT molecules via perforin regulate the ability of cytotoxic lymphocytes to kill.
It is well known that in lymphocytes IRF-1 transcription factor via STAT-1 is involved in the regulation of IFN-γ production, as well as in the synthesis of the lytic proteins.27 We show for the first time decreased mRNA level of IRF-1 in PBMC of patients with MM compared with HCs that is, in this study, followed by decreased STAT-1 level, as well as impaired perforin and IFN-γ expression in PBMC of these patients.
NKG2D, a potent activating receptor expressed on virtually all NK cells and CTL, determines capacity of these cells against tumours.9 In this study, we show that patients with MM have a significant decrease in NKG2D expression on total PBL, as well as on NK cells contrary to its expression on CTL. These alterations may cause impairments in effector functions of lymphocytes in patients with MM and also may suggest a significant role of NKG2D receptor in the regulation of lymphocyte activities.
We show for the first time significantly lower level of DAP10, NKG2D adaptor protein, in PBMC of patients with MM compared with HCs. It is well known that human NK cells express only one NKG2D splice variant protein, NKG2D-L, which is mainly engaged with DAP10 for its signalling.10 Recently, it has been documented that without DAP10 NKG2D is retained in the cytoplasm of the cell and degraded in the lysosome.28 Reduced level of DAP10 could be in association with decreased expression of NKG2D receptor on PBL, as well as on NK cells in patients with MM, shown in this study. Thus, reduced expression of DAP10 adaptor molecule might also lead to impaired effector functions of lymphocytes in response to tumour cells.
By analysing the expression of CD158a and CD158b receptors that belong to group A haplotype of KIR family with their main role in NK cell inhibition,29 we show that there is no significant difference in the expression of these receptors on PBL, NK cells, as well as on CTL between patients with MM and HCs. The function of KIRs in regulating T cell response has been studied to some extent. Most studies have been performed with KIR+CD8+ T cells, and it is clear that inhibitory KIRs can dampen CTL response, whereas activating KIRs can enhance effector functions of these cells.30 However, little is known about the role of human leucocyte antigen (HLA) class I molecules in shaping the KIR repertoire and function of CTL. Specifically, it is not known whether these molecules have a KIR-dependent educational effect on human CTL, as it has on NK cells.
As the data about the role of KIR phosphatase, SHP-1, in the regulation of human lymphocyte activities are still very heterogeneous, in accordance with our results of no significant difference in inhibitory KIRs expression, we show for the first time that there is no significant difference in the level of SHP-1 in PBMC between patients with MM and HCs. It could say that activating NK cell receptors and their signalling molecules are the main targets of tumour-mediated suppression.
By considering the importance of MDSC in the inhibition of anti-tumour immune response in this study, we show for the first time that patients with MM in both investigated groups, M1a+M1b and M1c, have significantly increased percentage of CD14+HLA-DR− cells in peripheral blood compared with HCs. It is shown that factors released by melanoma cells induce phenotypic changes in monocytes, leading them to resemble immature CD14+HLA-DR− monocyte MDSC found to be expanded in patients with advanced melanoma.17 ,31 However, in this study we did not find significantly higher prevalence of these immunosuppressive cells in M1c group of patients with MM with worse clinical outcome and prognosis compared with M1a+M1b group. Some authors showed that the percentage of CD14+HLA-DR−/low cells in peripheral blood increases gradually in parallel with melanoma progression and that the prevalence of these cells may be a determinant for predicting the prognosis of patients.32 ,33 However, contradictory reports exist.34 Furthermore, it is well known that MDSC by its membrane-bound transforming growth factor (TGF)-β1 impair NK cell development, as well as IFN-γ production and cytotoxicity of these cells against tumours.35 Additionally, TGF-β1 by down-regulating activating NKG2D receptor expression on NK cells impairs their cytotoxicity.36 Therefore, the down-regulation of NKG2D receptor on PBL and NK cells and its adaptor protein, DAP10, as well as the alterations in lymphocyte effector functions found in this study could be the consequence of immunosuppressive effect of TGF-β1 produced by substantial population of CD14+HLA-DR− immunosuppressive cells. So, the therapeutic targeting of these cells leading to their decrease may support the production of more potent NK cells and CTL against MM.37
The alterations in various signalling molecules, as well as an increase in the percentage of immunosuppressive MDSC in patients with MM shown in this study, may give new insight into the mechanisms of impaired cytotoxic and immunoregulatory function of PBL in these patients. Furthermore, these molecules and cells could represent biomarkers of impaired lymphocyte functions, indicating cancer-associated immunosuppression that may facilitate tumour progression. Moreover, they can be used as parameters for evaluation of the immunotherapy effects in patients with MM.
Take home messages
In patients with metastatic melanoma (MM), decreased expression of phosphorylated signal transducers and activators of transcription (pSTAT)-1/pSTAT-4/pSTAT-5 and interferon-regulating transcription factor-1 signalling molecules is associated with impaired perforin-mediated cytotoxicity and interferon-γ production of peripheral blood lymphocytes (PBLs).
Patients with MM have decreased expression of activating NKG2D receptor on PBL and natural killer cells and low level of its DAP10 signalling molecule contrary to no changes in the expression of CD158a and CD158b inhibitory killer immunoglobulin-like receptors.
Patients with MM have increased percentage of immunosuppressive CD14+HLA-DR− myeloid-derived suppressor cells (MDSCs) in PBL.
Decreased level of signalling molecules, as well as increased percentage of immunosuppressive MDSC in patients with MM, may represent biomarkers of impaired lymphocyte effector functions, indicating cancer-associated immunosuppression that may facilitate tumour progression.
Acknowledgments
The authors thank Mrs Jasna Popović Basić for excellent technical work and Mrs Dušica Gavrilović for extensive statistical analyses.
References
Supplementary materials
Abstract in Serbian
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
- Abstract in Serbian - Online abstract
Footnotes
Handling editor Des Richardson
Contributors KMM, RD, VJ and GK conceived the idea of the study. All authors contributed to the design of the research and were involved in data collection. KMM and TS-R performed flow cytometric and western blot analyses. KMM performed RT-PCT and ELISA analyses. KMM, TS-R and VJ analysed the data, prepared the manuscript, references and edited the final version of the manuscript. All authors approved the final version of the manuscript.
Funding This work was supported by the grants of the Ministry of Education, Science and Technological Development of the Republic of Serbia, numbers III41031 and 175056.
Competing interests None declared.
Patient consent Obtained.
Ethics Ethical committee of the Institute of Oncology and Radiology of Serbia..
Provenance and peer review Not commissioned; externally peer reviewed.