Background
Pancreatic cancer is one of the most common gastrointestinal malignancies worldwide and is the fifth leading cause of cancer-related deaths [
1]. It is characterized by rapid progression and intrinsic and acquired drug resistance [
2]. Pancreatic cancer is difficult to diagnose in its early stages, therefore the rate of surgical resection is low (<20%) [
3] and the five-year survival rate is poor (<6%) [
1]. The progression of pancreatic cancer is affected by the surrounding microenvironment which largely consists of cancer associated fibroblasts and infiltrating inflammatory cells [
4‐
6], including natural killer (NK) cells.
NK cells are key components in the innate immune system, acting as the first line of defense in the body. They are a type of cytotoxic lymphocyte that mediate a pro-inflammatory response and are closely associated with the progression of cancer, including pancreatic cancer [
7]. Evidence shows that NK cells not only kill target cancer cells directly, without prior sensitization [
8,
9], but also by binding to specific surface ligands expressed on the surface of cancer cells, such as major histocompatibility complex class I (MHC class I) molecules [
10]. The effectiveness of NK cells as anticancer agents depends not only on the number of circulating, activated NK cells [
11] but also on the expression of activating surface receptors, inhibitory surface receptors, cytotoxic granules and cytokines. NK cell dysfunction has been reported in several cancers, such as pancreatic cancer [
7], gastric cancer [
12], and colorectal cancer [
13]. Although cancer cells have been shown to induce NK cell dysfunction
via downregulation of specific activating surface receptors (e.g. NKG2D), natural cytotoxicity receptors (NCR) [
14], cytotoxic granules (e.g. Perforin and Granzyme B) [
15], decreased secretion of cytokines (e.g. TNF-α and IFN-γ) [
16] and upregulation of inhibitory surface receptors (e.g. KIR3DL1 and KIR2DL1/DS1) [
17,
18], the underlying mechanisms are largely unknown.
Several studies have suggested that matrix metallopeptidase 9 (MMP-9), a 92-kDa type IV collagenase which is secreted by mesenchymal stem cells (MSC), can significantly downregulate the cytotoxicity of NK cells
in vitro by targeting T cells [
19,
20]. Indoleamine 2, 3-dioxygenase (IDO) has also been reported to play a role in MSC-mediated immunosuppression [
21] by inhibiting NK cell accumulation and suppressing NK cell function [
22]. This suggests that MMP-9 and IDO may play similar roles in tumor-induced NK cell dysfunction. Decreased infiltration of inflammatory cells into the tumor microenvironment has been associated with increased expression of COX-2 [
23], which is known to promote tumor growth
via its major product prostaglandin E2 (PGE2) in a T cell or NK cell-dependent manner [
24,
25]. Together, MMP-9, IDO and PGE2 are potent effectors in the interaction between pancreatic cancer cells and NK cells. The mechanism by which they promote NK cell dysfunction is the focus of this investigation.
Methods
Antibodies and reagents
Anti-human CD3-FITC/CD16 + 56-PE mixed antibody were purchased from Beckman Coulter (Brea, CA, USA). Anti-human CD16-FITC, NKG2D-PE, NKp44-APC, DNAM-1-FITC, NKp46-Alexa Fluor 647, NKp30-PE, KIR3DL1-FITC, KIR2DL1/DS1-PE, NKp30-APC, Perforin-PE and Granzyme B-FITC antibodis were all purchased from BioLegend (San Diego, CA, USA), as well as Fixation Buffer, Wash Buffer, Annexin V binding buffer, Alexa Fluor 647 Annexin V and 7-AAD Viability Staining Solution. The MPP-9 and IDO inhibitors were 1-Methyl-DL-tryptopan (1-MT; Sigma-Aldrich, St. Louis, MO, USA) and Tissue inhibitor of metalloproteinases 1 (TIMP-1; PeproTech, Rocky Hill, NJ, USA). Human NK Cell Isolation Kit was purchased from Miltenyi Biotec (Auburn, CA, USA) and ELISA kits were purchased from Abcam (Cambridge, MA, USA). Trizol reagent and PrimeScript RT Master Mix (Perfect Real Time) were both obtained from TaKaRa (Shiga, Japan) and Power SYBR Green PCR Master mix was purchased from Applied Biosystems (Carlsbad, CA, USA).
NK cell isolation
Fresh peripheral blood samples from healthy volunteers were provided by Jiangsu Province Blood Center (Gu Jian, China). Peripheral blood mononuclear cells (PBMC) were isolated by Ficoll-Hypaque density gradient centrifugation. NK cells were selected from the PBMCs by negative magnetic selection. The purity of the NK cells was >92%, as quantified by multicolor flow cytometry (Gallios; Beckman Coulter).
Cells and cell culture
The normal human pancreatic ductal cell line hTERT-HPNE and pancreatic cancer cell lines SW1990 and BxPc-3 were obtained from the American Type Culture Collection (ATCC; Rockville, MD, USA). The hTERT-HPNE cells were cultured as previously described [
26]; SW1990 and BxPc-3 cell lines were cultured in DMEM supplemented with 10% FBS, penicillin (100 U/mL) and streptomycin (100 μg/mL). The myelogenous leukemia K562 cell line (ATCC) was cultured in RPMI 1640 supplemented with 10% FBS, penicillin (100 U/mL) and streptomycin (100 μg/mL). The NK-92 cell line was kindly donated by Professor Bin Gao and was cultured as previously described [
27]. Purified NK cells were cultured in 6-well plates (3 × 10
5 cells/well) in AIM-V media supplemented with 10% FBS, penicillin (100 U/mL), streptomycin (100 μg/mL) and interleukin 2 (IL-2; 100 U/mL; PeproTech, Rocky Hill, NJ, USA), either in absence or presence of hTERT-HPNE, SW1990 and BxPc-3 cells (5 × 10
5 cells/well). In order to investigate the roles of MMP-9 and IDO, NK-92 cells were co-cultured with SW1990 cells in 6-well plates (NK-92/SW1990 ratio: 3 × 10
5/5 × 10
5 cells/well) in the presence of 0.5 ug/ml TIMP-1 (a specific blocker for MMP-9) and/or 0.5 mM 1-MT (a specific blocker for IDO).
Flow cytometric analysis
NK and NK-92 cells, either cultured alone or co-cultured with normal or cancer pancreatic cell lines, were harvested after five days and divided into four tubes, labeled as T1, T2, T3 and T4. Firstly, T1, T2, T3 and T4 was respectively washed twice with PBS. Secondly, T1 was stained with anti-human CD16-FITC, NKG2D-PE, and NKp44-APC antibodies; T2 was stained with anti-human DNAM-1-FITC, NKp30-PE, and NKp46- Alexa Fluor 647 antibodies; T3 was stained with anti-human KIR3DL1-FITC, KIR2DL1/DS1-PE and NKp80-APC antibodies; T4 was added with 500 μl fixation Buffer and incubated in the dark at room temperature for 20 min, and then, T4 was washed twice with Wash Buffer and stained with anti-human Granzyme-B-FITC, and Perforin-PE antibodies. After incubating in the dark at room temperature for 15 min, the cells were washed twice with PBS. After staining, the tubes were incubated in the dark at room temperature for 15 min and washed twice with PBS. Lastly, all tubes were analyzed by multicolor flow cytometry. Data were analyzed using Kaluza software.
Apoptosis of NK cells
NK cells, either cultured alone or co-cultured with normal or cancer pancreatic cell lines, were harvested after three days and resuspended in 500 μL Annexin V binding buffer. The cells were stained with 5 μL Alexa Fluor 647 Annexin V and 7-AAD Viability Staining Solution and incubated in the dark at room temperature for 15 min before being analyzed by multicolor flow cytometry.
Apoptosis of K562 cells
NK and NK-92 cells, either cultured alone or co-cultured with normal or cancer pancreatic cell lines, were harvested after three days. K562 cells were treated with the harvested NK or NK-92 cells were in the presence of IL-2 (100 U/mL) for 2 h at different effector-to-target (E/T; NK/K562) ratios (1:1, 3:1, 9:1). The cells were then collected and stained with CD3-FITC/CD(16 + 56)-PE, incubated in the dark at room temperature for 15 min, washed 2 times with PBS and resuspended in 500 μL Annexin V binding buffer. The cells were stained with 5 μL Alexa Fluor 647 Annexin V and 7-AAD Viability Staining Solution and incubated in the dark at room temperature for a further 15 min. According to cells staining, cell subset which detected as CD3-/CD(16 + 56) + was NK cells, and another cell subset which detected as CD3-/CD(16 + 56)- was K562 cells. Percentage of apoptosis K562 cells before was analyzed by multicolor flow cytometry.
Enzyme-linked immunosorbent assay (ELISA)
The expression of some protein (IDO, MMP-9, GM-CSF, TNF-α, and IFN-γ) were detected by ELISA. The concentrations of MMP-9 and IDO in cell culture supernatants (NK cells cultured alone, hTERT-HPNE cells cultured alone or co-cultured with NK cells, SW1990 cells cultured aloneor co-cultured with NK cells; and NK-92 cells cultured alone or co-cultured with SW1990 cells in the absence or presence of TIMP-1 and/or 1-MT) were determined by specific ELISA kits. The concentrations of GM-CSF, TNF-α and IFN-γ in cell culture supernatants (NK cells cultured alone or co-cultured with hTERT-HPNE and SW1990 cells; and NK-92 cultured alone or co-cultured with SW1990 cells in the absence or presence of TIMP-1 and/or 1-MT) were also determined. All procedures were carried out according to the manufacturer’s protocols.
Quantitative real-time reverse-transcription polymerase chain reaction (qRT-PCR)
Total RNA was extracted from hTERT-HPNE or SW1990 cells cultured alone or co-cultured with NK cells using Trizol reagent and reverse-transcribed to cDNA using PrimeScript RT Master Mix (Perfect Real Time). The cDNA was amplified using Power SYBR Green PCR Master mix and the Step One Plus Real-Time PCR System (Applied Biosystems, Carlsbad, CA, USA). Each procedure was performed according to the manufacturer’s instructions. The sequences of gene specific primers for human MMP-9, IDO, COX-2, and β-actin were designed and purchased from Invitrogen (Carlsbad, CA, USA) using Primer Premier 5 and checked by Oligo 6 (Invitrogen). Expression levels were normalized relative to the cell line according to the formula 2-ΔΔCt, where Ct is the cycle threshold.
Statistical analysis
Statistical differences between data groups were determined by independent t-tests using SPSS 19.0 software (SPSS Inc., Chicago, IL, USA). Data are expressed as mean ± SD; P < 0.05 was considered statistically significant.
Discussion
In a previous clinical trial, we found that the development of pancreatic cancer was closely correlated with tumor immune escape, mediated by dysfunction in the circulating NK cells [
28]. In this study, we found that NK function was inhibited after exposure to pancreatic cancer cells
in vitro. This effect was greatest with SW1990 cells. Further investigation showed that MMP-9 and IDO were overexpressed in SW1990 cells and acted as negative factors in SW1990 cell-induced NK cell dysfunction.
The association between NK cell dysfunction and altered expression of surface receptors, cytotoxic granules and cytokines has previously been reported in other cancers, including melanoma and cervical cancer [
29,
30]. Activating surface receptors induce cell stress apoptosis by binding to specific ligands that are overexpressed on the surface of stressed cells, including cancer cells. For example, MHC class I polypeptide-related sequence A/B (MICA/MICB) and UL16-binding protein (ULBP) bind to NKG2D [
31]; whereas cellular heparin or heparin sulfate proteoglycans bind to NCRs [
32]. The cytotoxic granules Perforin and Granzyme-B cooperatively combine to form a complex which is released into the cytoplasm of infected cells and tumor cells [
33]. Cytokines TNF-α, IFN-γ and GM-CSF are abundantly secreted by NK cells [
34] and influence the recruitment and function of other hematopoietic cells (such as T cells) in the early stage of innate immune responses [
35].
Our findings showed that most of the surface receptors, cytotoxic granules and cytokines discussed above were significantly altered in NK cells following exposure to pancreatic cancer cells. For example, lower expressions of NKG2D, NKp30 and NKp46; downregulation of Perforin and Granzyme-B; and decreased secretion of TNF-α and IFN-γ. These data demonstrated that pancreatic cancer cells suppress NK cells through a variety of mechanisms. However, increased secretion of GM-CSF in NK cells may counteract the inhibiting effect of pancreatic cancer cells on NK cells [
36].
Inhibitory surface receptors remain hyporesponsive in healthy bodies but are upregulated when the body suffers infection or cancer to enable activated NK cells to perform anti-tumor or anti-infection functions [
37]. Two inhibitory surface receptors, KIR3DL1 and KIR2DL1/DS1, were investigated in this study, however we found no significant alteration in their expression in NK cells that following exposure to pancreatic cancer cells.
Investigations into the mechanisms by which SW1990 cells might modulate NK cells revealed that SW1990 cells produced high levels of negative soluble mediators when co-cultured with NK cells, including MMP-9 and IDO. Subsequent inhibition of MMP-9 and IDO could partially restore NK cell function, suggested that these mediators promoted pancreatic tumor growth by suppressing NK cells.
MMP-9 promotes cancer progression through a variety of mechanisms. Previous studies have reported that MMP-9 decreased apoptosis in cancer cells [
38]; reduced proliferation and metastasis in cancer cells derived from MMP-9 deficient mice compared to those from wild-type mice [
39,
40]; promoted angiogenesis in two transgenic models of cancer progression [
38,
39]; inhibited T cell activity against tumors by enhancing IL-2Rα and TGF-β production [
41]; and could significantly suppress the cytotoxicity of NK cells in oral squamous cell carcinoma cell lines [
20]. In support of these studies, we showed that MMP-9 negatively influenced NK cell function through decreased expression of NKG2D, NKp30 and Perforin and inhibition of IFN-γ and TNF-α. Conversely, we showed that inhibition of MMP-9 largely restored the levels of NKG2D-, NKp30- and Perforin-positive NK-92 cells, the secretion of TNF-α and IFN-γ and the cytotoxicity on NK cells against myelogenous leukemia K562 cells. However, our investigation was unable to determine whether MMP-9 targeted NK cells directly or acted
via other pathways; for example, through promotion of IL-2Rα and TGF-β, which would in turn suppress NK cells; or
via stimulation of cytokines, such as interleukin-8 (IL-8), connective-tissue activating peptide-III (CTAP-III), platelet factor-4 (PF4) and growth-related oncogene-α (GRO-α), which have previously been shown to alter infiltration and migration of leukocytes [
42].
IDO catalyzes the first and rate-limiting step in the kynurenine pathway of tryptophan catabolism. Previous studies have suggested that IDO contributes to immunotolerance in patients with autoimmune diseases and chronic infections [
43,
44], and that IDO expressed by hepatocellular carcinoma-associated fibroblasts could induce NK cell dysfunction
in vitro[
16]. Further studies have shown that IDO can facilitate cancer cells to escape immune surveillance by NK cells in cervical cancer [
45] and by regulatory T cells in pancreatic and breast cancer [
46,
47]. Our data has shown that IDO may promote the SW1990 cell-induced NK cell dysfunction by reducing the proportion of NKG2D-, NKp30- and Perforin-positive NK-92 cells; and by downregulating the secretion of IFN-γ and the cytotoxicity of NK cells against K562 cells. These findings were similar to our results with MMP-9, and there are no obvious synergic effects of MMP-9 and IDO. Therefore, their relationship in relation to NK cell dysfunction warrants further study.
The blocking effect of MMP-9 and IDO was incomplete suggesting that there may be additional factors involved in SW1990 cell-induced NK cell dysfunction. These could include PGE2, IL-8 and B7-H1 which have been reported to be overexpressed and act as immunosuppressants by reducing the cytotoxicity of T cells and NK cells in several solid cancers [
48‐
50]. Increasing our understanding of the underlying immunosuppressive mechanisms and pathways could facilitate the development of immune-dependent therapies for patients with specific cancers.
Acknowledgements
This work was supported in part by the grants from the National Natural Science Foundation of China (81170336, 81272239, 81101802, 81001079), the Natural Science Foundation of Jiangsu Province (BK2011845), the Program for Development of Innovative Research Team in the First Affiliated Hospital of NJMU, the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD, JX10231801), and the research Special Fund For public welfare industry of health (201202007).
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
Y-PP, YZ and J-JZ carried out the studies, participated in the statistical analysis and drafted the manuscript. W-bL participated in the sample collection and statistical analysis. MT carried out Flow cytometric analysis. Z-PL carried out Enzyme-linked immunosorbent assay. J-SW carried out quantitative real-time reverse-transcription polymerase chain reaction. K-RJ, W-TG, J-LW and Z-KX participated in the sample collection. YZ participated in the design of this study. YM conceived of the study, and participated in its design and coordination and helped to draft the manuscript. All authors read and approved the final manuscript.