Tumor-derived immuno-modulators induce overlapping pro-tolerogenic gene expression signatures in human dendritic cells

Human Immunology xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

www.ashi-hla.org

journal homepage: www.elsevier.com/locate/humimm

Tumor-derived immuno-modulators induce overlapping pro-tolerogenic gene expression signatures in human dendritic cells Stefanie Trojandt, Iris Bellinghausen, Angelika B. Reske-Kunz, Matthias Bros ⇑ Department of Dermatology, Johannes Gutenberg-University, Mainz, Obere Zahlbacher Str. 63, D-55131 Mainz, Germany

a r t i c l e

i n f o

Article history: Received 19 November 2015 Revised 18 August 2016 Accepted 30 August 2016 Available online xxxx Keywords: Dendritic cells Tolerogenic Dexamethason Interleukin-10 Prostaglandin E2

a b s t r a c t Immature dendritic cells (iDCs) and tolerogenic DCs are essential for the induction and maintenance of peripheral tolerance. Tumors produce immuno-modulatory factors which imprint a pro-tolerogenic, maturation-resistant state in DCs. Here we asked for common markers of differentially tolerized human monocyte-derived DC populations. For this, PBMC-derived monocytes were differentiated to DCs in the presence of established immuno-modulators as released by tumors (IL-6, IL-10, TGF-b, glucocorticoid [GC], prostaglandin E2 [PGE2]). Most unstimulated pro-tolerogenic DC populations commonly overexpressed some tolerance-associated markers (ILT-4, IL-10, HO-1) as compared with iDCs. These markers may contribute to imprint a pro-tolerogenic state in DCs. Furthermore, some tolerance markers were overexpressed in an immuno-modulator specific manner in DCs differentiated in the presence of TGF-b (overexpressed tolerance markers: B7-H3, CD103, TGF-b1), IL-10 (B7-DC, ILT-3) and PGE2 (IDO). Upon stimulation, matured control DCs (mDCs) down-regulated most pro-tolerogenic markers monitored, while some were upregulated (IkBa, IDO, B7-H1, B7-DC). In contrast, the different groups of tolerized DCs largely retained expression of pro-tolerogenic markers after stimulation. In contrast to mDCs, most groups of tolerized DCs showed impaired upregulation of CD80, and all groups retained IL-10 cytokine production after stimulation. All tolerized DC populations commonly exerted an attenuated allogenic T cell stimulatory capacity as compared with mDCs. Ó 2016 Published by Elsevier Inc. on behalf of American Society for Histocompatibility and Immunogenetics.

1. Introduction Under steady state conditions, DCs induce and maintain peripheral tolerance by presenting self and harmless environmental antigens in the context of low costimulation and the absence of proinflammatory cytokines [1]. By this, natural regulatory T cells (Treg) that are specific for self antigens are expanded, and naive T cells that recognize environmental antigens are converted to Treg. In response to pathogen-associated and endogenous danger signals DCs are activated, and upregulate expression of costimulators and T cell-polarizing proinflammatory mediators to induce T cell responses [2]. The immune system has also developed strategies to limit the extent and duration of immune responses [3]. These mechanisms aim to reduce inflammation-associated tissue destruction, and the induction of autoimmune responses. In this regard, proinflammatory cytokines stimulate the hypothalamicpituitary-adrenal axis to upregulate production of cortisol that

⇑ Corresponding author. E-mail address: [email protected] (M. Bros).

exerts broad anti-inflammatory properties [4]. Hence, synthetic GCs are frequently prescribed in the treatment of autoimmune diseases and severe allergic diseases. Activation of DCs also results in upregulated expression of endogenous inhibitors like IkBa to limit sustained NF-jB dependent DC activation [5]. Besides inhibitors that exert autocrine effects, other stimulation-induced inhibitors may affect surrounding T cells in a paracrine manner as well. In this regard, stimulation-induced IDO which depletes tryptophan from the DC micro-environment to generate pro-apoptotic kynurenines plays an important role [6]. Monocyte-derived DCs constitute the model of choice for analysis of human DCs in vitro, and after differentiation they may be stimulated by a well-established cocktail of endogenously produced proinflammatory mediators [7]. However, when applied to DC progenitors, components of this maturation cocktail were reported to induce myeloid-derived suppressor cells, termed MDSCs (PGE2 [8]), and pro-tolerogenic DCs (TNF-a [9], IL-6 [10]). Therefore, these mediators exert differential effects on DC progenitors (inhibitory) and differentiated DCs (stimulatory). Cytokines like IL-10 [11] and TGF-b [12] tolerize DCs, and promote Treg

http://dx.doi.org/10.1016/j.humimm.2016.08.014 0198-8859/Ó 2016 Published by Elsevier Inc. on behalf of American Society for Histocompatibility and Immunogenetics.

Please cite this article in press as: S. Trojandt et al., Tumor-derived immuno-modulators induce overlapping pro-tolerogenic gene expression signatures in human dendritic cells, Hum. Immunol. (2016), http://dx.doi.org/10.1016/j.humimm.2016.08.014

2

S. Trojandt et al. / Human Immunology xxx (2016) xxx–xxx

induction. All of the aforementioned immuno-modulators have been found expressed by tumors to induce MDSCs and to tolerize DCs [10]. Besides, different types of tumors were shown to produce adrenocorticotropic hormone which may stimulate production of anti-inflammatory cortisol [13]. In this comparative study, we analysed the characteristics of pro-tolerogenic human monocyte-derived DCs as induced by long-term treatment with different immuno-modulators in the course of DC differentiation. These DC populations convergingly upregulated expression of HO-1, ILT-4, and IL-10 on transcriptional level. Furthermore, groups of DCs showed considerable overlaps in their patterns of pro-tolerogenic marker expression at unstimulated state. In contrast to control DCs, DC populations differentiated in the presence of immuno-modulators retained expression of pro-tolerogenic markers and production of IL-10 cytokine after stimulation. Accordingly, these pro-tolerogenic DCs exerted impaired allo CD4+ T cell stimulatory capacity.

2. Materials and methods 2.1. Reagents GM-CSF was obtained from Berlex (Seattle, WA), IL-4, IL-6, IL-10, and TGF-b3 from ImmunoTools (Friesoythe, Germany), and IL-1b and TNF-a from Peprotech (Hamburg, Germany). All cytokines used were of recombinant human origin. Prostaglandin E2 (PGE2) was purchased from Alexis (Lausen, Switzerland) and dexamethason (DEX) from Sigma-Aldrich (Deisenhofen, Germany). 2.2. Cells Human monocyte-derived DCs were generated from peripheral blood mononuclear cells (PBMCs) as described [7] with some minor modifications. PBMCs were depleted of CD4+ T cells by immunomagnetic separation using anti-CD4 antibody-conjugated immunomagnetic beads (Miltenyi, Bergisch-Gladbach, Germany) as recommended. T cell-depleted PBMC from three donors were pooled for each experiment. Monocytes were enriched by plastic adherence, and iDCs were generated from these by a 6-day culture in IMDM, supplemented with 2% autologous plasma, 200 IU/ml GM-CSF, and 1000 IU/ml IL-4. In parallel cultures, IL-6 (1000 IU/ ml), IL-10 (20 ng/ml), TGF-b3 (20 ng/ml), PGE2 (1 lg/ml), or DEX (1 lM) was added at the onset of DC culture. These agents were replenished in the course of change of culture media on days 3 and 6. Maturation of DCs was induced with a stimulation cocktail (1000 IU/ml IL-1b, 10 ng/ml TNF-a, and 1 lg/ml PGE2) for 2 days. mDC cultures contained >90% of HLA-DR+ cells, and <5% of CD14+ cells as assessed by flow cytometry (see Fig. S1). CD4+ T cells positively isolated by MACS technology (purity >90% as assessed by flow cytometry) served as responders in allogenic mixed lymphocyte reactions (MLR; see below). 2.3. Flow cytometry DCs washed in staining buffer (PBS/2% FCS) were incubated with labelled monoclonal antibody (mAb) for 20 min at 4 °C. Fluorescein isothiocyanate-conjugated mAb recognized HLA-DR (L243; BioLegend, San Diego, CA). Phycoerythrin-conjugated mAbs were specific for CD14 (M5E2), CD80 (2D10), and CD86 (IT2.2), all purchased from BioLegend (San Diego, CA), and for CD83 (HB15e) obtained from BD Pharmingen (Heidelberg, Germany). Appropriate isotype controls were used. All mAbs were of rat origin. Flow cytometric analysis was performed using a FACScan flow cytometer (BD Biosciences) equipped with CellQuest Software.

2.4. IL-10 detection IL-10 was quantified using a sandwich-ELISA. ELISA capture (JES3-9D7 and biotinylated detection Abs for IL-10 were obtained from BD Pharmingen. Recombinant mouse IL-10 used for ELISA standards was purchased from ImmunoTools. 2.5. Real-time RT-PCR analysis Total RNA was isolated from at least 5  105 DCs per isolation by using the RNeasy MiniPrep kit and performing on-column DNase treatment (both from Qiagen, Hilden, Germany) as recommended. RNA was reverse-transcribed applying a 1:1 mix of Oligo-dT and random hexamer primers by using iScript (Bio-Rad, Munich, Germany) as recommended. Gapdh served as an internal control. Primers were purchased from eurofins MWG Synthesis (Ebersberg, Germany) and according sequences are listed in Tab. S1. Real-time PCR reactions were performed and analysed as described [14]. 2.6. MLR Graded numbers of irradiated (30 Gy) DCs were cocultured in triplicates with 105 sorted allogenic CD4+ T cells for five days on 96-well tissue culture plates. DCs and T cells cultured alone served as internal negative controls. Cell proliferation was assessed by the uptake of [3H] thymidine (0.25 lCi/well) for the last 16 h of MLR cultures. Cells were harvested onto glass fiber filters and retained radioactivity was measured in a liquid scintillation counter (1205 Betaplate; LKB Wallac, Turcu, Finland). 2.7. Database analysis Potential interactions of tolerance markers as upregulated in the differentially tolerized DC populations was visualized using STRING database v10 (http://string-db.org [15]). The probability of interaction correlates with the thickness of connecting lines (confidence view). 2.8. Statistical analysis Data are presented as means ± SEM and were analysed for statistically significant differences by applying one-way ANOVA followed by Student’s t-test for comparison of groups (SigmaPlot 12.3, Systat Software, San Jose, CA). 3. Results 3.1. DCs differentiated in the presence of immuno-modulators express less costimulators, except for IL6-DCs PBMC-derived monocytes were differentiated using GM-CSF and IL-4 in the presence of immuno-modulatory agents at doses established in the literature (see Section 2.2, and legend of Fig. 1). Application of the various immuno-modulators yielded no major influence on cell morphology, viability, and numbers of cells harvested (data not shown). All cell populations displayed similar levels of expression of HLA-DR which was slightly increased in response to stimulation, except for cells conditioned with IL-10 (Fig. 1). Major fractions of HLA-DR+ cells of all groups were CD86+. Expression of CD86 increased in all groups after stimulation with the maturation cocktail. These findings indicate that neither immuno-modulator affected the acquisition of a typical DC phenotype (HLA-DR+CD86+). Under unstimulated conditions, only minor fractions of the different DC populations showed CD80 surface expression. In

Please cite this article in press as: S. Trojandt et al., Tumor-derived immuno-modulators induce overlapping pro-tolerogenic gene expression signatures in human dendritic cells, Hum. Immunol. (2016), http://dx.doi.org/10.1016/j.humimm.2016.08.014

S. Trojandt et al. / Human Immunology xxx (2016) xxx–xxx

3

Fig. 1. DCs differentiated in the presence of IL-10, PGE2 and DEX show impaired upregulation of CD80 and CD83 in response to stimulation. DCs were differentiated from plate-adherent monocytes using GM-CSF (200 IU/ml) and IL-4 (1000 IU/ml). In parallel cultures, immuno-modulators were applied at the onset of DC culture (IL-6: 1000 IU/ml [26], IL-10: 20 ng/ml [21,27], TGF-b3: 20 ng/ml [12], PGE2: 1 lg/ml [8], DEX: 1 lM [14,30]). Untreated DCs served as a control (Ctrl). On day 6, aliquots were stimulated with a maturation cocktail (1000 IU/ml IL-1b, 10 ng/ml TNF-a, and 1 lg/ml PGE2) for 2 days. Cell populations were incubated with mAb specific for the indicated cell surface markers and subjected to flow cytometry. Graphs represent the frequencies of cells expressing HLA-DR (upper left graph), and HLA-DR+ cells that coexpressed the indicated surface marker (other graphs). Data represent the mean ± SEM of 3–4 independent experiments each. Statistically significant differences: *versus unstimulated Ctrl DCs (iDC), and DCs differentiated in the presence of immuno-modulators at stimulated state $versus corresponding unstimulated state, and #versus stimulated Ctrl DCs (*,$,#p < 0.05; $$ p < 0.01; ***,###p < 0.001).

response to stimulation, the majority of mDCs (which represent iDCs stimulated with the cocktail) and IL6-DCs expressed CD80 to a comparable extent. However, the stimulation-associated increase in CD80+ expression was lower in the other groups of DCs. At basal state, only small subsets of HLA-DR+ cells of either DC population expressed CD83. Most groups of DCs upregulated CD83 in response to stimulation, but to variable extent. While the majority of mDCs, IL6-DCs and TGFb-DCs were CD83+, only smaller fractions of IL10-DCs and DEX-DCs expressed this DC maturation marker. PGE-DCs were largely refractory to stimulationdependent upregulation of CD83. These findings suggest that IL10 and PGE2 potently prevent stimulation-associated upregulation of the maturation markers CD80 and CD83 on human DCs, followed by DEX in potency. 3.2. DCs differentiated in the presence of immuno-modulators retain IL-10 production after stimulation Antigen presenting cells (APCs) shape T cell responses largely by production of cytokines. iDCs produced considerable amounts

of the dual Treg/Th2-promoting cytokine IL-10 (Fig. 2). DCs differentiated in the presence of immuno-modulators showed unaltered IL-10 production, except for TGFb-DCs which generated significantly less IL-10 than iDCs. mDCs produced markedly less IL-10 than iDCs. In contrast, DCs differentiated with either immunomodulator after stimulation generated IL-10 at comparable or higher concentrations than without stimulation. Therefore, all immuno-modulators convergingly prevented stimulationassociated down-regulation of IL-10 production.

3.3. Differentiation and stimulation of DCs in the presence of immunomodulators impairs stimulation-induced acquisition of APC activity At unstimulated state, all DC populations induced proliferation of allogenic CD4+ T cells (Fig. 3) to a low extent. mDCs displayed a significantly higher T cell stimulatory capacity than iDCs. On the contrary, all groups of DC differentiated and stimulated in the presence of an immuno-modulator congruently showed an attenuated T cell stimulatory capacity as compared with mDCs. This

Please cite this article in press as: S. Trojandt et al., Tumor-derived immuno-modulators induce overlapping pro-tolerogenic gene expression signatures in human dendritic cells, Hum. Immunol. (2016), http://dx.doi.org/10.1016/j.humimm.2016.08.014

4

S. Trojandt et al. / Human Immunology xxx (2016) xxx–xxx

Fig. 2. DCs differentiated in the presence of immuno-modulators retain IL-10 production after stimulation. DC populations were generated as described (see legend of Fig. 1). DC culture supernatants were assessed for contents of IL-10 by ELISA. Data represent the mean ± SEM of 3–4 independent experiments each, and were normalized to the cytokine content in iDCs (102.41 ± 43.81 ng/ml) arbitrarily set to one. Statistically significant differences: *versus iDCs (**p < 0.01; ***p < 0.001). nd = not done.

observation confirms the induction of a pro-tolerogenic state of activity in DCs by the immuno-modulators. 3.4. pro-tolerogenic DCs display distinct gene expression signatures Next, the transcriptional pattern of pro-tolerogenic markers in the differentially tolerized groups of DCs was assessed. For this, genes that encode modulators of signalling and gene expression (IkBa [5]; MKP-1 [16], GILZ [17], surface receptors (B7-H1 [18], B7-DC [19], B7-H3 [20], ILT-3 and ILT-4 [21], CD103 [22]), cytokines (IL10 [11], IL1RA [23], TGF-b1 [12]) and intracellular enzymes (HO-1 [24], IDO [6]) known to exert pro-tolerogenic activities were selected. mRNA expression levels of these pro-tolerogenic markers in DCs that were differentiated and stimulated in the presence of the various immuno-modulators are depicted in Fig. 4. Based on these results, qualitative alterations in the transcriptional pattern of pro-tolerogenic markers in the different DC populations are visualized in Fig. 5. At unstimulated state, several of the markers monitored were commonly upregulated (HO-1, ILT-4, IL-10) by 4 out of 5 groups of DCs that were differentiated in the presence of immunomodulators, while only one (ILT-3) was down-regulated by most groups (Figs. 4, 5A). Of all groups, IL10-DCs (7 markers) and TGFb-DCs (8 markers) upregulated most of the markers assessed, while DEX-DCs (7 markers) down-regulated most of the markers. IL6-DCs and DEX-DCs (up: HO-1, ILT-4, IL-10, down: IDO, B7-H3, ILT-3), as well as IL10-DCs and DEX-DCs (up: GILZ, HO-1, ILT-4, IL-10; down: B7-H3) showed the largest overlaps in coordinately regulated pro-tolerogenic markers. Besides, only TGFb-DCs (B7H3, CD103, TGF-b1), IL10-DCs (B7-DC, ILT-3) and PGE-DCs (IDO) showed unique upregulation of pro-tolerogenic markers. Database analysis suggested that within the groups of upregulated tolerogenic markers most molecules may interact on protein level, except for IDO, ILT-3, ILT-4, and B7-H3 (Fig. 5B). In this regard, IL-10 showed the highest number of potential interactants, followed by TGF-b, and MKP-1 and IL1RA. Stimulation of iDCs with the maturation cocktail in the absence of immuno-modulators to generate mDCs yielded down-regulation

of most pro-tolerogenic markers, while only a subset was upregulated (IkBa, IDO, B7-H1, B7-DC) (Figs. 4, S1). In contrast, DCs differentiated in the presence of immuno-modulators preserved the expression state of most of these pro-tolerogenic markers after stimulation. Most DC groups commonly upregulated IDO (except for PGE-DCs) and B7-DC (except for IL10-DCs) in a stimulationdependent manner. Furthermore, the other pro-tolerogenic markers that were elevated in expression by mDCs (IkBa, B7-H1) were also upregulated by IL6-DCs and DEX-DCs in a stimulationassociated manner. As compared with mDCs, all groups of DCs that were differentiated and stimulated in the presence of immuno-modulators showed either an unaltered or upregulated expression of pro-tolerogenic markers, with the exception of B7-H1 as downregulated by PGE-DCs (Figs. 4, 5C). Most groups upregulated ILT4, GILZ, HO-1, and IL-10. The group of stimulated DEX-DCs upregulated the highest number of markers monitored (GILZ, MKP-1, HO-1, ILT-4, IL-10), and displayed the most pronounced overlap with others groups, namely PGE-DCs (GILZ, HO-1, ILT-4), IL10DCs (MKP-1, ILT-4, IL-10) and IL6-DCs (HO-1, ILT-4, IL-10). Of these coordinately upregulated tolerogenic markers GILZ, IL-10 and MKP-1 potentially interact with each other on protein level (Fig. 5D). Altogether, these results indicate that differentiation of DCs in the presence of different immuno-modulators resulted in upregulated expression of partially overlapping sets of pro-tolerogenic markers. Stimulation-associated down-regulation of most protolerogenic markers in mDCs may constitute a prerequisite to acquire potent T cell stimulatory capacity. Preserved expression of these molecules in stimulated DCs differentiated and stimulated in the presence of immuno-modulators may contribute to their impaired T cell stimulatory activity. In this regard, retained IL-10 cytokine production by tolerized DCs may play an important role.

4. Discussion Human monocyte-derived iDCs have been demonstrated to exert tolerogenic activity by inducing Treg in an IL-10 and TGF-b dependent manner [25]. In response to stimulation with a maturation cocktail these DCs strongly upregulate activation markers and accordingly acquire potent T cell stimulatory capacity [7]. For distinct immuno-modulators that act via different signalling pathways imprintment of a pro-tolerogenic, partially maturation-resistant state on DCs has been shown (IL-6 [26], IL-10 [11], TGF-b [12], PGE2 [8], DEX [14]). Here we investigated the efficacy of each immuno-modulator to alter the immuno-phenotype and T cell stimulatory capacity of DCs in a comparative study. We aimed to identify pro-tolerogenic markers that were commonly regulated in DC populations tolerized by different immuno-modulators. Such markers may constitute key molecules essential for DC tolerization. We show that DCs differentiated in the presence of the immuno-modulators IL-10, DEX, PGE2, or IL-6 convergingly upregulated HO-1, IL-10 and ILT-4 on the transcriptional level. The heme-degrading enzyme HO-1 exerts broad antiinflammatory effects by direct inhibition of NF-jB on the protein level, and by the activities of the heme-degradation products biliverdin, ferritin and CO [24]. HO-1 may play a general role in dampening immune responses based on its inducible expression in different (immune) cell types by a great variety of ligands. Constitutive IL-10 production by tolerogenic DCs is a hallmark of so-called DC-10 that induce Tr1-termed Treg in an IL-10 dependent manner [27]. Of note, Tr1 cells exert inhibitory effects on immune cells via IL-10 secretion as well. In our study, elevated IL-10 mRNA production as induced by differentiation of DCs in the presence of immuno-modulators was not accompanied by

Please cite this article in press as: S. Trojandt et al., Tumor-derived immuno-modulators induce overlapping pro-tolerogenic gene expression signatures in human dendritic cells, Hum. Immunol. (2016), http://dx.doi.org/10.1016/j.humimm.2016.08.014

S. Trojandt et al. / Human Immunology xxx (2016) xxx–xxx

5

Fig. 3. Differentiation of DCs in the presence of immuno-modulators prevents the acquisition of potent stimulation-induced T cell proliferation inducing capacity. DC populations were generated as described (see legend of Fig. 1). Irradiated DCs (2x104) were cocultured with 105 sorted allogenic CD4+ T cells for 5 days in 0.2 ml culture medium in triplicates. T cell proliferation was assessed by uptake of [3H] thymidine for the final 16 h of culture. Data represent the mean ± SEM of triplicates. Graphs are representative of 4–6 independent experiments each. Statistically significant differences: *versus iDCs, #DCs differentiated in the presence of immuno-modulators at stimulated state versus stimulated control DCs (mDCs) (*,#p < 0.05; **,##p < 0.01; ***,###p < 0.001). Other statistically significant differences were omitted for reasons of clarity. DCs and T cells cultured alone served as negative controls, and showed no proliferation.

enhanced IL-10 cytokine levels in DC culture supernatants. This could be due to post-transcriptional inhibition of IL-10 mRNA translation [28]. In addition, enhanced consumption of IL-10 by the producer cells may play a role. In this regard, for IL-10 a positive feedback loop has been demonstrated in human macrophages that involved IL-10-induced STAT-3 activation, which in turn trans-activated the IL-10 gene promoter [29]. In human B cells, GCs were shown to activate STAT-3 as well, which yielded enhanced IL-10 expression [30]. PGE2, which engaged EP2/4 receptors of human DCs, resulted in activation of CREB in a cAMP/PKAdependent manner [31]. Activated CREB trans-activated IL-10 expression via two CRE sites within the human IL-10 promoter.

PGE2-induced IL-10 production may be enhanced by cAMP/ PKA-induced expression of COX-2 [8], which constitutes the ratelimiting enzyme of PGE2 production. Furthermore, cAMP was reported to induce expression of C/EBP transcription factors in monocytic cell lines, which trans-activated the human IL-10 promoter by binding to according regulatory sites [32]. The main signalling pathways known to be engaged by tolerizing factors to elevate IL-10 and other tolerance markers (ILT-4, GILZ) in immune cells are summarized in Fig. S3. ILT-4 expression is a characteristic of pro-tolerogenic DC populations, including DC10 [27]. ILT-4 binds HLA-G receptors of contacting immune cells. Engagement of ILT-4 on DCs was

Please cite this article in press as: S. Trojandt et al., Tumor-derived immuno-modulators induce overlapping pro-tolerogenic gene expression signatures in human dendritic cells, Hum. Immunol. (2016), http://dx.doi.org/10.1016/j.humimm.2016.08.014

6

S. Trojandt et al. / Human Immunology xxx (2016) xxx–xxx

Fig. 4. Expression of pro-tolerogenic markers in pro-tolerogenic DC populations. Expression of mRNA species that encode molecules with pro-tolerogenic function was monitored in DCs differentiated and stimulated in the presence of different immuno-modulators as described (see legend of Fig. 1). Data represent mean ± SEM of 3–6 experiments each, and were normalized to the according expression level in iDCs arbitrarily set to one. Statistically significant differences: *versus iDCs. DCs differentiated in the presence of immuno-modulators at stimulated state $versus corresponding unstimulated state, and #versus mDCs (*,$,#p < 0.05; **,$$,##p < 0.01; ***,$$$,###p < 0.001).

Please cite this article in press as: S. Trojandt et al., Tumor-derived immuno-modulators induce overlapping pro-tolerogenic gene expression signatures in human dendritic cells, Hum. Immunol. (2016), http://dx.doi.org/10.1016/j.humimm.2016.08.014

S. Trojandt et al. / Human Immunology xxx (2016) xxx–xxx

7

Fig. 5. Differentially tolerized DCs show partially overlapping gene expression signatures. Based on the data shown in Fig. 4, qualitative changes in mRNA expression levels of pro-tolerogenic molecules in unstimulated DCs (A) and stimulated DCs (C) differentiated in the presence of immuno-modulators as compared with the respective control cell population (A: iDCs, C: mDCs) are indicated. (Stimulation-induced alterations in pro-tolerogenic marker expression in the different DC populations are given in Fig. S2). (B, D) Based on database analysis, potential interaction of pro-tolerogenic markers as upregulated by unstimulated DCs (B) and stimulated DCs (D) in the presence of immunomodulators is visualized. The probability of protein-protein interaction correlates with the thickness of marker-connecting lines.

reported to induce IL-6 expression, which in an autocrine manner activated the tolerance-associated transcription factor STAT-3 [33]. Activated STAT-3 in turn preserved a maturation-resistant protolerogenic state in DCs. As shown in our study, three of the four groups of DCs differentiated in the presence of the aforementioned immuno-modulators upregulated ILT-4, but down-regulated ILT-3 mRNA expression. ILT-3 and ILT-4 belong to a group of colocalized genes that encode immunoglobulin-like receptors [34]. Both receptors bear a high degree of structural homology and exert inhibitory

functions. Hence, upregulation of ILT-4 may (over-)compensate down-regulation of ILT-3. Analysis of the ILT-4 gene core promoter identified myeloidassociated transcription factors as relevant for constitutive basal expression [35]. IL-10 was previously shown to upregulate ILT-4 in human unstimulated DCs [21]. IL-10 is known to signal primarily via the JAK/STAT-3 pathway [11]. For human iDCs that were cocultured with tumor cells, cell contact-dependent induction of STAT-3 was observed [36]. This was accompanied by enhanced

Please cite this article in press as: S. Trojandt et al., Tumor-derived immuno-modulators induce overlapping pro-tolerogenic gene expression signatures in human dendritic cells, Hum. Immunol. (2016), http://dx.doi.org/10.1016/j.humimm.2016.08.014

8

S. Trojandt et al. / Human Immunology xxx (2016) xxx–xxx

expression of ILT-4. Based on these findings, we suggest that in IL10-DCs IL-10 may induce ILT-4 via STAT-3 activation as well. For the ILT-4 gene, histone acetylation was required to allow access of transcription factors to yield gene expression [35]. PGE2/PKA was reported to inhibit the activity of histone deacetylases in a global manner, and thereby favours histone acetylation [37]. GCs were demonstrated to regulate the histone acetylation state in a promoter-specific manner [38]. Hence, in our study in DCs differentiated in the presence of either agent (PGE2, DEX) increased ILT-4 mRNA expression may depend on epigenetic regulation. DCs differentiated in the presence of DEX, PGE2, or IL-10 (but not IL-6) upregulated GILZ mRNA in addition to HO-1, IL-10, and ILT-4. GILZ expression is induced by GC, and binds RAF and RAS proteins that interfere with AKT and ERK signalling [17]. In addition, GILZ was demonstrated to engage NF-jB and AP-1, and to sequester the activity of either transcription factor. Overexpression of GILZ in human DCs promoted the generation of maturationresistant, pro-tolerogenic human DCs that induced Treg [39]. GC-mediated expression of GILZ mRNA is facilitated by binding of the activated GC receptor to several GRE sites within the GILZ promoter region [40]. Besides these GRE sites, the proximal GILZ promoter region was reported to harbour a functional CRE site [41] which may explain PGE2-induced GILZ mRNA expression. Furthermore, IL-10 was shown to elevate GILZ expression in human macrophages by a yet non-identified signalling pathway [42]. IL6-DCs and DEX-DCs, besides upregulating the three aforementioned pro-tolerogenic markers (HO-1, ILT-4, IL-10) were characterized by diminished expression of an overlapping set of markers (IDO, B7-H3, ILT-3). Similar to IL-10 [29], IL-6 was shown to induce JAK-dependent activation of STAT3 and of ERK that may induce C/EBPb expression [43]. GC activate the GC receptor, which in turn binds NF-jB, AP-1, and other transcription factors to inhibit their activity. Further studies are required to elucidate whether IL6-DCs and DEX-DCs bear yet unknown common signalling pathways that hold responsible for the down-regulation of the overlapping set of pro-tolerogenic markers. Some tolerance markers were upregulated in an immunomodulator-specific manner. Thus, TGF-b-induced SMAD activation [12] may contribute to elevated expression of a subset of protolerogenic markers (B7-H3, CD103, TGF-b1) in TGFb-DCs. Of these, only for CD103 SMAD-mediated induction has been described [44]. Unstimulated PGE-DCs were characterized by specific upregulation of IDO mRNA expression which is in accordance with a previous study [45]. Unstimulated IL10-DCs showed elevated expression levels of B7-DC and ILT-3 mRNA. The latter finding is in line with a recent report on Tr1-induced ILT-3 expression mediated by IL-10 [46]. Further studies are required to elucidate whether assessment of markers that are specifically induced by single immuno-modulators may serve to delineate, which immunomodulator(s) play an immuno-suppressive role in cancer patients, by testing PBMCs or isolated primary DCs. Imprintment of a maturation-resistant state in human DCs as induced by immuno-modulators was evidenced by an impaired T cell stimulatory capacity as compared with mDCs. In accordance, most pro-tolerogenic DC populations were characterized by attenuated upregulation of the activation surface marker CD80 in response to stimulation. Only IL6-DCs were fully responsive towards stimulation in this regard. Therefore, IL-6 which induces STAT-3 activation, attributed to be responsible for inducing tolerogenic DCs, may be less effective than IL-10, reported to induce the same signalling pathway [11]. Alternatively, other IL-6 triggered signalling pathways [43] may counter-act in part STAT-3mediated inhibition. As compared with iDCs, mDCs were characterized by downregulated IL-10 production. In contrast, all other stimulated pro-tolerogenic DC populations retained production of this STAT-

3 activating anti-inflammatory cytokine. Stimulation of iDCs resulted in down-regulation of most pro-tolerogenic markers assessed, except for IkBa [5], IDO [6], B7-H1 [18] and B7-DC [19]. These molecules constitute stimulation-induced negative regulators in mDCs aimed to prevent sustained DC activation. In contrast, the pro-tolerogenic DC populations in most cases retained expression levels of pro-tolerogenic markers after stimulation, or even exhibited enhanced levels. Intriguingly, in response to stimulation, IL6-DCs and DEX-DCs commonly upregulated the same set of negative regulators that were found elevated in mDCs. As stated above, no common signalling pathway for IL-6 and GC has yet been identified. Therefore, in IL6-DCs and DEX-DCs on one hand stimulation-associated down-regulation of pro-tolerogenic markers is actively suppressed. On the other hand, in these DC populations stimulation-induced upregulation of those molecules that serve to limit sustained (control) DC activation is not prevented. Our observations on the immuno-phenotype and functional properties of the different pro-tolerogenic DC populations suggest that convergingly upregulated pro-tolerogenic markers may contribute to imprint a pronounced, maturation-resistant protolerogenic state of activation in DCs. In this regard, IL-10 as upregulated on transcriptional level by most tolerized DC populations, and commonly retained on protein level by all tolerized DC populations after stimulation, may essentially contribute to DC tolerization, and constitute a common key denominator of tolerized DC. The high number of potential interaction partners of IL-10 on protein level underscores its potential influence on the expression of other pro-tolerogenic markers. Besides, markers as identified in this study to be upregulated in a immune modulator-specific manner may bear the potential to be exploited as biomarkers of cancer patients´ blood DCs to deduce the identity of tumor-derived immuno-modulators. Authors’ contributions Stefanie Trojandt and Iris Bellinghausen performed and analysed the experiments. Angelika B. Reske-Kunz and Matthias Bros designed and supervised the study and wrote the manuscript. Acknowledgements This study was supported by an intramural grant of the University Medical Center of the Johannes Gutenberg University Mainz to MB (no. 97282830). The authors declare no conflict of interest. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.humimm.2016. 08.014. References [1] K.L. Flannigan, D. Geem, A. Harusato, T.L. Denning, Intestinal antigenpresenting cells: key regulators of immune homeostasis and inflammation, Am. J. Pathol. 185 (2015) 1809–1819. [2] M. Escamilla-Tilch, G. Filio-Rodríguez, R. García-Rocha, I. Mancilla-Herrera, N. A. Mitchison, J.A. Ruiz-Pacheco, F.J. Sánchez-García, D. Sandoval-Borrego, E.A. Vázquez-Sánchez, The interplay between pathogen-associated and dangerassociated molecular patterns: an inflammatory code in cancer?, Immunol Cell Biol. 91 (2013) 601–610. [3] A.K. Hopp, A. Rupp, V. Lukacs-Kornek, Self-antigen presentation by dendritic cells in autoimmunity, Front. Immunol. 5 (2014) 55. [4] A. Bazsó, Á. Szappanos, A. Patócs, G. Poór, Y. Shoenfeld, E. Kiss, The importance of glucocorticoid receptors in systemic lupus erythaematosus. A systematic review, Autoimmun. Rev. 14 (2015) 349–351. [5] M.A. van Delft, L.F. Huitema, S.W. Tas, The contribution of NF-jB signalling to immune regulation and tolerance, Eur. J. Clin. Invest. 45 (2015) 529–539. [6] J.L. Harden, N.K. Egilmez, Indoleamine 2,3-dioxygenase and dendritic cell tolerogenicity, Immunol. Invest. 41 (2012) 738–764.

Please cite this article in press as: S. Trojandt et al., Tumor-derived immuno-modulators induce overlapping pro-tolerogenic gene expression signatures in human dendritic cells, Hum. Immunol. (2016), http://dx.doi.org/10.1016/j.humimm.2016.08.014

S. Trojandt et al. / Human Immunology xxx (2016) xxx–xxx [7] H. Jonuleit, U. Kühn, G. Müller, K. Steinbrink, L. Paragnik, E. Schmitt, J. Knop, A. H. Enk, Pro-inflammatory cytokines and prostaglandins induce maturation of potent immunostimulatory dendritic cells under fetal calf serum-free conditions, Eur. J. Immunol. 27 (12) (1997) 3135–3142. [8] N. Obermajer, R. Muthuswamy, J. Lesnock, R.P. Edwards, P. Kalinski, Positive feedback between PGE2 and COX2 redirects the differentiation of human dendritic cells toward stable myeloid-derived suppressor cells, Blood 118 (2011) 5498–5505. [9] M. Menges, S. Rössner, C. Voigtländer, H. Schindler, N.A. Kukutsch, C. Bogdan, K. Erb, G. Schuler, M.B. Lutz, Repetitive injections of dendritic cells matured with tumor necrosis factor alpha induce antigen-specific protection of mice from autoimmunity, J. Exp. Med. 195 (2002) 15–21. [10] E. Gottfried, M. Kreutz, A. Mackensen, Tumor-induced modulation of dendritic cell function, Cytokine Growth Factor Rev. 19 (2008) 65–77. [11] D.M. Mosser, X. Zhang, Interleukin-10: new perspectives on an old cytokine, Immunol. Rev. 226 (2008) 205–218. [12] L. Yang, Y. Pang, H.L. Moses, TGF-beta and immune cells: an important regulatory axis in the tumor microenvironment and progression, Trends Immunol. 31 (2010) 220–227. [13] F.B. Patel, S. Khagi, K.P. Daly, R.M. Lechan, V. Ummaritchot, M.W. Saif, Pancreatic neuroendocrine tumor with ectopic adrenocorticotropin production: a case report and review of literature, Anticancer Res. 33 (2013) 4001–4005. [14] M. Bros, F. Jährling, A. Renzing, N. Wiechmann, N.A. Dang, A. Sutter, R. Ross, J. Knop, S. Sudowe, A.B. Reske-Kunz, A newly established murine immature dendritic cell line can be differentiated into a mature state, but exerts tolerogenic function upon maturation in the presence of glucocorticoid, Blood 109 (2007) 3820–3829. [15] D. Szklarczyk, A. Franceschini, S. Wyder, K. Forslund, D. Heller, J. Huerta-Cepas, M. Simonovic, A. Roth, A. Santos, K.P. Tsafou, M. Kuhn, P. Bork, L.J. Jensen, C. von Mering, STRING v10: protein-protein interaction networks, integrated over the tree of life, Nucleic Acids Res. 43 (2015) D447–D452. [16] Q. Zhao, X. Wang, L.D. Nelin, Y. Yao, R. Matta, M.E. Manson, R.S. Baliga, X. Meng, C.V. Smith, J.A. Bauer, C.H. Chang, Y. Liu, MAP kinase phosphatase 1 controls innate immune responses and suppresses endotoxic shock, J. Exp. Med. 203 (2006) 131–140. [17] E. Beaulieu, E.F. Morand, Role of GILZ in immune regulation, glucocorticoid actions and rheumatoid arthritis, Nat. Rev. Rheumatol. 7 (2011) 340–348. [18] S.J. Wölfle, J. Strebovsky, H. Bartz, A. Sähr, C. Arnold, C. Kaiser, A.H. Dalpke, K. Heeg, PD-L1 expression on tolerogenic APCs is controlled by STAT-3, Eur. J. Immunol. 41 (2011) 413–424. [19] E.N. Rozali, S.V. Hato, B.W. Robinson, R.A. Lake, W.J. Lesterhuis, Programmed death ligand 2 in cancer-induced immune suppression, Clin. Dev. Immunol. 2012 (2012) 656340. [20] M. Loos, D.M. Hedderich, H. Friess, J. Kleeff, B7-h3 and its role in antitumor immunity, Clin. Dev. Immunol. 2010 (2010) 683875. [21] J.S. Manavalan, P.C. Rossi, G. Vlad, F. Piazza, A. Yarilina, R. Cortesini, D. Mancini, N. Suciu-Foca, High expression of ILT3 and ILT4 is a general feature of tolerogenic dendritic cells, Transpl. Immunol. 11 (2003) 245–258. [22] G. Huang, Y. Wang, H. Chi, Control of T cell fates and immune tolerance by p38a signaling in mucosal CD103+ dendritic cells, J. Immunol. 191 (2013) 650– 659. [23] C.A. Dinarello, J.W. van der Meer, Treating inflammation by blocking interleukin-1 in humans, Semin. Immunol. 25 (2013) 469–484. [24] A. Schumacher, A.C. Zenclussen, Effects of heme oxygenase-1 on innate and adaptive immune responses promoting pregnancy success and allograft tolerance, Front. Pharmacol. 5 (2015) 288. [25] N. Cools, V.F. Van Tendeloo, E.L. Smits, M. Lenjou, G. Nijs, D.R. Van Bockstaele, Z.N. Berneman, P. Ponsaerts, Immunosuppression induced by immature dendritic cells is mediated by TGF-beta/IL-10 double-positive CD4+ regulatory T cells, J. Cell Mol. Med. 12 (2008) 690–700. [26] U. Bharadwaj, M. Li, R. Zhang, C. Chen, Q. Yao, Elevated interleukin-6 and G-CSF in human pancreatic cancer cell conditioned medium suppress dendritic cell differentiation and activation, Cancer Res. 67 (2007) 5479–5488. [27] S. Gregori, D. Tomasoni, V. Pacciani, M. Scirpoli, M. Battaglia, C.F. Magnani, E. Hauben, M.G. Roncarolo, Differentiation of type 1 T regulatory cells (Tr1) by

[28]

[29]

[30]

[31]

[32]

[33]

[34] [35]

[36]

[37]

[38] [39]

[40]

[41]

[42]

[43]

[44]

[45]

[46]

9

tolerogenic DC-10 requires the IL-10-dependent ILT4/HLA-G pathway, Blood 116 (2010) 935–944. A. Gaba, S.I. Grivennikov, M.V. Do, D.J. Stumpo, P.J. Blackshear, M. Karin, Cutting edge: IL-10-mediated tristetraprolin induction is part of a feedback loop that controls macrophage STAT3 activation and cytokine production, J. Immunol. 189 (2012) 2089–2093. K.J. Staples, T. Smallie, L.M. Williams, A. Foey, B. Burke, B.M. Foxwell, L. ZieglerHeitbrock, IL-10 induces IL-10 in primary human monocyte-derived macrophages via the transcription factor Stat3, J. Immunol. 178 (2007) 4779–4785. C. Unterberger, K.J. Staples, T. Smallie, L. Williams, B. Foxwell, A. Schaefer, B. Kempkes, T.P. Hofer, M. Koeppel, M. Lohrum, H. Stunnenberg, M. Frankenberger, L. Ziegler-Heitbrock, Role of STAT3 in glucocorticoid-induced expression of the human IL-10 gene, Mol. Immunol. 45 (2008) 3230–3237. Y. Alvarez, C. Municio, S. Alonso, M. Sánchez Crespo, N. Fernández, The induction of IL-10 by zymosan in dendritic cells depends on CREB activation by the coactivators CREB-binding protein and TORC2 and autocrine PGE2, J. Immunol. 183 (2009) 1471–1479. S. Brenner, S. Prösch, K. Schenke-Layland, U. Riese, U. Gausmann, C. Platzer, CAMP-induced Interleukin-10 promoter activation depends on CCAAT/ enhancer-binding protein expression and monocytic differentiation, J. Biol. Chem. 278 (2003) 5597–5604. S. Liang, V. Ristich, H. Arase, J. Dausset, E.D. Carosella, A. Horuzsko, Modulation of dendritic cell differentiation by HLA-G and ILT4 requires the IL-6–STAT3 signaling pathway, Proc. Natl. Acad. Sci. U.S.A. 105 (2008) 8357–8362. R. Thomas, T. Matthias, T. Witte, Leukocyte immunoglobulin-like receptors as new players in autoimmunity, Clin. Rev. Allergy Immunol. 38 (2010) 159–162. H. Nakajima, A. Asai, A. Okada, L. Ping, F. Hamajima, T. Sata, K. Isobe, Transcriptional regulation of ILT family receptors, J. Immunol. 171 (2003) 6611–6620. D. Gur-Wahnon, Z. Borovsky, M. Liebergall, J. Rachmilewitz, The induction of APC with a distinct tolerogenic phenotype via contact-dependent STAT3 activation, PLoS One 4 (2009) e6846. H. Lee, N. Rezai-Zadeh, E. Seto, Negative regulation of histone deacetylase 8 activity by cyclic AMP-dependent protein kinase A, Mol. Cell. Biol. 2004 (24) (2004) 765–773. J.L. Eddy, K. Krukowski, L. Janusek, H.L. Mathews, Glucocorticoids regulate natural killer cell function epigenetically, Cell. Immunol. 290 (2014) 120–130. N. Cohen, E. Mouly, H. Hamdi, M.C. Maillot, M. Pallardy, V. Godot, F. Capel, A. Balian, S. Naveau, P. Galanaud, F.M. Lemoine, D. Emilie, GILZ expression in human dendritic cells redirects their maturation and prevents antigen-specific T lymphocyte response, Blood 107 (2006) 2037–2044. J.C. Wang, M.K. Derynck, D.F. Nonaka, D.B. Khodabakhsh, C. Haqq, K.R. Yamamoto, Chromatin immunoprecipitation (ChIP) scanning identifies primary glucocorticoid receptor target genes, Proc. Natl. Acad. Sci. U.S.A. 101 (2004) 15603–15608. S.H. Tynan, S.G. Lundeen, G.F. Allan, Cell type-specific bidirectional regulation of the glucocorticoid-induced leucine zipper (GILZ) gene by estrogen, J. Steroid Biochem. Mol. Biol. 91 (2004) 225–239. D. Berrebi, S. Bruscoli, N. Cohen, A. Foussat, G. Migliorati, L. Bouchet-Delbos, M. C. Maillot, A. Portier, J. Couderc, P. Galanaud, M. Peuchmaur, C. Riccardi, D. Emilie, Synthesis of glucocorticoid-induced leucine zipper (GILZ) by macrophages: an anti-inflammatory and immunosuppressive mechanism shared by glucocorticoids and IL-10, Blood 101 (2003) 729–738. M. Mihara, M. Hashizume, H. Yoshida, M. Suzuki, M. Shiina, IL-6/IL-6 receptor system and its role in physiological and pathological conditions, Clin. Sci. (Lond.) 122 (2012) 143–159. M. Mokrani, J. Klibi, D. Bluteau, G. Bismuth, F. Mami-Chouaib, Smad and NFAT pathways cooperate to induce CD103 expression in human CD8 T lymphocytes, J. Immunol. 192 (2014) 2471–2479. M. Lanzinger, B. Jürgens, U. Hainz, B. Dillinger, J. Raberger, D. Fuchs, A. Heitger, Ambivalent effects of dendritic cells displaying prostaglandin E2-induced indoleamine 2,3-dioxygenase, Eur. J. Immunol. 42 (2012) 1117–1128. L. Cousens, N. Najafian, W.D. Martin, A.S. De Groot, Tregitope: immunomodulation powerhouse, Hum. Immunol. 75 (2014) 1139–1146.

Please cite this article in press as: S. Trojandt et al., Tumor-derived immuno-modulators induce overlapping pro-tolerogenic gene expression signatures in human dendritic cells, Hum. Immunol. (2016), http://dx.doi.org/10.1016/j.humimm.2016.08.014