https://orcid.org/0000-0001-6408-4061
Figures
Abstract
Background
Immunotherapy is now considered as the new pillar in treatment of cancer patients. Dendritic cells (DCs) play an essential role in stimulating anti-tumor immune responses, as they are capable of cross-presenting exogenous tumor antigens in MHCI complexes to activate naïve CD8+ T cells. Analgesics, like non-steroid anti-inflammatory drugs (NSAIDs), are frequently given to cancer patients to help relieve pain, however little is known about their impact on DC function.
Methods
Here, we investigated the effect of the NSAIDs diclofenac, ibuprofen and celecoxib on the three key processes of DCs required for proper CD8+ cytotoxic T cell induction: antigen cross-presentation, co-stimulatory marker expression, and cytokine production.
Citation: Raaijmakers TK, van den Bijgaart RJE, Scheffer GJ, Ansems M, Adema GJ (2022) NSAIDs affect dendritic cell cytokine production. PLoS ONE 17(10): e0275906. https://doi.org/10.1371/journal.pone.0275906
Editor: Silke Appel, University of Bergen, NORWAY
Received: April 24, 2022; Accepted: September 26, 2022; Published: October 13, 2022
Copyright: © 2022 Raaijmakers et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: The author(s) received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Cancer patients often receive analgesics to treat acute or chronic cancer-related pain [1]. In many cases, non-opioids are sufficient to relieve pain, especially when the medication is taken regularly, and pain can be controlled and maintained [2]. Non-steroid anti-inflammatory drugs (NSAIDs) are a specific class of analgesics. NSAIDs are not recommended during chemotherapy as they can cover up a fever. Similarly, they are not advised closely before and after surgery because of their anti-coagulant effects [2]. However, their analgesic effect [3, 4], their effect on the immune system and on tumor progression deems them as interesting analgesic candidates for cancer patients. NSAIDs exert their function through inhibition of cyclooxygenase (COX) enzymes (COX1 and/or COX2). COX1 is a constitutively expressed enzyme, whereas COX2 is an inducible enzyme and increases upon cell activation during inflammatory processes, e.g. by pro-inflammatory cytokines, as well as damage-associated molecular patterns [5]. Non-selective NSAIDs inhibit the activity of both COX1 and COX2 (e.g. diclofenac and ibuprofen), while selective NSAIDs specifically target COX2 (celecoxib). COX enzymes convert arachidonic acid to prostanoids, including thromboxane and prostaglandins (PGE), in lipid droplets [6]. High COX expression levels in the tumor therefore often correlate with overproduction of PGE [7, 8]. The prostaglandin E2 (PGE2) lowers nociceptor thresholds [9], promotes tumor development and progression by evasion of apoptosis [10, 11], sustains proliferation [12, 13], induces angiogenesis [14–16], and promotes cancer cell adhesion, migration, and invasion [17]. Furthermore, PGE2 acts as a major mediator in inflammatory responses [18], mainly promoting immunosuppressive environments [19]. High COX levels in tumors [20–26] have been linked to lower survival rates [27, 28]. The use of NSAIDs in turn has been linked to lower cancer incidence [29–33], lower mortality [34, 35], and increased anti-tumor effects [22, 36, 37].
The immune system plays an essential role in the elimination of tumor cells. Initiation of the adaptive arm of the immune system, and specifically the activation of CD8+ cytotoxic T lymphocytes (CTL), is important, as these cells are able to recognize and kill tumor cells specifically. Dendritic cells (DCs) play a crucial role in orchestrating anti-tumor immunity as they are specialized cross-presenting cells, capable of priming tumor-specific CTLs. Three signals crucial for efficient T cell priming are: (cross)-presentation of extracellular tumor antigens in MHCI-molecules, co-stimulation, and cytokines [38–40]. Antigen presenting DCs are able to provide these signals, as they phagocytose tumor debris, cross-present these antigens and mature (express co-stimulatory markers) while migrating to the lymph nodes, and produce cytokines to activate naïve CD8+ T cells locally [41, 42]. Toll-like receptor (TLR) activation educates DCs to initiate effector T cell differentiation and expansion [43]. The immunomodulatory effect of several NSAIDs has been explored [44–47], but further insight is needed into the effect of different NSAIDs on the three signals exploited by DCs in one system.
Here, we set out to explore the effects of non-selective COX inhibitors diclofenac and ibuprofen and specific COX2 inhibitor celecoxib on the cross-presentation, co-stimulatory potential, and cytokine production by DCs. We report that TLR-induced cytokine production by murine bone-marrow derived DCs (mBMDCs) and human moDCs is impaired upon preincubation with NSAIDs. Understanding the role of NSAIDs on DC functioning might further improve DC-based cancer immunotherapy.
Material and methods
Mice, cell lines, and murine and human DC cultures
Female C57BL/6J mice (6–8 weeks old) were purchased from Charles River Wiga (Sulzfeld, Germany). Drinking water and standard laboratory food pellets were provided ad libitum and mice were allowed to settle for at least 1 week. The experiments were approved by the Animal Experiments Committee of the Radboud University Medical Center, and were performed in accordance with institutional, national and European guidelines. All mice were maintained under specific pathogen-free barrier conditions at the Central Animal Laboratory (Nijmegen, The Netherlands). Mice were sacrificed by cervical dislocation. mBMDCs were cultured as according to previously described protocol [48, 49]. In short, bone marrow was flushed out of tibia and femur and mashed over a nylon mesh with pore size 100 μm. Cells (4x106 in 10 cm dish) were cultured for 8 days in RPMI-1640 (Gibco, #42401–018), supplemented with 10% heat-inactivated fetal bovine serum (FBS, Greiner Bio-One), 2mM L-glutamine (Lonza, #BE17-605E/U1), 1% penicillin/streptomycin (pen/strep, Gibco, #15140–122) and 50 μM beta-mercaptoethanol (Gibco, #21985–023) (BMDC medium) in the presence of 20 ng/ml rmGM-CSF (PeproTech, # 315–03), at 37°C with 5% CO2. On day 3 and 6, rmGM-CSF was added to the culture. Non-adherent cells were harvested and used for assays (mBMDCs). B3Z cells, a T cell hybridoma specific for the immunodominant OVA Kb peptide in H-2Kb, which carries a β-galactosidase construct driven by NF-AT elements from the interleukin-2 (IL-2) promotor [50], were cultured in IMDM (Gibco, #21980–032) supplemented with 5% FBS, 2 mM L-glutamine, 1% pen/strep, 50 μM beta-mercaptoethanol, and 0.5 mg/ml hygromycin (Invitrogen, #10687010). Human monocyte-derived DCs (moDCs) were generated as described previously [51, 52], from cells isolated from buffy coats obtained from healthy volunteers (Sanquin, Nijmegen, The Netherlands) after written informed consent as per the Declaration of Helsinki. moDCs were acquired by culturing ficoll gradient (SepMate™, StemCell, #85450) obtained peripheral blood mononuclear cells (PBMCs) with 450 U/ml rhGM-CSF (Immunotools, #11343128) and 300 U/ml rhIL-4 (Immunotools, #11340047) for 6 days. rhGM-CSF and rhIL-4 were added again at day 3. moDCs were cultured in RPMI-1640 supplemented with 10% FCS, 2 mM glutamine, and 1% pen/strep. Adherent cells were used for assays. During all incubation steps, cells were kept at 37°C with 5% CO2, unless stated otherwise.
Adjuvants, reagents and antibodies
Lipopolysaccharide (LPS) was purchased from Sigma-Aldrich (St Louis, MO, USA), and used at a concentration of 0.5 μg/ml. CpG-ODN 1668 (‘5-TCCATGACGTTCCTGATGCT-3’) with total phosphorothioate-modified backbone was purchased from Sigma Genosys (Haverhill, UK), and used at a concentration of 1 μg/ml. R848 was purchased from Enzo Life Sciences, and used at a concentration of 4 μg/ml. The NSAIDs diclofenac sodium salt (PHR1144-1G), celecoxib (PHR1683-1G), and ibuprofen (PHR-1004-1G) were purchased from Sigma-Adrich. For every experiment, a fresh dilution in medium was made. Celecoxib was first diluted in DMSO and thereafter further diluted in medium (DMSO concentration <0.03%). For flow cytometry experiments the following antibodies (clone name in brackets, followed by supplier, and dilution) conjugated to various fluorophores were used: MHCII-BV510 (M5/114.15.2, Antibodychain, 1:500), CD80-A488 (16-10A1, Antibodychain, 1:1000), CD11b-PerCP (M1/70, Biolegend, 1:600), CD115-PeCy7 (AFS98, eBioscience, 1:200), CD11c-APC (HL3, BD, 1:400), and CD86-APCCy7 (GL-1, Biolegend, 1:800). Viability dye used was eFluor™ 450 (ThermoFisher, 1:4000).
RNA isolation and RT-qPCR
Total RNA was isolated 0.5x106 mBMDCs stimulated with TLR for 6 or 18 hours (hr), using TRIzol reagent (Invitrogen Life Technologies) according to the manufacturer’s instructions, with minor modifications. mRNA levels for the genes of interest were determined with a CFX96 sequence detection system (Bio-Rad) using the Faststart SYBR green mastermix (Roche) with SYBR Green as the fluorophore and gene-specific oligonucleotide primers. Primers used are COX1 FW TTACTATCCGTGCCAGAACCA, COX1 REV CCCGTGCGAGTACAATCACA, COX2 FW TTCAACACACTCTATCACTGGC, COX2 REV AGAAGCGTTTGCGGTACTCAT, Rer1 FW GCCTTGGGAATTTACCACCT, Rer1 REV CTTCGAATGAAGGGACGAAA. Quantitative PCR data were analyzed with the CFX Manager V1.6.541.1028 software (Bio-Rad) and checked for correct amplification and dissociation of the products. mRNA levels of the genes of interest were normalized to mRNA levels of Rer1 and were calculated according to the cycle threshold method.
In vitro cross-presentation assay
For in vitro cross-presentation assays 80x103 NSAID preincubated mBMDCs were pulsed with 80 μg/ml of endotoxin-free chicken egg ovalbumin (OVA protein, Endograde, Hyglos GmbH, Germany) in the presence of 400 ng/ml immune stimulatory complexes (ISCOMs). After pulsing, cells were washed and cultured overnight with 80x103 B3Z cells. As a control for cell viability and/or MHC-I expression levels, DCs were pulsed with 5 ng/ml OVA Kb peptide (SIINFEKL, 257–264, AS-60193, Tebu-bio) 30 min before adding the B3Z cells. The presentation of OVA Kb peptide in H-2Kb results in production of β-galactosidase (LacZ) by B3Z cells, which can be detected by adding 0.15 mM chlorophenolred-h-D-galactopyranoside (Calbiochem), 9 mM MgCl2, 0.125% NP40, and 7.5 mM DTT in PBS. Plates were incubated for 3 to 5 hr and absorbance values were measured at 595 nm using a photo spectrometer (Biorad).
Flow cytometry
Expression of maturation markers was assessed using flow cytometry. 15x104 NSAID preincubated mBMDCs were stimulated with TLR ligands, washed with medium, and rested overnight in fresh medium. Subsequently, medium was removed, and cells were washed with phosphate buffered saline (PBS). Samples were incubated with viability dye for 15 minutes on ice, and washed with PBS and thereafter washed with PBS supplemented with 0,5% bovine serum albumin and 0,05% sodium azide (PBA). Samples were incubated with Fc-block (CD16/32, clone 2.4G2, BD, 1:800) for 10 minutes on ice. Next, cells were incubated with fluorescently labeled antibodies in PBA for 20 min on ice. Subsequently, cells were washed twice, diluted in PBA and measured on FACS Canto.
Measurement soluble factors
15x104 NSAID preincubated mBMDCs were stimulated with TLR ligands, washed with medium, and rested overnight in fresh medium. Supernatant was collected 16–24 hr after incubation with adjuvants and stored at -80°C. Cytokines (IL-6, IL-10, IL-12, and TNF-α) were measured using ELISA kits (mIL-6: 88–7064, mIL-10: 88–7105, mIL-12p70: 88–7121, mTNF-α: 88–7324, hIL-6: 88–7066, hIL-10: 88–7106, hIL-12: 88–7126, hTNF-α: 88–7346; all from Invitrogen) according to the manufacturer’s instructions. PGE2 was measured using the R&D kit (KGE004B), according to manufacturer’s instructions, with minor modifications. In short, prior to the PGE2 measurement, large size proteins in the supernatant interfering with the assay were removed using a perchloric-acid (PCA) precipitation. For this, 13.7% PCA (SIGMA, 244252-100M) was added to supernatant. Supernatants were spun down for 5 minutes at maximum speed, aspirated to new plate, and neutralization was performed using 4N NaOH.
Viability measurement
Metabolic activity as a measure for cell viability was assessed using the CCK8 kit (96992-3000TESTS-F, Sigma-Aldrich). In short, 15x104 GMCSF DCs were cultured with NSAIDs for 6h. The cells were washed, medium was replaced (100 μL) and 20 μL CCK8 reagent was added to each well. Plates were incubated for 1 to 3 hr and absorbance values were measured at 450 nm using a photo spectrometer (Biorad).
Statistical analysis
Depending on the experimental layout, data were analyzed using a two-tailed Student’s t-test, a one-way or 2-way ANOVA or mixed effects analysis with post hoc Dunnett’s or Sidak’s multiple comparisons test (medium versus rest) or Tukey’s multiple comparisons test, as indicated in the figure legends. Differences were considered significant when P values were smaller than 0.05, while the following symbols were used: *P < 0.05; **P < 0.01; ***P < 0.001. The statistical analyses were performed in Graphpad Prism 8.0.1.
Results
TLR matured mBMDCs provide 3 key signals to induce CTLs: the presentation of extracellular (tumor) antigens in the MHCI complex, costimulation, and cytokines [38]. mBMDCs also respond to TLR triggering by a strong upregulation of COX2 mRNA expression, while COX-1 expression is down regulated (Fig 1A). This effect of TLR stimulation on the expression of COX enzymes in mBMDC is also reflected by the increased production of PGE2 (Fig 1B) and is in line with previous findings [50]. We then set out to explore the effect of NSAIDs on cross-presentation, costimulation and cytokine production by DCs. For this, we pretreated mBMDCs with NSAIDs for 6 hr, followed by a 2.5 hr induction of the DC cross-presentation adjuvant ISCOMs or a 2.5 hr induction of DC maturation by the TLR ligands LPS, CpG, or R848 (Fig 2A for experimental setup). This timing regimen was chosen as a 2.5 hr incubation with ISCOMs or TLR-ligands is sufficient for the induction of cross-presentation, co-stimulatory marker expression and cytokine secretion by DCs (S1 Fig). Exposure of DC for 6 hr to NSAIDs did not affect their viability (S2A Fig) nor the COX1 or COX2 expression (S2B Fig). A 6 hr pre-incubation period with NSAIDs allowed us to study the early effects of NSAIDs on DC function.
mBMDCs were treated for 6 hr (a) and 18 hr (a, b) with the TLR adjuvants LPS, CpG, and R848. (a) RT-QPCR was performed for mRNA expression of COX1 and COX2 (mBMDCs, n = 6). (b) PGE2 was measured using a competitive enzyme immunoassay in PCA-treated supernatant (n = 3). Results are shown as means with SEM. Statistical significance was calculated using a one-way ANOVA with Dunnett’s multiple comparison test, medium versus rest.
(a) Timeline shows the experimental setup for assessing cross-presentation, maturation markers and cytokine production by mBMDCs. (b-d) mBMDCs were first pretreated with NSAIDs for 6 hr, washed and subsequently treated with OVA protein and ISCOMs for 2.5 hr, and then co-cultured with B3Z T cells for 18 hr. As a positive control for viability and MHC-I levels, mBMDCs were pulsed with OVA peptide (SIINFEKL) 0.5 hr before coculture with B3Z T cells (n = 4). ISCOM induced cross-presentation of OVA protein and stable loading of exogenous SIINFEKL by celecoxib (b), diclofenac (c), and ibuprofen (d) pretreated DCs. Results are shown as means with SEM. Statistical significance was calculated using a one-way ANOVA with Dunnett’s multiple comparison test, medium versus rest.
NSAIDs do not affect ISCOM induced cross-presentation by mBMDCs
The effect of NSAIDs on the DC’s capacity to cross-present antigens was assessed using mBMDCs treated with the model antigen OVA in combination with the cross-presentation inducing ISCOM adjuvant as described previously [45]. Cross-presentation of the OVA peptide (SIINFEKL) in MHCI molecules by DCs was detected upon co-culture with the OVA-specific B3Z reporter T cell hybridoma as a readout system (Fig 2A). As expected, OVA protein cross-presentation by mBMDCs is greatly enhanced by the addition of ISCOMs (Fig 2B–2D). Preincubation of DCs with the NSAIDs celecoxib (Fig 2B), diclofenac (Fig 2C) or ibuprofen (Fig 2D) prior to exposure to OVA protein and ISCOMs did not affect cross-presentation of OVA protein. B3Z activation by OVA peptide loaded control DC was similar in all conditions, indicating that MHCI expression and viability was not affected by preincubation of the DCs with NSAIDs. These results demonstrate that NSAIDs do not affect ISCOM induced cross-presentation by mBMDCs.
Maturation marker profiles remain unaffected by NSAID exposure
The bone-marrow-derived DC cultures with GM-CSF give rise to two major DC subpopulations [53], the GM-MAC (MHCIIlowCD11bhiCD115hi) and GM-DC (MHCIIhiCD11bintCD115low). This prompted us to assess whether NSAIDs specifically affect the development or maturation of these different DC subpopulations (S3A Fig). Our flow cytometry data shows that the percentage of total CD11c+ cells was slightly lowered by the 6 hr diclofenac preincubation, when stimulated with CpG (Fig 3A). Preincubation with ibuprofen dose-dependently lowered CD11c+ population in all TLR-stimulated conditions (Fig 3A). The ratio between the subpopulations of GM-DCs versus GM-MACs was not significantly affected upon NSAID exposure (Fig 3B).
(a-d) mBMDCs were first pretreated with NSAIDs for 6 hr, followed by a 2.5 hr TLR stimulation with LPS, CpG, or R848. After overnight rest in fresh medium, composition of the mBMDC culture (a, b) and maturation marker expression (c, d) were analyzed using flow cytometry (n = 4). Results are shown as means with SEM. (a) Percentage CD11c+ cells in BMDC culture. (b) GM-DC/GM-MAC ratio within CD11c+ population. (c) Mean Fluorescent Intensity (MFI) ± SEM of CD80 and (d) CD86 in CD11c+ population. Statistical significance was calculated using a one-way ANOVA with Dunnett’s multiple comparison test, medium versus rest.
Next, we assessed the expression of maturation markers CD80 and CD86 on total CD11c+ DCs and the GM-DC subpopulation as this is the subpopulation expressing highest levels of CD80 and CD86 (S3B Fig) following TLR stimulation. CD80 and CD86 are important co-stimulatory molecules expressed on mature DCs that help the activation of CD8+ T cells towards effective CTLs. NSAIDs pretreatment did not significantly affect the expression levels of CD80 and CD86 on CD11c+ DCs (Fig 3C and 3D). In the GMDC subpopulation only minor differences were observed (S3C–S3E Fig). Together these results imply that preincubation with NSAIDs does not affect TLR-induced maturation marker expression.
NSAID exposure reduces TLR-induced cytokine production by mBMDCs
Cytokine secretion by DCs is the third signal essential for T cell activation [54], and also plays an important role in the differentiation towards specific subtypes of T cells [55, 56]. Therefore, we investigated cytokine production by mBMDCs after preincubation with different NSAIDs and stimulation with different TLR adjuvants. As expected, DCs pulsed for 2.5 hr with adjuvants increase their overnight production of IL-6, IL-10, IL-12 and TNF-α (S4A Fig). Interestingly, preincubation with celecoxib reduced TLR-induced production of IL-10, IL-12, TNF-α, and IL-6 (Fig 4A). Diclofenac pre-exposure significantly decreased IL-12 production upon TLR stimulation (Fig 4B). Also a trend towards lower IL-10, IL-6 and TNF-α production after CpG stimulation by diclofenac pretreated DCs was observed. Ibuprofen also strongly inhibited IL-12 production of DCs (Fig 4C). In addition, also TNF-α and IL-6 production was decreased after ibuprofen preincubation (Fig 4C). These results indicate that preincubation with NSAIDs leads to decreased TLR-induced cytokine production by mBMDCs.
(a-c) mBMDCs were first pretreated with NSAIDs for 6 hr, followed by a 2.5 hr TLR stimulation with LPS, CpG, or R848. After overnight rest in fresh medium, cytokines were measured using ELISA (n = 4). IL-10, IL-12, TNF-α, and IL-6 production after TLR stimulation by (a) celecoxib, (b) diclofenac, and (c) ibuprofen preincubated DCs. Results are shown as means with SEM. Statistical significance was calculated using a one-way ANOVA with Dunnett’s multiple comparison test, medium versus rest.
Cytokine secretion by mature human DCs is affected by exposure to NSAIDs
To extrapolate our data on the effect of NSAIDs on cytokine production by mBMDCs to the human situation, human monocyte derived DCs (moDCs) were cultured from human blood and exposed to NSAIDs according to the same schedule (Fig 2A). Many reports have described that myeloid DCs and moDCs do not express TLR9 or do not respond to its ligand CpG [57–62]. We therefore used LPS and R848 as TLR stimuli. As expected, also human moDCs secrete IL-6, IL-10 and TNF-α after 2.5 hr stimulation with adjuvants (S4B Fig). TNF-α was mainly produced within the 2.5 hr stimulation period (S4B Fig) with R848, while IL-12 secretion was not significantly detected after LPS nor R848 stimulation (S4B Fig). In line with our murine data, also human moDCs show a trend towards diminished cytokine expression profile when exposed to celecoxib or diclofenac prior to the TLR stimulation (Fig 5A and 5B). The decreased production of TNF-α upon celecoxib or diclofenac preincubation was mainly detected in the first 2.5 hr during TLR stimulation (S4C Fig). For ibuprofen treated moDCs, the effect was less pronounced (Fig 5C). Altogether, TLR-stimulated human moDCs, like mBMDCs, appear to produce less cytokines when pretreated with celecoxib and diclofenac.
(a-c) moDCs were first pretreated with NSAIDs for 6 hr, followed by a 2.5 hr TLR stimulation with LPS or R848. After overnight rest in fresh medium, cytokines were measured using ELISA (n = 5–6). IL-10, TNF-α, and IL-6 production after TLR stimulation by (a) celecoxib, (b) diclofenac, and (c) ibuprofen preincubated moDCs. Results are shown as means with SEM. Statistical significance was calculated using Mixed-effects analysis with Dunnett’s multiple comparisons test.
Discussion
There is compelling evidence that NSAIDs lower different cancer types (colon, breast, prostate and lung) incidence and improve survival [29–35]. NSAIDs are suggested to play a role in the modulation of anti-tumor immunity [22, 36, 37, 63]. Since DCs play a vital role in the anti-tumor immunity cascade, we have studied the effects of NSAIDs on the three pathways (cross-presentation, co-stimulation, cytokines) applied by DCs to induce tumor-specific CD8+ CTLs. Our data demonstrate that celecoxib (selective COX2 inhibitor) and diclofenac and ibuprofen (COX1/2 inhibitors) have no or little effect on DC cross-presentation and CD80/CD86 co-stimulatory molecule expression, but reduce TLR-induced cytokine production by both murine bone marrow-derived DCs and human moDCs.
Cross-presentation is a process primarily executed by DCs and is essential to initiate a cytotoxic T cell response to exogenous tumor antigens. In this study, we show that diclofenac, ibuprofen and celecoxib had no effect on ISCOMs-induced cross-presentation and TLR-induced (murine) DC maturation. Interestingly, in contrast to our results on freshly differentiated DCs, Kim et al. has shown that immortalized DCs (DC2.4) are slightly restricted in cross-presentation of OVA present in biodegradable microspheres when cultured with ibuprofen [44]. Their basal CD80 and CD86 expression slightly increased upon 18 hr ibuprofen exposure but not their phagocytic capacity [44]. The use of a different source of DCs, antigen delivery vehicle and stimulation schedule (prolonged dosed and higher doses of ibuprofen) may explain these differences. In line with our data, diclofenac does not alter nickel-induced human DC maturation and CD86 expression [46].
Furthermore, we report that diclofenac strongly reduced the production of TLR-induced pro- and anti-inflammatory cytokines by murine DCs, and a similar trend towards reduced production was observed for human DCs. Statistical significance for NSAID-induced reduction in cytokine production by TLR stimulated moDCs was not reached, as the magnitude of the effects were variable between donors.
The secretion of distinct soluble and membrane-bound molecules determines the downstream polarization of T cells towards type 1 or type 2 responses [64]. Th1 cells secrete IL-2 and IFN-γ, cytokines promoting differentiation and maturation of CD8+ T cells into CTLs [65, 66]. Interfering with DC polarization and thereby cytokine production can thus affect T cell skewing and thereby modify the outcome of anti-tumor immune responses. Previous studies showed that diclofenac induces the polarization towards Th1 instead of Th2 cells, as it suppresses type 2 (CCL17) DC cytokine secretion [46], and not type 1 cytokines (TNF-α, IL-12p70). Another group showed that murine bone marrow-derived DCs co-cultured with glioma cells show enhanced IL-12 and decreased IL-10 secretion in the presence of diclofenac after stimulation with R848 [45]. Furthermore, ex vivo analyses revealed that tumor-infiltrating DCs regained their capacity to produce IL-12 on R848 stimulation after diclofenac exposure [45]. In contrast to our results where we saw a decrease in TLR-induced IL-12 production after pretreatment with diclofenac, these studies indicate that diclofenac promotes Th1 differentiation. This discrepancy possibly depends on the exposure time with NSAIDs, specific DC subsets used and the presence of glioma cells. The environmental context is important for the effect of NSAIDs on cytokine production by DCs. Future research will have to elucidate whether (tumor)microenvironmental factors influence pro- and anti-inflammatory cytokine production upon NSAID exposure.
Previously, it was shown that also celecoxib may modulate the balance between Th1 cytokines and Th2 cytokines by increasing Th1 cytokine IL-12 and reducing Th2 cytokine IL-10 [47], although its effect on cytokine production was most evidently seen in combination with tumor lysate. In contrast to the effect of celecoxib on tumor lysate induced DCs, we here show that celecoxib reduced TLR-induced production of IL-10, IL-12, TNF-α, and IL-6. It will be interesting to study the effect of altered cytokine production by NSAIDs through DCs on T cell skewing/differentiation in future experiments.
Our study underscores that combination of NSAIDs with DC-based immunotherapies should be explored in more detail, as NSAIDs are able to alter DC function, and thereby could affect (immuno)therapy outcome. Since NSAIDs also reduce immunosuppressive PGE2 levels in the tumor microenvironment, their combination with different immunotherapies has been studied preclinically. Both celecoxib and aspirin reduced PGE2 levels in the melanoma tumor microenvironment, inhibited PGE2-dependent suppression of myeloid cell activation, and increased anti-PD-1 efficacy [67]. Immature human DCs differentiate towards stable myeloid-derived suppressor cells (MDSC) when exposed to PGE2 [68]. Celecoxib refined DC-based immunotherapy by preventing the local and systemic expansion of all MDSC subtypes, reducing levels of reactive oxygen species and nitric oxide, and reversing T cell tolerance for mesothelioma [69]. Also in COX2 inhibitor SC5836-treated 4T1 bearing mice, MDSC accumulation in the tumor was reduced, and primary outgrowth was delayed [70]. Celecoxib administered after primary tumor establishment synergized with tumor lysate‐pulsed DCs and GM‐CSF [71]. This combination therapy suppressed primary 4T1 murine mammary tumor growth and markedly reduced the occurrence of lung metastases, due to an enhanced tumor-specific immune response [71].
The direct effect of NSAIDs on DC polarization may play a pivotal role in the initiation of anti-tumor immunity. Now that NSAIDs are explored for their beneficial properties in diverse anti-tumor immunotherapies in different tumor types, it will be important to further study their potential influence on treatment outcome.
Conclusion
Altogether we show that preincubation with NSAIDs does not affect ISCOM-induced cross-presentation nor TLR-induced CD80 or CD86 co-stimulatory molecule expression. Cytokine production by both mBMDCs and human moDCs, however, is diminished when DCs were pretreated with celecoxib, diclofenac, or ibuprofen. Since the presence of NSAIDs can affect DC polarization, giving NSAIDs to cancer patients might affect the initiation of the anti-tumor immunity cascade, and thereby could affect treatment outcome.
Supporting information
S1 Fig. 2.5 hr stimulation induces cross-presentation, maturation markers, and sufficient cytokine production after replacement medium.
(a) Experimental design with different incubation durations and stimuli. (b) mBMDCs were treated with OVA protein and ISCOMs for indicated durations, and then co-cultured with B3Z T cells for 18 hr. As a positive control for viability and MHC-I levels, mBMDCs were pulsed with OVA peptide (SIINFEKL) 0.5 hr before coculture with B3Z T cells (n = 3). Statistical significance calculated using 2-way ANOVA, Sidak’s multiple comparisons test (for medium versus ISCOM), Tukey’s multiple comparisons test (for ISCOMs versus ISCOMs). mBMDCs were stimulated with LPS, CpG, or R848 for indicated durations (c-e). After overnight rest in fresh medium (for 1, 2.5, and 5 hr pulse) or overnight stimulation without rest in fresh medium (O/N stimulation), maturation marker expression in CD11c+ cells was analyzed using flow cytometry (n = 5). Statistical significance was calculated using a one-way ANOVA with Dunnett’s multiple comparison test, medium versus rest. Directly after the pulse (d), and during the first and second 6 hr after refreshment of the medium, cytokines IL-6, IL-10, IL-12 and TNF-α were measured using ELISA (n = 2–3). (b-d) Results are shown as means with SEM.
https://doi.org/10.1371/journal.pone.0275906.s001
(TIF)
S2 Fig. Viability and COX1/2 expression is maintained after 6 hr NSAID stimulation.
(a-b) mBMDCs were treated for 6 hr with NSAIDs. (a) CCK8 assay as a read out for cell viability and metabolic activity, after 6 hr stimulation with NSAIDs. Raw data–blanco depicted (n = 4). Statistical significance was calculated using one-way ANOVA with Dunnett’s multiple comparisons test, medium versus rest. (b) RT-QPCR was performed for mRNA expression of COX1 and COX2 (mBMDCs, n = 3–4). Results are shown as means with SEM. Statistical significance was calculated using a two-tailed Student’s t-test.
https://doi.org/10.1371/journal.pone.0275906.s002
(TIF)
S3 Fig. CD80 and CD86 expression in GMMAC and GMDC subsets.
(a) Gating strategy of mBMDCs to GMMAC (MHCIIlowCD11bhiCD115hi) and GMDCs (MHCIIhiCD11bintCD115low). mBMDCs were not pretreated (b), and first pretreated with NSAIDs (c) celecoxib, (d) diclofenac, or (e) ibuprofen for 6 hr, followed by a 2.5 hr TLR stimulation with LPS, CpG or R848. After overnight rest in fresh medium, maturation marker expression was analyzed using flow cytometry (n = 4). Mean Fluorescent Intensity (MFI) ± SEM of CD80 and CD86 in CD11c+ GMMAC (b) and GMDC (b-e) population. Statistical significance was calculated using a one-way ANOVA with Dunnett’s multiple comparison test, medium versus rest.
https://doi.org/10.1371/journal.pone.0275906.s003
(TIF)
S4 Fig. TLR induced cytokine production by mBMDCs and moDCs.
mBMDCs (a, n = 4) and moDCs (b-c, n = 5–6) were first pretreated with NSAIDs for 6 hr, followed by a 2.5 hr TLR stimulation with LPS or R848. After overnight rest in fresh medium (a-b) and directly after pulse (b-c), cytokines IL-6, IL-10, IL-12 (a-b), and TNF-α (a-c), were measured using ELISA. Results are shown as means with SEM. Statistical significance was calculated using Mixed-effects analysis with Dunnett’s multiple comparisons test, on raw data. (b) Detection limit for hIL-12 is 50–4000 pg/ml.
https://doi.org/10.1371/journal.pone.0275906.s004
(TIF)
Acknowledgments
We would like to thank M. Wassink, L. Huis in’t Veld, L. Cornelissen, M. Looiman and E. Kers-Rebel for technical assistance and M. H. den Brok for helpful discussions.
References
- 1.
WebMD. Pain Management: Cancer Pain. Available from: https://www.webmd.com/cancer/pain-management-cancer-pain#1.
- 2.
Society AC. Non-opioids and Other Drugs Used to Treat Cancer Pain [Available from: https://www.cancer.org/treatment/treatments-and-side-effects/physical-side-effects/pain/non-opioids-and-other-drugs-to-treat-cancer-pain.html.
- 3. Vane JR. Inhibition of prostaglandin synthesis as a mechanism of action for aspirin-like drugs. Nat New Biol. 1971;231(25):232–5. pmid:5284360
- 4. Higgs GA. Arachidonic acid metabolism, pain and hyperalgesia: the mode of action of non-steroid mild analgesics. Br J Clin Pharmacol. 1980;10 Suppl 2:233S–5S. pmid:7002184
- 5. Rodriguez M, Domingo E, Municio C, Alvarez Y, Hugo E, Fernandez N, et al. Polarization of the innate immune response by prostaglandin E2: a puzzle of receptors and signals. Mol Pharmacol. 2014;85(1):187–97. pmid:24170779
- 6. den Brok MH, Raaijmakers TK, Collado-Camps E, Adema GJ. Lipid Droplets as Immune Modulators in Myeloid Cells. Trends in immunology. 2018;39(5):380–92. pmid:29478771
- 7. Prescott SM, Fitzpatrick FA. Cyclooxygenase-2 and carcinogenesis. Biochimica et biophysica acta. 2000;1470(2):M69–78. pmid:10722929
- 8. Chell S, Kaidi A, Williams AC, Paraskeva C. Mediators of PGE2 synthesis and signalling downstream of COX-2 represent potential targets for the prevention/treatment of colorectal cancer. Biochimica et biophysica acta. 2006;1766(1):104–19. pmid:16859832
- 9. England S, Bevan S, Docherty RJ. PGE2 modulates the tetrodotoxin-resistant sodium current in neonatal rat dorsal root ganglion neurones via the cyclic AMP-protein kinase A cascade. J Physiol. 1996;495 (Pt 2):429–40. pmid:8887754
- 10. Tsujii M, DuBois RN. Alterations in cellular adhesion and apoptosis in epithelial cells overexpressing prostaglandin endoperoxide synthase 2. Cell. 1995;83(3):493–501. pmid:8521479
- 11. Tang X, Sun YJ, Half E, Kuo MT, Sinicrope F. Cyclooxygenase-2 overexpression inhibits death receptor 5 expression and confers resistance to tumor necrosis factor-related apoptosis-inducing ligand-induced apoptosis in human colon cancer cells. Cancer research. 2002;62(17):4903–8. pmid:12208739
- 12. Kinoshita T, Takahashi Y, Sakashita T, Inoue H, Tanabe T, Yoshimoto T. Growth stimulation and induction of epidermal growth factor receptor by overexpression of cyclooxygenases 1 and 2 in human colon carcinoma cells. Biochimica et biophysica acta. 1999;1438(1):120–30. pmid:10216286
- 13. Castellone MD, Teramoto H, Williams BO, Druey KM, Gutkind JS. Prostaglandin E2 promotes colon cancer cell growth through a Gs-axin-beta-catenin signaling axis. Science. 2005;310(5753):1504–10. pmid:16293724
- 14. Gallo O, Franchi A, Magnelli L, Sardi I, Vannacci A, Boddi V, et al. Cyclooxygenase-2 pathway correlates with VEGF expression in head and neck cancer. Implications for tumor angiogenesis and metastasis. Neoplasia. 2001;3(1):53–61. pmid:11326316
- 15. Leung WK, To KF, Go MY, Chan KK, Chan FK, Ng EK, et al. Cyclooxygenase-2 upregulates vascular endothelial growth factor expression and angiogenesis in human gastric carcinoma. Int J Oncol. 2003;23(5):1317–22. pmid:14532971
- 16. Wang D, DuBois RN. Cyclooxygenase 2-derived prostaglandin E2 regulates the angiogenic switch. Proceedings of the National Academy of Sciences of the United States of America. 2004;101(2):415–6. pmid:14707264
- 17. Wang D, Dubois RN. Eicosanoids and cancer. Nat Rev Cancer. 2010;10(3):181–93. pmid:20168319
- 18. Kalinski P. Regulation of immune responses by prostaglandin E2. Journal of immunology. 2012;188(1):21–8. pmid:22187483
- 19. Pockaj BA, Basu GD, Pathangey LB, Gray RJ, Hernandez JL, Gendler SJ, et al. Reduced T-cell and dendritic cell function is related to cyclooxygenase-2 overexpression and prostaglandin E2 secretion in patients with breast cancer. Ann Surg Oncol. 2004;11(3):328–39. pmid:14993030
- 20. Gupta RA, Tejada LV, Tong BJ, Das SK, Morrow JD, Dey SK, et al. Cyclooxygenase-1 is overexpressed and promotes angiogenic growth factor production in ovarian cancer. Cancer research. 2003;63(5):906–11. pmid:12615701
- 21. Eberhart CE, Coffey RJ, Radhika A, Giardiello FM, Ferrenbach S, DuBois RN. Up-regulation of cyclooxygenase 2 gene expression in human colorectal adenomas and adenocarcinomas. Gastroenterology. 1994;107(4):1183–8. pmid:7926468
- 22. Molina MA, Sitja-Arnau M, Lemoine MG, Frazier ML, Sinicrope FA. Increased cyclooxygenase-2 expression in human pancreatic carcinomas and cell lines: growth inhibition by nonsteroidal anti-inflammatory drugs. Cancer research. 1999;59(17):4356–62. pmid:10485483
- 23. Hida T, Yatabe Y, Achiwa H, Muramatsu H, Kozaki K, Nakamura S, et al. Increased expression of cyclooxygenase 2 occurs frequently in human lung cancers, specifically in adenocarcinomas. Cancer research. 1998;58(17):3761–4. pmid:9731479
- 24. Aab A, Abreu P, Aglietta M, Ahn EJ, Samarai IA, Albuquerque IF, et al. Search for patterns by combining cosmic-ray energy and arrival directions at the Pierre Auger Observatory. Eur Phys J C Part Fields. 2015;75(6):269. pmid:26120280
- 25. Zimmermann KC, Sarbia M, Weber AA, Borchard F, Gabbert HE, Schror K. Cyclooxygenase-2 expression in human esophageal carcinoma. Cancer research. 1999;59(1):198–204. pmid:9892207
- 26. Kirschenbaum A, Klausner AP, Lee R, Unger P, Yao S, Liu XH, et al. Expression of cyclooxygenase-1 and cyclooxygenase-2 in the human prostate. Urology. 2000;56(4):671–6. pmid:11018637
- 27. Ke HL, Tu HP, Lin HH, Chai CY, Chang LL, Li WM, et al. Cyclooxygenase-2 (COX-2) up-regulation is a prognostic marker for poor clinical outcome of upper tract urothelial cancer. Anticancer Res. 2012;32(9):4111–6. pmid:22993369
- 28. Achiwa H, Yatabe Y, Hida T, Kuroishi T, Kozaki K, Nakamura S, et al. Prognostic significance of elevated cyclooxygenase 2 expression in primary, resected lung adenocarcinomas. Clinical cancer research: an official journal of the American Association for Cancer Research. 1999;5(5):1001–5. pmid:10353732
- 29. Harris RE. Cyclooxygenase-2 (cox-2) blockade in the chemoprevention of cancers of the colon, breast, prostate, and lung. Inflammopharmacology. 2009;17(2):55–67. pmid:19340409
- 30. Arber N. Cyclooxygenase-2 inhibitors in colorectal cancer prevention: point. Cancer Epidemiol Biomarkers Prev. 2008;17(8):1852–7. pmid:18708371
- 31. Johnson TW, Anderson KE, Lazovich D, Folsom AR. Association of aspirin and nonsteroidal anti-inflammatory drug use with breast cancer. Cancer Epidemiol Biomarkers Prev. 2002;11(12):1586–91. pmid:12496048
- 32. Harris RE, Chlebowski RT, Jackson RD, Frid DJ, Ascenseo JL, Anderson G, et al. Breast cancer and nonsteroidal anti-inflammatory drugs: prospective results from the Women’s Health Initiative. Cancer research. 2003;63(18):6096–101. pmid:14522941
- 33. Ratliff TL. Aspirin, ibuprofen, and other non-steroidal anti-inflammatory drugs in cancer prevention: a critical review of non-selective COX-2 blockade (review). J Urol. 2005;174(2):787–8.
- 34. Chan AT, Ogino S, Fuchs CS. Aspirin use and survival after diagnosis of colorectal cancer. JAMA. 2009;302(6):649–58. pmid:19671906
- 35. Huang WW, Hsieh KP, Huang RY, Yang YH. Role of cyclooxygenase-2 inhibitors in the survival outcome of colorectal cancer patients: A population-based cohort study. Kaohsiung J Med Sci. 2017;33(6):308–14. pmid:28601236
- 36. Hasegawa K, Ohashi Y, Ishikawa K, Yasue A, Kato R, Achiwa Y, et al. Expression of cyclooxygenase-2 in uterine endometrial cancer and anti-tumor effects of a selective COX-2 inhibitor. Int J Oncol. 2005;26(5):1419–28. pmid:15809736
- 37. Masferrer JL, Leahy KM, Koki AT, Zweifel BS, Settle SL, Woerner BM, et al. Antiangiogenic and antitumor activities of cyclooxygenase-2 inhibitors. Cancer research. 2000;60(5):1306–11. pmid:10728691
- 38. Sallusto F, Lanzavecchia A. The instructive role of dendritic cells on T-cell responses. Arthritis Res. 2002;4 Suppl 3:S127–32. pmid:12110131
- 39. Banchereau J, Steinman RM. Dendritic cells and the control of immunity. Nature. 1998;392(6673):245–52. pmid:9521319
- 40. Iwasaki A, Medzhitov R. Control of adaptive immunity by the innate immune system. Nature immunology. 2015;16(4):343–53. pmid:25789684
- 41. Chen DS, Mellman I. Oncology meets immunology: the cancer-immunity cycle. Immunity. 2013;39(1):1–10. pmid:23890059
- 42. Kim R, Emi M, Tanabe K. Functional roles of immature dendritic cells in impaired immunity of solid tumour and their targeted strategies for provoking tumour immunity. Clin Exp Immunol. 2006;146(2):189–96. pmid:17034569
- 43. Hemmi H, Akira S. TLR signalling and the function of dendritic cells. Chem Immunol Allergy. 2005;86:120–35. pmid:15976491
- 44. Kim HJ, Lee YH, Im SA, Kim K, Lee CK. Cyclooxygenase Inhibitors, Aspirin and Ibuprofen, Inhibit MHC-restricted Antigen Presentation in Dendritic Cells. Immune network. 2010;10(3):92–8. pmid:20631879
- 45. Chirasani SR, Leukel P, Gottfried E, Hochrein J, Stadler K, Neumann B, et al. Diclofenac inhibits lactate formation and efficiently counteracts local immune suppression in a murine glioma model. International journal of cancer. 2013;132(4):843–53. pmid:22752934
- 46. Toebak MJ, de Rooij J, Moed H, Stoof TJ, von Blomberg BM, Bruynzeel DP, et al. Differential suppression of dendritic cell cytokine production by anti-inflammatory drugs. The British journal of dermatology. 2008;158(2):225–33. pmid:18028503
- 47. Zhang H, Tian M, Xiu C, Wang Y, Tang G. Enhancement of antitumor activity by combination of tumor lysate-pulsed dendritic cells and celecoxib in a rat glioma model. Oncology research. 2013;20(10):447–55. pmid:24308155
- 48. Lutz MB, Kukutsch N, Ogilvie AL, Rossner S, Koch F, Romani N, et al. An advanced culture method for generating large quantities of highly pure dendritic cells from mouse bone marrow. J Immunol Methods. 1999;223(1):77–92. pmid:10037236
- 49. Raaijmakers TK, van den Bijgaart RJE, den Brok MH, Wassink M, de Graaf A, Wagenaars JA, et al. Tumor ablation plus co-administration of CpG and saponin adjuvants affects IL-1 production and multifunctional T cell numbers in tumor draining lymph nodes. Journal for immunotherapy of cancer. 2020;8(1). pmid:32461350
- 50. Sanderson S, Shastri N. LacZ inducible, antigen/MHC-specific T cell hybrids. Int Immunol. 1994;6(3):369–76. pmid:8186188
- 51. Hontelez S, Ansems M, Karthaus N, Zuidscherwoude M, Looman MW, Triantis V, et al. Dendritic cell-specific transcript: dendritic cell marker and regulator of TLR-induced cytokine production. Journal of immunology. 2012;189(1):138–45. pmid:22615205
- 52. Tel-Karthaus N, Kers-Rebel ED, Looman MW, Ichinose H, de Vries CJ, Ansems M. Nuclear Receptor Nur77 Deficiency Alters Dendritic Cell Function. Frontiers in immunology. 2018;9:1797. pmid:30123220
- 53. Helft J, Bottcher J, Chakravarty P, Zelenay S, Huotari J, Schraml BU, et al. GM-CSF Mouse Bone Marrow Cultures Comprise a Heterogeneous Population of CD11c(+)MHCII(+) Macrophages and Dendritic Cells. Immunity. 2015;42(6):1197–211. pmid:26084029
- 54. Curtsinger JM, Mescher MF. Inflammatory cytokines as a third signal for T cell activation. Current opinion in immunology. 2010;22(3):333–40. pmid:20363604
- 55. Gutcher I, Becher B. APC-derived cytokines and T cell polarization in autoimmune inflammation. J Clin Invest. 2007;117(5):1119–27. pmid:17476341
- 56. de Jong EC, Smits HH, Kapsenberg ML. Dendritic cell-mediated T cell polarization. Springer Semin Immunopathol. 2005;26(3):289–307. pmid:15609003
- 57. Ito T, Amakawa R, Kaisho T, Hemmi H, Tajima K, Uehira K, et al. Interferon-alpha and interleukin-12 are induced differentially by Toll-like receptor 7 ligands in human blood dendritic cell subsets. The Journal of experimental medicine. 2002;195(11):1507–12. pmid:12045249
- 58. Hartmann G, Weiner GJ, Krieg AM. CpG DNA: a potent signal for growth, activation, and maturation of human dendritic cells. Proceedings of the National Academy of Sciences of the United States of America. 1999;96(16):9305–10. pmid:10430938
- 59. Jarrossay D, Napolitani G, Colonna M, Sallusto F, Lanzavecchia A. Specialization and complementarity in microbial molecule recognition by human myeloid and plasmacytoid dendritic cells. European journal of immunology. 2001;31(11):3388–93. pmid:11745357
- 60. Kadowaki N, Ho S, Antonenko S, Malefyt RW, Kastelein RA, Bazan F, et al. Subsets of human dendritic cell precursors express different toll-like receptors and respond to different microbial antigens. The Journal of experimental medicine. 2001;194(6):863–9. pmid:11561001
- 61. Bauer M, Redecke V, Ellwart JW, Scherer B, Kremer JP, Wagner H, et al. Bacterial CpG-DNA triggers activation and maturation of human CD11c-, CD123+ dendritic cells. Journal of immunology. 2001;166(8):5000–7. pmid:11290780
- 62. Krug A, Towarowski A, Britsch S, Rothenfusser S, Hornung V, Bals R, et al. Toll-like receptor expression reveals CpG DNA as a unique microbial stimulus for plasmacytoid dendritic cells which synergizes with CD40 ligand to induce high amounts of IL-12. European journal of immunology. 2001;31(10):3026–37. pmid:11592079
- 63. Tegeder I, Pfeilschifter J, Geisslinger G. Cyclooxygenase-independent actions of cyclooxygenase inhibitors. FASEB J. 2001;15(12):2057–72. pmid:11641233
- 64. Kaiko GE, Horvat JC, Beagley KW, Hansbro PM. Immunological decision-making: how does the immune system decide to mount a helper T-cell response? Immunology. 2008;123(3):326–38. pmid:17983439
- 65. Wang JC, Livingstone AM. Cutting edge: CD4+ T cell help can be essential for primary CD8+ T cell responses in vivo. Journal of immunology. 2003;171(12):6339–43. pmid:14662830
- 66. Abbas AK, Murphy KM, Sher A. Functional diversity of helper T lymphocytes. Nature. 1996;383(6603):787–93. pmid:8893001
- 67. Zelenay S, van der Veen AG, Bottcher JP, Snelgrove KJ, Rogers N, Acton SE, et al. Cyclooxygenase-Dependent Tumor Growth through Evasion of Immunity. Cell. 2015;162(6):1257–70. pmid:26343581
- 68. Obermajer N, Muthuswamy R, Lesnock J, Edwards RP, Kalinski P. Positive feedback between PGE2 and COX2 redirects the differentiation of human dendritic cells toward stable myeloid-derived suppressor cells. Blood. 2011;118(20):5498–505. pmid:21972293
- 69. Veltman JD, Lambers ME, van Nimwegen M, Hendriks RW, Hoogsteden HC, Aerts JG, et al. COX-2 inhibition improves immunotherapy and is associated with decreased numbers of myeloid-derived suppressor cells in mesothelioma. Celecoxib influences MDSC function. BMC cancer. 2010;10:464. pmid:20804550
- 70. Sinha P, Clements VK, Fulton AM, Ostrand-Rosenberg S. Prostaglandin E2 promotes tumor progression by inducing myeloid-derived suppressor cells. Cancer Res 2007;67(9):4507–13. pmid:17483367
- 71. Hahn T, Alvarez I, Kobie JJ, Ramanathapuram L, Dial S, Fulton A, et al. Short-term dietary administration of celecoxib enhances the efficacy of tumor lysate-pulsed dendritic cell vaccines in treating murine breast cancer. International journal of cancer. 2006;118(9):2220–31. pmid:16331615