SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

Immunotherapy using dendritic cells (DC) has shown promising results. However, the use of an appropriate DC population is critical for the outcome of this treatment, and the search for an optimal DC subset is still ongoing. The DC used in immunotherapy today are usually matured with a cytokine cocktail consisting of TNF-α, IL-1β, IL-6 and PGE2. These cells have deficits in their cytokine production, particularly IL-12p70, mainly because of the presence of PGE2. Bromelain is a pineapple stem extract containing a mixture of proteases that has been used clinically in adjuvant cancer treatment. In this study, we analysed the effect of bromelain on human monocyte-derived DC. We added bromelain to the cytokine cocktail and modified cytokine cocktails with either no PGE2 or reduced amounts of PGE2, respectively. Combining bromelain with the cytokine cocktails containing PGE2 resulted in an increased surface expression of CD83, CD80 and CD86. The chemokine receptor CCR7 was also considerably upregulated in these DC populations compared with DC treated with the cytokine cocktail alone. Removal or reduction of PGE2 from the cytokine cocktail did not increase the IL-12p70 secretion from stimulated DC, and addition of bromelain to the different cytokine cocktails resulted in only a minor increase in IL-12p70 production. Moreover, combining bromelain with the cytokine cocktails did not improve the T cell stimulatory capacity of the generated DC populations. In conclusion, bromelain treatment of monocyte-derived DC does not improve the functional quality compared with the standard cytokine cocktail.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

Dendritic cells (DC) are professional antigen-presenting cells with the unique ability to stimulate naïve T cells [1]. Immature DC circulate in our bodies constantly sampling the surroundings for potential antigens. Upon encounter with an antigen in the presence of danger signals, DC start to mature and migrate toward the lymph node to present the captured antigens to T cells. When receiving this maturation stimulus, an upregulation of major histocompatibility complex (MHC) and costimulatory molecules are detected on the surface of DC, in addition to higher amounts of secreted inflammatory cytokines such as IL-12p70. These changes increase the ability of DC to stimulate T cells and activate the immune response [2].

One problem concerning immune responses towards tumours is that cancer cells have the ability to evade the immune system surveillance and thereby avoid being eliminated by effector cells [3, 4]. Owing to their outstanding ability to initiate immune responses, DC have, for a long time, been in the focus of immunotherapy. The development of protocols for the ex vivo generation of DC [5–7] led to the design and clinical application of tumour vaccination therapies using DC. Such DC vaccines aim to activate tumour-specific effector T cells [8]. Several trials have been performed the last decade [9–12]. However, the different steps of the protocol still need to be optimized. One element that needs improvement is the maturation of the DC. Cells used in trials today are often stimulated with the Jonuleit cytokine cocktail consisting of interleukin (IL)-1β, IL-6, tumour necrosis factor (TNF)-α and prostaglandin E2 (PGE2) [13]. Because these cells are lacking IL-12p70 production in addition to having low migratory capacity [14, 15], they are not optimal for inducing strong cell-mediated immune responses. Studies indicate that PGE2 is necessary for CCR7 surface expression on DC and for their potential to migrate [16]. Nevertheless, it has also been shown that PGE2 can be the cause for low IL-12p70 secretion [17, 18].

It is therefore an ongoing quest to find the optimal DC population for cancer immunotherapy. Bromelain is an extract from the stem of the pineapple plant (Ananas comosus). Immunological and enzymological data indicate that the crude extract contains different cysteine proteases and other compounds with distinct characteristics [19, 20]. Bromelain has been used in tropical health regimens for centuries, and the last decades, it has been used clinically as an additive to cancer treatment [19]. It has been shown to reduce side effects of chemotherapy, reduce skin tumour formation as well as to reduce oedema and improve wound healing after radiotherapy and surgery [19, 21, 22]. In human glioblastoma cells treated with bromelain, reduced adhesion, migration and invasive capacity were noted [23]. In addition to modulating cancer cells, bromelain has been shown to trigger and regulate cytokine production from different immune cells and affect the function of adhesion molecules on endothelial and blood cells [19]. As bromelain has the potential to activate and stimulate several different cell types, we have examined how it affects DC maturation. The aim was to analyse the DC maturation effect of bromelain, with respect to phenotype, cytokine production and T cell stimulatory capacity. Moreover, we investigated the possibility to replace PGE2 in the cytokine cocktail with bromelain.

We discovered that bromelain can be used as stimulus to induce DC maturation. Bromelain-stimulated DC revealed an upregulation of surface maturation markers, as well as an increased secretion of IL-12p70. When DC were stimulated with a combination of bromelain and the cytokine cocktail, an even more mature phenotype was detected. The T cell stimulatory capacity was, however, not changed. When PGE2 was removed from the cytokine cocktail, DC showed a less mature phenotype and lower ability to stimulate T cells. Addition of bromelain to this modified cytokine cocktail did not restore the DC maturation. We conclude that maturing DC with bromelain in vitro does not improve the functional quality of DC aimed to be used in cancer immunotherapy.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

Generation of DC.  Monocyte-derived DC were generated from buffy coat preparations obtained from healthy blood donors at the Blood Bank, Haukeland University Hospital, Bergen, Norway, as described previously [24]. In short, peripheral blood mononuclear cells (PBMC) were isolated by density gradient centrifugation before the monocytes were purified by plastic adherence. To generate immature DC, these cells were then cultured for 6 days in RP10 medium (RPMI 1640 (Cambrex Bioscience, Verviers, Belgium) with 10% FCS (PAA, Pasching, Austria), 100 units/ml penicillin and 100 μg/ml streptomycin (Sigma-Aldrich, St Louis, MO, USA), with IL-4 (20 ng/ml; ImmunoTools, Friesoythe, Germany) and GM-CSF (100 ng/ml; ImmunoTools). The cytokines were replenished every 2–3 days.

In initial experiments, different amounts of bromelain (100, 50, 25, 10 and 5 μg/ml; CPC W. Mühlbauer, Hamburg, Germany) were tested to analyse the effect of bromelain on DC and to determine the most suitable concentration. The maturation stimulus was given for 24 h, and cells were compared with immature DC. DC stimulated with the Jonuleit cytokine cocktail consisting of IL-1β (10 ng/ml), IL-6 (1000 U/ml), TNF-α (10 ng/ml; all from ImmunoTools) and PGE2 (1 μg/ml; Sigma-Aldrich) were used as a control. We next analysed the effect of combining bromelain with the cytokine cocktail. Included in this set-up were DC populations stimulated with the cytokine cocktail with less (¼) PGE2 (250 ng/ml) or without PGE2, both alone and in combination with bromelain. During harvesting of the generated DC populations, aliquots of conditioned medium were collected and stored at −20 °C. An automated CASY cell counter (Innovatis, Ueticon am See, Switzerland) was used to determine the amount of cells, cell size and viability.

Immunostaining.  The phenotypes of the generated cell populations were analysed using flow cytometry. The cells were stained for 10 min at room temperature with titrated amounts of antibodies in FACS buffer (PBS + 0.5% BSA) before washing and immediately analysed on a FACS Canto I cytometer (BD Biosciences, Heidelberg, Germany). The following antibodies were used: CD1-PE (NA1/34-HLK), CD8-PE (LT8), CD14-FITC (UCHM1), HLA-DR-APC (HL-39), CD38-Alexa Fluor 647(AT13/5), CD86-FITC (BU63), CD83-PE (HB15e), CD40-FITC (LOB7/6) (all from AbD Serotec, Düsseldorf, Germany), CCR7-PE (150503) (R&D systems, Minneapolis, MN, USA) and CD4-APC (MEM-241), CD80-APC (MEM-233) (ImmunoTools). FlowJo software (Tree Star, Ashland, OR, USA) was used for analyses. One percent false-positive events were accepted throughout the experiments.

Mixed lymphocyte reaction (MLR).  Twenty thousand DC were cultured with 2 × 105 allogeneic PBMC depleted for monocytes and stained with CFDA-SE (Invitrogen, Carlsbad, CA, USA). To improve the survival of the T cells, IL-2 (50 U/ml) and IL-7 (10 ng/ml; both from ImmunoTools) were added on the first day of coculture. On the fifth day of incubation, cells were harvested and analysed on a FACS Canto I flow cytometer.

Cytokine measurements.  The level of secreted IL-12p70 was measured in the conditioned medium by a sandwich ELISA according to the manufacturer’s protocol (BioLegend, San Diego, CA, USA).

Statistical analyses.  Statistical analyses were performed using GraphPad Prism, and the results were analysed using the Kruskal–Wallis test. Dunn’s post hoc test was used for comparisons of median values. The difference between groups was considered significant if P < 0.05.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

Bromelain treatment leads to maturation of DC

Five different concentrations of bromelain (100, 50, 25, 10 and 5 μg/ml) were tested to identify the bromelain concentration that would be the best stimulus. After 24 h of stimulation, cells were harvested and analysis of the cell size and viability revealed that cells stimulated with 25 μg/ml bromelain had the largest cell size and showed the highest viability of the different concentrations tested, comparable with cells stimulated with the cytokine cocktail (cytokine DC) (data not shown). DC matured with 100 μg/ml bromelain showed very low viability; we therefore did not include this concentration in further experiments.

Phenotypic analyses showed a concentration-dependent upregulation of costimulatory molecules and maturation markers after stimulation with bromelain (Fig. 1). The generated cells were all CD14 (not shown), confirming that the generation of DC had been successful. CD80 was higher expressed on bromelain-stimulated cells than on cytokine DC. In addition to CD80/CD86 expression, the costimulatory molecule CD40 is required for the induction of powerful T cell activation [25]. Stimulation with bromelain resulted in higher median fluorescence intensity (MFI) for CD40 compared with cytokine DC (Fig. 1D). Expression of the migration marker CCR7 was not increased upon bromelain treatment, but CD38 surface expression was significantly upregulated when compared with cytokine DC (Fig. 1C).

image

Figure 1.  Bromelain-stimulated dendritic cells (DC) show concentration-dependent upregulation of maturation markers and costimulatory molecules. Monocyte-derived DC were incubated for 24 h with different concentrations of bromelain and compared with immature DC and DC stimulated with the Jonuleit cytokine cocktail consisting of TNF-α, IL-1β, IL-6 and PGE2 (Cytokine DC). (A) Representative plots showing size and granularity of immature DC, cytokine cocktail–treated DC and DC stimulated with 25 μg/ml bromelain. (B) Representative histograms showing surface marker staining of immature DC, cytokine cocktail–treated DC and DC stimulated with 25 μg/ml bromelain. Percentage of cells expressing the surface markers (C) and the median fluorescence intensity (MFI) (D) were determined by flow cytometry. Results from six independent experiments (n = 4 for 5 μg/ml bromelain) are shown, and median is indicated (max/min and 25, 75 quartiles). The generated cell populations were CD14 (not shown).

Download figure to PowerPoint

None of the groups secreted high amounts of IL-12p70; however, cells stimulated with 25 μg/ml bromelain secreted slightly elevated amounts of IL-12p70 compared with cytokine DC (median 14.3 to 0 pg/ml, n = 7, data not shown). DC stimulated with higher and lower amounts of bromelain (50, 10 and 5 μg/ml) secreted less IL-12p70 than the cytokine cocktail-stimulated cells (data not shown).

DC stimulated with bromelain in combination with the cytokine cocktail have increased expression of surface maturation markers

We next analysed the effect of bromelain in combination with the cytokine cocktail. Because cytokine cocktail stimulation resulted in the most mature phenotype and stimulation with bromelain lead to a higher IL-12p70 secretion, we were interested to find out whether an additive or synergistic effect could be detected. We also tested bromelain combined with two modified versions of the cytokine cocktail containing less or no PGE2 as it has been stated that PGE2 is responsible for the lack of IL-12p70 production [17, 18]. The phenotype of the cells revealed that all DC populations stimulated with a combination of bromelain and the cytokine cocktail (original cocktail, ¼ of PGE2 and without PGE2) had a mature phenotype (Fig. 2), but the population with the least mature phenotype among these was the group that was stimulated with bromelain and the cytokine cocktail without any PGE2 (Fig. 2). The DC populations stimulated with bromelain in combinations with the cytokine cocktail and the cytokine cocktail with ¼ of PGE2 showed an even more mature phenotype compared with cytokine DC, with the highest CD86, CD80, CD83 and CCR7 surface expression (Fig. 2). Interestingly, a synergistic effect was detected on CD83 and CCR7 surface expression when bromelain was added to the original or modified cytokine cocktail with ¼ PGE2. We also analysed the migratory potential of the generated DC populations but could not detect any clear differences between the populations (data not shown).

image

Figure 2.  Dendritic cells (DC) stimulated with a combination of bromelain and cytokine cocktails containing PGE2 have a more mature phenotype than DC stimulated with only the cytokine cocktail. Monocyte-derived DC were stimulated for 24 h with the Jonuleit cytokine cocktail consisting of TNF-α, IL-1β, IL-6 and PGE2 (Cytokine DC) or modified versions of this cocktail either with ¼ PGE2 (Cytokine ¼ PGE2) or without PGE2 (Cytokine-PGE2). These three stimulating cocktails were also combined with 25 μg/ml bromelain (B+Cyto, B+Cyto-PGE2 and B+Cyto ¼ PGE2), and one DC population was stimulated with bromelain alone, or left immature, respectively. Percentage of cells expressing the surface markers (A) and the median fluorescence intensity (MFI) (B) were determined by flow cytometry. Results from four to seven independent experiments are shown, and median is indicated (max/min and 25, 75 quartiles). The generated cell populations were CD14 (not shown).

Download figure to PowerPoint

PGE2 is needed to ensure appropriate maturation with the cytokine cocktail

Removal of PGE2 from the cytokine cocktail resulted in reduced surface levels for most of the markers analysed compared with the original cytokine cocktail (Fig. 2). When ¼ of PGE2 was included in the cocktail, the surface expression was restored (Fig. 2). We also determined the MFI of these markers (Fig. 2B). All populations expressed comparable amounts of CD40. The density of surface CD38 was highest upon treatment with bromelain alone or in combination with the modified cytokine cocktail without PGE2. Treatment of the cells with the modified cytokine cocktail without PGE2 resulted in lowest surface expression of HLA-DR, similar to that of immature cells. HLA-DR was highest expressed on DC treated with a combination of bromelain and the cytokine cocktail (Fig. 2B).

DC stimulated with bromelain in combination with a modified cytokine cocktail without PGE2 secrete slightly higher amounts of IL-12p70

DC stimulated with a combination of bromelain and the cytokine cocktail did only produce higher amounts of IL-12p70 when PGE2 was completely removed from the cocktail (Fig. 3). However, this DC population had a less mature phenotype (Fig. 2). As expected, immature DC and DC stimulated with the cytokine cocktail alone did not produce considerable amounts of IL-12p70.

image

Figure 3.  Dendritic cells (DC) stimulated with bromelain in combination with the cytokine cocktail without PGE2 secrete highest, but still low, amounts of IL-12p70. Monocyte-derived DC were stimulated for 24 h with the Jonuleit cytokine cocktail consisting of TNF-α, IL-1β, IL-6 and PGE2 (Cytokine DC) or modified versions of this cocktail either with ¼ PGE2 (Cytokine ¼ PGE2) or without PGE2 (Cytokine-PGE2). These three cocktails were also combined with 25 μg/ml bromelain (B+Cyto, B+Cyto-PGE2 and B+Cyto ¼ PGE2). One DC population was stimulated with bromelain alone, and one was left immature. IL-12p70 production was analysed using a sandwich ELISA. Results from four to nine independent experiments are shown, and median is indicated.

Download figure to PowerPoint

All stimulated DC populations have a similar T cell stimulatory capacity

To analyse the functionality of the generated DC populations, we performed allogeneic MLR to assess the T cell stimulatory capacity. As shown in Fig. 4, immature DC had, as expected, the lowest capacity to stimulate allogeneic T cells. DC stimulated with the cytokine cocktail without PGE2 showed only a small increase in T cell stimulatory capacity compared with immature DC. All other DC populations had a slightly better ability to stimulate T cells.

image

Figure 4.  All mature dendritic cell (DC) populations have a similar capacity in stimulating T cells. Monocyte-derived DC were stimulated for 24 h with the Jonuleit cytokine cocktail consisting of TNF-α, IL-1β, IL-6 and PGE2, (Cytokine DC) or modified versions of this cocktail either with ¼ PGE2, (Cytokine ¼ PGE2) or without PGE2 (Cytokine-PGE2). These three stimulating cocktails were also combined with 25 μg/ml bromelain (B+Cyto, B+Cyto-PGE2 and B+Cyto ¼ PGE2). One DC population was stimulated with bromelain alone, and one was left immature. After 24 h stimulation, 2 × 104 DC were cocultured with 2 × 105 CFSE-labelled allogeneic peripheral blood mononuclear cells depleted of monocytes for 5 days before the number of proliferated T cells was analysed by flow cytometry. Representative figures from one of four independent experiments are shown.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

The maturation status of DC has an important role in initiating and directing antitumor immune responses [26]. A proper mature DC population is essential, because the quality of the DC vaccine-induced immune response never can be better than the quality of the DC population used. DC used in most clinical trials today are stimulated with the Jonuleit cytokine cocktail [13] referred to as the ‘gold standard’. The discussion concerning this cytokine cocktail is related to the use of PGE2. This inflammatory mediator has been shown to augment survival [27] and migration [28] of DC, in addition to be responsible for surface expression of the costimulatory molecules CD252 (OX40L) and CD70 needed for the stimulation of T cell proliferation [29]. However, PGE2 has also been demonstrated to be responsible for the lack of secreted IL-12p70 [17, 18], which is crucial for the activation of strong immune responses through the induction of Th1-type responses.

The intentions behind this study were to analyse the effect of bromelain on DC maturation and to investigate whether bromelain could replace PGE2 in the cytokine cocktail to overcome the negative effects of PGE2. Previous experiments performed with bromelain on glioma cells had shown that bromelain affects and alters glioma cells without causing any cellular toxicity at 50 μg/ml [23]. This was only partly confirmed during our experiments, as DC treated with 100 and 50 μg/ml of bromelain showed lower viability compared with cells treated with lower concentrations of bromelain. Stimulation with 25 μg/ml bromelain resulted in phenotypic mature DC that secreted more IL-12p70 than DC matured with the cytokine cocktail. When bromelain was combined with the cytokine cocktail, we discovered the existence of a synergistic effect, influencing the expression of some of the analysed surface markers. Clearly, higher levels of CCR7 and CD83 were detected when using bromelain in combination with the original cytokine cocktail or bromelain in combination with the cocktail with reduced amount of PGE2 as maturation stimulus. This synergistic effect was lost when bromelain was used in combination with the cytokine cocktail without any PGE2.

The migratory capacity of DC has been shown to be dependent on their surface expression of CCR7 [30], although we could recently show that CCR7 is not directly correlated with its ligand CCL19-driven chemotaxis [24]. PGE2 was shown to be responsible for the upregulation of CCR7 on the surface of DC [16]. In addition to the effect of CCR7 expression on DC, PGE2 was found to be important for induction of metalloproteinase-9, which is also important for the migration of DC [31]. This is consistent with our data, showing that surface expression of CCR7 is strikingly reduced when PGE2 is completely removed from the cytokine cocktail. However, migration of DC is complex, and another surface marker that has been linked to influence DC migration is the ectoenzyme and signalling receptor CD38 [32]. As with CCR7, we showed previously that the level of CD38 expression does not correlate with chemotaxis towards CCL19 [24]. Nevertheless, we could see that DC stimulated with bromelain or with bromelain in combination with the cytokine cocktail without PGE2 had noticeably higher MFI values for CD38 (Fig. 2B). Addition of reduced amounts of PGE2 did not increase the MFI. Thus, PGE2 had an inhibitory effect of CD38 expression on DC, similar to IL-12p70 production. Interestingly, a correlation between CD38 expression and IL-12p70 secretion of DC has been described previously [33], in agreement with our data.

The only DC population capable of producing higher amounts of IL-12p70 was DC stimulated with bromelain in combination with the cytokine cocktail without PGE2. We expected to find a higher secretion of IL-12p70 in the group stimulated with the cytokine cocktail without PGE2, as PGE2 has been claimed to be responsible for the lack of this cytokine, but our results indicate that it is not enough to only remove PGE2. In addition to not producing any notable amounts of IL-12p70, these DC also showed a less mature phenotype compared with the other groups, so obviously PGE2 is necessary for inducing (phenotypic) maturation. However, addition of bromelain could overcome this lack of stimulation. On the other hand, bromelain alone was not potent enough to induce both phenotypic maturation and high IL-12p70 production. The lack of IL-12p70 production was not a result of a general inability of the DC, as we detected large amounts of IL-12p70 after stimulation with the bacterial compound OK432 using DC from the same preparation [24].

Comparing the functionality of the generated DC populations in a MLR, we could show that PGE2 also influenced the T cell stimulatory capacity of the DC. When DC stimulated with the modified cytokine cocktail without PGE2 were cocultured with lymphocytes, fewer proliferative T cells were detected. Addition of ¼ of PGE2 to the cocktail improved this stimulatory capacity. This was also true regarding the phenotype of the cells. Use of ¼ of the amount of PGE2 in the cocktail increased the expression of surface maturation markers, and some markers had even higher surface expression using this stimulation than with the original cytokine cocktail. Addition of bromelain to both the original and the modified cytokine cocktail with reduced PGE2 resulted in an even more mature phenotype, but this phenotype had an insufficient secretion of IL-12p70.

Because IL-12p70 is essential for a strong induction of cytotoxic T lymphocyte (CTL) responses, several other attempts to generate DC with high IL-12p70 secretion have been made by other research groups. Stimulation with polyriboinosinic polyribocytidylic acid (poly I:C) has shown to generate DC capable of producing high amounts of IL-12p70 [34, 35]. However, these DC have poor migratory abilities compared with PGE2-stimulated DC [36]. Attempts to utilize the strength of poly I:C has been made by stimulation with poly I:C in combination with TLR 7/8 ligands in addition to PGE2 [37] and in a two-step maturation where poly I:C was added after the Jonuleit cytokine cocktail [38]. These studies showed that combining poly I:C with PGE2 stimulation results in DC with both high IL-12p70 secretion and enhanced migratory capacity, although it has been claimed that mature DC differentiate into either cytokine-producing or migratory cells [39]. As we discovered a synergistic effect when bromelain was combined with the cytokine cocktail, it might also be interesting to test bromelain in combination with other stimulating agents in a two-step maturation protocol.

In conclusion, we could show that bromelain can be used to stimulate DC, but these DC have a less mature phenotype than those stimulated with the ‘gold standard’ cytokine cocktail. Addition of bromelain to the cytokine cocktail or to a modified cytokine cocktail with reduced amounts of PGE2 resulted in cells with a more mature phenotype than that of cytokine DC characterized by higher CD83 and CCR7 expression, but without sufficient IL-12p70 secretion. Removal of PGE2 from the cocktail did not increase the IL-12p70 secretion from DC, but addition of bromelain did result in detectable amounts of IL-12p70. Moreover, PGE2 was found to augment T cell responses in the MLR assay and to induce synergistic effects on CD83 and CCR7 expression on DC stimulated with bromelain in combination with the cytokine cocktail. However, bromelain treatment of monocyte-derived DC does not seem to improve the functional quality of DC significantly compared with the standard cytokine cocktail.

Acknowledgment

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

This work was supported by Bergen Translational Research Fund, The Bergen Research Foundation, The Norwegian Cancer Society, Kreftforeningens paraplystiftelse for kreftforskning and the Broegelmann Legacy. We thank Dagny Ann Sandnes for excellent technical assistance.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References