The first 3 authors and the last 2 authors contributed equally to this article. The first 3 authors and the last 2 authors initiated the work, designed the experiments, and wrote the article; the first 7 authors performed the experiments and statistical analyses; and the eighth author provided critical suggestions for this article.
Mantle cell lymphoma (MCL) is an incurable B-cell malignancy, and patients with this disease have the poorest prognosis among all patients with B-cell lymphomas. The signaling pathways that trigger MCL escape from immune surveillance are unclear. Because Toll-like receptors (TLRs) initiate innate and adaptive immune responses against invading pathogens, the authors investigated the impact of TLR signaling in MCL cells.
TLR expression was examined in MCL cell lines and in primary tumors. The examination focused on TLR4 and its ligand lipopolysaccharide (LPS) on MCL cells and their function on MCL proliferation and immune evasion.
MCL cells expressed multiple TLRs, and TLR4 was among the highest expressed molecules. The activation of TLR4 signaling in MCL cells by LPS induced MCL proliferation and up-regulated the secretion of cytokines like interleukin-6 (IL-6), IL-10, and vascular endothelial growth factor (VEGF). LPS-pretreated MCL cells inhibited the proliferation and cytolytic activity of T cells by secreted IL-10 and VEGF, and neutralizing antibodies against these cytokines restored their functions. Similar results were observed in TLR4-positive/myeloid differentiation 88 (MyD88)-positive primary lymphoma cells but not in TLR4-positive/MyD88-negative primary lymphoma cells from patients with MCL. Knockdown of TLR4 on MCL cells abrogated the effect of LPS on MCL cells in term of cell growth or secretion of the cytokines and evasion of the immune system.
Mantle cell lymphoma (MCL) is an incurable B-cell neoplasm that constitutes about 6% of non-Hodgkin lymphoma.1 It is characterized by a specific t(11;14)(q13;q32) translocation, leading to the overexpression of cyclin D1.2 Most patients manifest with advanced-stage disease at initial diagnosis, and the prognosis is poorest among patients with B-cell lymphoma, with a short median survival of approximately 3 to 5 years.3, 4 Although progress has been made in MCL treatment in the past decade, persistent remissions usually are not achieved, and the treatment of patients with relapsed or refractory MCL is still challenging. Therefore, new insights into the biology of MCL and novel treatment options are needed to improve the clinical outcome in MCL patients.
Dysregulated chronic inflammation can cause oncogenesis and often provides growth and angiogenic factors, which enhance proliferation and metastasis of tumor cells.5 Many of these events are mediated by direct recognition of pathogen-associated molecular patterns (PAMPs).6 Toll-like receptors (TLRs) are type-I transmembrane receptors that are composed of an ectodomain with leucine-rich repeats and a cytoplasmic domain called the Toll/interleukin-1 receptor (TIR) domain.6 TLRs directly recognize a series of PAMPs, including foreign pathogens from bacteria, virus, and fungi; induce the release of inflammatory cytokines; and link innate and adaptive immune responses.7, 8 Recently it was reported that, TLRs, which normally have expression restricted to immune cells or nonimmune cells, also are expressed in tumors and play important roles in tumor biology.9, 10 In pituitary tumors and glioblastoma, the TLR4 ligand lipopolysaccharide (LPS) inhibited the growth of TLR4-positive tumor cells11 and induced antitumor effects.12 Another study demonstrated that the activation of TLR4 on tumor cells in vitro inhibited subsequent tumor growth in vivo in a rat prostate cancer model and in a B16 murine melanoma model.13 However, phase 2 clinical trials in lung cancer indicated that the administration of LPS did not have any antitumor effects.14 Conversely, other evidence indicated that triggering of TLR4 on human myeloma and ovarian cancer promotes tumor proliferation and induces chemoresistance.15-17 Activation of TLR5 by flagellin reportedly suppresses cell proliferation and tumor growth in breast cancer18 but promotes the proliferation of gastric cancer cells.19 Therefore, the functional roles of TLR signaling in cancer cells remain to be further elucidated. The objective of the current study was to investigate the role of TLRs and their signaling in MCL cells. We observed that TLRs, especially TLR4, were highly expressed on human MCL cells and that TLR4 signaling was functional in the cells.
MATERIALS AND METHODS
Patients and Cell Lines
The sources of primary MCL cells for this study included bone marrow aspirates and peripheral blood samples obtained from patients with newly diagnosed and relapsed MCL after obtaining their informed consent. Among these samples, 7 bone marrow aspirates were derived from 4 patients with newly diagnosed MCL and 3 patients with relapsed MCL who had bone marrow involvement. Samples of peripheral blood that contained leukemic MCL cells were collected from 5 patients with relapsed MCL who also had bone marrow involvement. This study was approved by the Institutional Review Board at The University of Texas M. D. Anderson Cancer Center.
Mononuclear cells were separated by Ficoll-Hypaque density centrifugation, and MCL cells were isolated using anti-B-lymphocyte antigen (anticluster of differentiation 19 [anti-CD19]) magnetic microbeads (Miltenyi Biotec, Auburn, Calif). Purified MCL cells were cryopreserved in liquid nitrogen until use. Four human MCL lines (SP53, Mino, Granta 519 [G519], and Jeko-1; American Type Culture Collection, Manassas, Va) were maintained in RPMI-1640 medium (Sigma Chemical Company, St. Louis, Mo) supplemented with 10% fetal bovine serum (Atlanta Biologicals, Lawrenceville, Ga), penicillin (100 U/mL; Sigma Chemical Company), and streptomycin (100 μg/mL; Sigma Chemical Company). The bone marrow stromal cell line S17 was kindly provided by Dr. Richard Eric Davis (The University of Texas M. D. Anderson Cancer Center, Houston, Tex) and used in coculture with primary MCL cells to sustain their survival ex vivo.
The cells were lysed, and total RNAs were isolated from the cells using an RNeasy Mini Kit (QIAGEN, Valencia, Calif) according to the manufacturer's instructions. Combinational DNAs were generated using the M-MuLV Reverse Transcription System (Fermentas Inc., Hanover, Md) and then were amplified by polymerase chain reaction (PCR). The TLR primers that we used are listed in Table 1. PCR conditions consisted of 35 cycles at 94°C for 30 seconds, 55°C for 40 seconds, and 72°C for 50 seconds for denaturing, annealing, and extension, respectively, followed by an extension at 72°C for 7 minutes. Glyceraldehyde 3-phosphate dehydrogenase was used as a control. The PCR products were analyzed on 1.5% agarose gels.
Table 1. Primer Sequences of Toll-Like Receptors for Reverse Transcriptase-Polymerase Chain Reaction Analysis
Phycoerythrin-labeled antihuman TLR4 and isotype-matched control antibody were obtained from eBioscience (San Diego, Calif). Data were acquired with a flow cytometer (FACS Calibur; BD Biosciences, Sparks, Md).
TLR4 Short-Hairpin RNA Transfection of Tumor Cells by Lentivirus
MCL cells were transfected using human TLR4 or control short-hairpin RNA (shRNA) lentiviral particles (Santa Cruz Biotechnology, Santa Cruz, Calif) according to the manufacturer's protocol to knockdown TLR4 expression in the cells. In this study, these untreated cells were divided into wild-type (wt) cells, control shRNA-transfected cells, and TLR4-specific shRNA-transfected cells, respectively.
Cells (1 × 104/100 μL per well) were seeded into 96-well, flat-bottom tissue culture plates (Corning Inc., Corning, NY) and cultured for 4 days. Primary MCL cells were cocultured with the irradiated bone marrow stromal cell line S17. LPS (0-1000 ng/mL; InvivoGen; San Diego, Calif) was added to the culture. At the end of each treatment, 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS) reagent (Promega, Fitchburg, Wis) was added to the culture medium for 2 hours, and the absorption was qualified using an automatic microplate reader (Biotek, Winooski, Vt) at 490 nm.
Preparation of Tumor-Cell Culture Medium
MCL cells were cultured in RPMI-1640 complete medium with or without LPS stimulation. Forty-eight hours later, supernatants were harvested, filtered, and concentrated 10-fold using an Amicon Ultra Filter (Millipore, Bedford, Mass). Concentrated tumor-cell culture medium (TCCM) was divided into aliquots and stored at −80°C until use. Medium control was prepared from freshly prepared RPMI-1640 complete medium in a manner similar to that used for TCCM preparation.
Western Blot Analysis
Western blot analysis was used to detect protein expression in MCL cells, as described previously.20 Antihuman myeloid differentiation 88 (MyD88), nuclear factor κB (NF-κB), c-Jun N-terminal kinase (JNK), extracellular signal-regulated kinase (ERK), the p38 mitogen activated protein kinase (p38), protein kinase B (Akt), phosphorylated p38 (p-p38), p-JNK, p-ERK, p-NF-κB, and p-Akt (all from Cell Signaling Technology, Danvers, Mass), TLR4 (Santa Cruz Biotechnology), and β-actin (Sigma Chemical Company) monoclonal antibodies were used.
Enzyme-Linked Immunosorbent Assay
Enzyme-linked immunosorbent assays for interleukin 6 (IL-6), IL-10, IL-18, vascular endothelial growth factor (VEGF), or transforming growth factor beta (TGF-β) were used to measure the secreted cytokines. Cell culture supernatants were collected on day 2, and the amounts of secreted IL-6, IL-10, IL-18, VEGF, and TGF-β in the supernatants were quantified using a commercially available enzyme-linked immunosorbent assay kit (R&D Systems, Rockland, Md).
T-Cell Proliferation Assay
Allogeneic CD3-positive T cells were purified from peripheral blood mononuclear cells (PBMCs) from healthy donors using magnetic cell sorting (Miltenyi Biotec, Cambridge, Mass). CD3-positive T cells were labeled with 56-carboxyfluorescein diacetate succinimidyl ester (CFSE) (5 μM; Invitrogen, Carlsbad, Calif) for 10 minutes at 37°C. After washing, T cells (1 × 105/100 μL per well) were seeded into 96-well, U-bottomed tissue culture plates (Corning Inc.) for 5 days in 5% CO2 in Aim-V medium supplemented with 10% pooled human AB serum (T-cell medium). T cells were activated with anti-CD3 and CD28 antibodies. Medium or TCCM from MCL cells were added to the cultures. Flow cytometric analysis was used to detect the dilution of CFSE.
Generation of Tumor-Reactive, Alloantigen-Specific, Cytotoxic T-Lymphocyte Lines and Cytotoxicity Assay
Allogeneic CD3-positive T cells were cocultured in T-cell medium with irradiated SP53 or G519-wt, respectively. After 7 days of coculture, CD3-positive T cells were harvested and restimulated with newly irradiated tumor cells. The cultures were fed with fresh T-cell medium containing recombinant IL-2 (10 IU/mL), IL-7 (5 ng/mL), and IL-15 (5 ng/mL; all from R&D Systems). After at least 4 repeated cycles of in vitro restimulation, T-cell lines were generated. The T-cell lines were expanded in T-cell medium that contained recombinant IL-2, IL-7, and IL-15 for 2 weeks and were subjected to functional tests. A standard, 4-hour chromium-51 release assay was performed to measure cytolytic activity of the T-cell lines with target cells, including the stimulatory MCL cell lines SP53 and G519, and primary tumor cells isolated from patients with MCL, as described previously.20
The Student t test was used to compare various experimental groups. A P value < .05 was considered statistically significant. Unless otherwise indicated, the values provided are means and standard deviations (SDs).
Toll-Like Receptors Are Expressed on Human Mantle Cell Lymphoma Cells
We first analyzed the expression of TLR1-10 in human primary MCL cells and cell lines by reverse transcriptase (RT)-PCR. Surprisingly, all tested MCL cells expressed multiple TLRs (Fig. 1A). We focused our study on TLR4, because it has a high level of mRNA expression among all TLRs. Consistent with the RT-PCR results, Western blot analysis (Fig. 1B) and flow cytometry (Fig. 1C-D) also revealed that total and surface TLR4 proteins were present in human MCL cell lines and primary MCL cells from 11 patients, whereas the bone marrow stromal cell line S17 expressed little TLR4 surface protein. The levels of TLR4 were significantly higher in primary MCL cells than in normal PBMCs (P < .01) (Fig. 1D) or B cells (P < .01). An important contributing protein in the TLR4 signaling cascade, MyD88, also was expressed in MCL cell lines and in 2 of 4 primary cells from patients with MCL (Fig. 1B). The widespread expression of TLR4 on human MCL cells indicated that TLR4 signaling may play a role in tumor biology.
TLR4-Specific Ligand Lipopolysaccharide Promotes Proliferation of Human Mantle Cell Lymphoma Cells
Next, we examined whether TLR4 signaling was functional in human MCL cells. Upon stimulation with TLR4 ligand LPS at different concentrations, the proliferation of SP53, Jeko-1, Mino, and G519 cells was increased in a dose-dependent manner (Fig. 2A). LPS also promoted the proliferation of TLR4-positive/MyD88-positive primary MCL cells but not TLR4-positive/MyD88-negative primary MCL cells (P < .01 compared with medium control) (Fig. 2B). However, LPS did not increase the proliferation of S17 cells or normal PBMCs from blood donors (Fig. 2C).
To confirm the role of TLR4 in LPS-induced MCL proliferation, we knocked down tlr4 gene expression in SP53 and G519 cells by using TLR4-specific shRNA lentiviral particles. Upon transfection, TLR4-specific shRNA reduced the expression of TLR4 total protein (Fig. 2D) and surface protein (Fig. 2E) by 78% in SP53 cells and by 66% in G519 cells, respectively, whereas the control shRNA did not. SP53-knockdown and G519-knockdown cell lines had a reduced response to LPS stimulation compared with shRNA control cells (from P < .05 to P < .01) (Fig. 2F).
Lipopolysaccharide Up-Regulates the Secretion of Inflammatory Cytokines by Mantle Cell Lymphoma Cells
We next investigated whether TLR4 signaling in MCL could activate the expression of cytokines. It is noteworthy that LPS stimulation significantly up-regulated the secretion of IL-6, IL-10, and VEGF in MCL cell lines (P < .01) (Fig. 3A) and in TLR4-positive/MyD88-positive primary MCL cells (P < .01) (Fig. 3B, Patient 1), but not in TLR4-knockdown cell lines or TLR4-positive/MyD88-negative primary MCL cells (Fig. 3B, Patient 3). MCL cells did not express or secret indoleamine-2, 3-dyoxigenase (IDO), TGF-β, or IL-18 with or without LPS stimulation (data not shown).
TLR4 Signaling in Mantle Cell Lymphoma Cells Involves Mitogen-Activated Protein Kinase, Nuclear Factor κB, and Phosphoinositide 3-Kinase Pathways
To investigate the molecular pathways involved in MCL cells in response to LPS stimulation, mitogen-activated protein kinase (MAPK), p-NF-κB, and phosphoinositide 3-kinase (PI3K) pathway kinases were analyzed. All MCL cells that we examined constitutively expressed p-p38, p-JNK, p-ERK, p-NF-κB, and p-Akt (Fig. 4). After short stimulation with LPS, the levels of p-p38, p-JNK, p-ERK, p-NF-κB, and p-Akt were increased significantly in control cells but not in TLR4-knockdown MCL cells.
TLR4 Activation in Human Mantle Cell Lymphoma Cells Facilitates Immune Evasion
To examine the impact of TLR4-activated MCL cells on the immune system, purified and CFSE-labeled, CD3-positive T cells from blood donors were activated by anti-CD3 and CD28 antibodies in cultures in the presence or absence of MCL-derived TCCM., TCCM from wild-type, unstimulated MCL cells suppressed T-cell proliferation (P < .05,compared with T-cell medium control), as indicated in Figure 5A; whereas TCCM from wild-type, LPS-stimulated MCL cells displayed even stronger inhibition of T-cell proliferation (P < .01 compared with T-cell medium control; P < .05 compared with TCCM from unstimulated MCL cells). Similarly, TCCM from LPS-stimulated, TLR4-positive/MyD88-positive primary MCL cells, but not TLR4-positive/MyD88-negative primary MCL cells, displayed strong inhibitory activities on T-cell proliferation (Fig. 5B). This was not caused by LPS contamination in the TCCM, because the addition of LPS to control medium did not inhibit T-cell proliferation (Fig. 5C). Furthermore, although TCCM from unstimulated wild-type or TLR4-knockdown MCL cells displayed similar inhibitory activities, TCCM from LPS-stimulated TLR4-knockdown MCL cells had a much weaker inhibitory effect on T-cell proliferation compared with TCCM from LPS-stimulated wild-type or control cells (P < .05) (Fig. 5A). These results further confirmed that LPS contamination in TCCM was not responsible for inhibited T-cell proliferation. Collectively, these findings indicated that the suppression of T-cell proliferation was induced by soluble factors in TCCM of unstimulated MCL cells and LPS-stimulated MCL cells dependent on TLR4 and MyD88 signaling. Because we demonstrated that MCL cells secreted IL-6, IL-10, and VEGF and that LPS up-regulated their secretion through TLR4 on MCL cells, neutralizing antibodies against these cytokines were used. Figure 5D indicates that neutralizing IL-10 or VEGF partially reversed the inhibition of T-cell proliferation by LPS-stimulated TCCM. Furthermore, the combination of both antibodies to neutralize IL-10 and VEGF further reduced the inhibition of T-cell proliferation by LPS-stimulated TCCM. IL-6 had no effect on the inhibition of T-cell proliferation, because the addition of IL-6–neutralizing antibodies to LPS-stimulated TCCM did not increase T-cell proliferation (data not shown). These findings indicated that MCL cells inhibited T-cell response by secreting the cytokines, and LPS-stimulated MCL cells had a stronger inhibitory effect on T-cell response.
Next, we explored whether the activation of MCL TLR4 affected the sensitivity of cells to cytotoxic T lymphocyte (CTL) attack. MCL-reactive allogeneic T-cell lines were generated by coculturing CD3-positive T cells with irradiated wild-type or primary MCL cells and were assayed for their cytolytic activity against their target cells. The killing of LPS-stimulated (wild-type or vector control) MCL cells was inhibited significantly compared with that of unstimulated MCL cells (P < .01), as illustrated in Figure 6A; however, the killing of LPS-stimulated or unstimulated TLR4-knockdown MCL cells was not different. Similarly, LPS-stimulated, TLR4-positive/MyD88-positive primary MCL cells, but not TLR4-positive/MyD88-negative primary MCL cells, were more resistant to the killing by T cells (P < .01 compared with unstimulated cells) (Fig. 6B). Neutralizing antibodies against IL-10, or VEGF, or a combination of both could partially reverse LPS-induced inhibition of the killing of MCL cells by T cells (from P < .05 to P < .01 compared with LPS-stimulated and control immunoglobulin G-treated cells), as illustrated in Figure 6C. These results indicated that LPS-stimulated MCL cells were less sensitive to the killing of tumor-reactive CTLs.
The current study demonstrates that TLR4 expressed on MCL cells may contribute to tumor progression. We demonstrated that MCL cells expressed multiple TLRs, especially TLR4. We focused our study on TLR4 and used the LPS ligand of TLR4 to explore the activity of TLR4 signaling in MCL cells. Our results indicate that the activation of TLR4 by LPS in MCL cells induces proliferation in MCL cell lines and in TLR4-positive/MyD88-positive primary MCL cells, but not in TLR4-positive/MyD88-negative primary MCL cells. Furthermore, knockdown of TLR4 in MCL cells retarded LPS-induced tumor cell growth.
In the tumor microenvironment, tumor-derived factors, including IL-6, IL-10, VEGF, and TGF-β, may bias the induction of immune response and lead to a state of tolerance against tumors.21 Moreover, it has been demonstrated that the immune-modulatory enzyme IDO is an important mediator of immune evasion.22 It also has been reported that IL-18, a new member of the IL-1 cytokine super family, plays an important role in immune responses.23 In our study, MCL cells produced IL-6, IL-10, and VEGF; and the activation of TLR4 signaling in MCL cells up-regulated the secretion of these cytokines. However, IDO, TGF-β, or IL-18 secretion was not detected in MCL supernatants, and LPS stimulation could not induce their secretion. These cytokines may contribute to the inhibition of T-cell response induced by LPS-pretreated MCL cells.
Because the current findings suggested that TLR4 ligand LPS may facilitate MCL cell evasion of the immune surveillance, we also examined the effects of LPS-pretreated MCL cells in T-cell proliferation and CTL function. The results indicated that T-cell proliferation induced by anti-CD3 and CD28 antibodies was inhibited in TCCM of unstimulated MCL cells and more so in TCCM of LPS-stimulated MCL cells. Furthermore, LPS-pretreated MCL cells were more resistant than untreated cells to CTL-mediated killing. Neutralizing antibodies against IL-10 and/or VEGF could partially restore T-cell proliferation or CTL cytolytic activity. Similarly, knocking down TLR4 on MCL cells could also partially restore their sensitivity to CTLs. The inhibition of T-cell proliferation and CTL function by LPS-stimulated MCL cells also was observed in TLR4-positive/MyD88-positive primary MCL cells, but not in TLR4-positive/MyD88-negative primary MCL cells. Taken together, these findings suggest that TLR4 ligand activation may facilitate MCL cell evasion of immune surveillance, which is MyD88-dependent.
Recent studies have indicated that TLRs play a role in carcinogenesis, but different TLRs may play different roles, having either antitumor or protumor activities.24, 25 Similar to our findings in MCL, He et al demonstrated that TLR4 signaling promoted immune escape of human lung cancers,26 and Szajnik et al demonstrated that TLR4 signaling induced by LPS or paclitaxel supported ovarian cancer progression and chemoresistance.15 However, Nunez et al reported that TLR4-activated tumor cells produced interferon-β and enhanced induction of the antitumor immune response.27 The discrepancy between these observations is not obvious but may be attributed to tumor heterogeneity and the secretion of immunosuppressive (IL-10 and VEGF) or proimmune (interferon-β) cytokines upon TLR4 signaling in the tumors.
In conclusion, our current results indicate that MCL cells express TLRs and that recurrent bacterial infections in patients may activate MCL-associated TLR4 through LPS, which promotes tumor growth and shields MCL cells from immune surveillance. Therefore, TLR4 signaling molecules may be novel therapeutic targets for cancer therapy in patients with MCL. It is noteworthy that targeting TLR4 signaling through the TLR4 Toll/IL-1 receptor domain-derived decoy peptides has been developed.28 Sheedy et al demonstrated that targeting of the proinflammatory tumor suppressor PDCD4 (programmed cell death 4) by microRNA-21 may negatively regulate TLR4.29 Lysakova-Devine et al demonstrated that a viral inhibitory peptide of TLR4 specifically inhibited TLR4 by directly targeting MyD88 adaptor-like and TIR-domain–containing, adapter-inducing interferon-β (TRIF)-related adaptor molecules.30 Tang et al also identified negative regulation of TLR4 through targeting of the key adaptor molecule MyD88 by microRNA-155.31
We thank The University of Texas M. D. Anderson Cancer Center Lymphoma Tissue Bank for providing patient samples.
This work was supported by National Cancer Institute grants R01 CA163881, R01 CA138402, R01 CA138398, and P50 CA142509; by Leukemia and Lymphoma Society Translational Research Grants; the Multiple Myeloma Research Foundation; the Commonwealth Foundation for Cancer Research; and the Center for Targeted Therapy of The University of Texas M. D. Anderson Cancer Center. This work also was supported by Special Funds for International Cooperation from the National Natural Science Foundation of China (81120108018); the Major Research Plan of the Chinese National Natural Science Foundation (91029740); the Key Program of Natural Science Foundation of Zhejiang, China (2009C03012-2); and the Cultivation Program for Distinguished Talented Persons of Health of Zhejiang, China. The University of Texas M. D. Anderson Cancer Center Lymphoma Tissue Bank is supported by the National Institutes of Health Lymphoma Specialized Programs of Research Excellence (SPORE) (grant P50CA136411) and by the Fredrick B. Hagemeister Research Fund.