Impaired activation of plasmacytoid dendritic cells via toll-like receptor 7/9 and STING is mediated by melanoma-derived immunosuppressive cytokines and metabolic drift

Introduction Plasmacytoid dendritic cells (pDCs) infiltrate a large set of human cancers. Interferon alpha (IFN-α) produced by pDCs induces growth arrest and apoptosis in tumor cells and modulates innate and adaptive immune cells involved in anti-cancer immunity. Moreover, effector molecules exert tumor cell killing. However, the activation state and clinical relevance of pDCs infiltration in cancer is still largely controversial. In Primary Cutaneous Melanoma (PCM), pDCs density decreases over disease progression and collapses in metastatic melanoma (MM). Moreover, the residual circulating pDC compartment is defective in IFN-α production. Methods The activation of tumor-associated pDCs was evaluated by in silico and microscopic analysis. The expression of human myxovirus resistant protein 1 (MxA), as surrogate of IFN-α production, and proximity ligation assay (PLA) to test dsDNA-cGAS activation were performed on human melanoma biopsies. Moreover, IFN-α and CXCL10 production by in vitro stimulated (i.e. with R848, CpG-A, ADU-S100) pDCs exposed to melanoma cell lines supernatants (SN-mel) was tested by intracellular flow cytometry and ELISA. We also performed a bulk RNA-sequencing on SN-mel-exposed pDCs, resting or stimulated with R848. Glycolytic rate assay was performed on SN-mel-exposed pDCs using the Seahorse XFe24 Extracellular Flux Analyzer. Results Based on a set of microscopic, functional and in silico analyses, we demonstrated that the melanoma milieu directly impairs IFN-α and CXCL10 production by pDCs via TLR-7/9 and cGAS-STING signaling pathways. Melanoma-derived immunosuppressive cytokines and a metabolic drift represent relevant mechanisms enforcing pDC-mediated melanoma escape. Discussion These findings propose a new window of intervention for novel immunotherapy approaches to amplify the antitumor innate immune response in cutaneous melanoma (CM).

Introduction: Plasmacytoid dendritic cells (pDCs) infiltrate a large set of human cancers.Interferon alpha (IFN-a) produced by pDCs induces growth arrest and apoptosis in tumor cells and modulates innate and adaptive immune cells involved in anti-cancer immunity.Moreover, effector molecules exert tumor cell killing.However, the activation state and clinical relevance of pDCs infiltration in cancer is still largely controversial.In Primary Cutaneous Melanoma (PCM), pDCs density decreases over disease progression and collapses in metastatic melanoma (MM).Moreover, the residual circulating pDC compartment is defective in IFN-a production.

Methods:
The activation of tumor-associated pDCs was evaluated by in silico and microscopic analysis.The expression of human myxovirus resistant protein 1 (MxA), as surrogate of IFN-a production, and proximity ligation assay (PLA) to test dsDNA-cGAS activation were performed on human melanoma biopsies.Moreover, IFN-a and CXCL10 production by in vitro stimulated (i.e. with R848, CpG-A, ADU-S100) pDCs exposed to melanoma cell lines supernatants (SN-mel) was tested by intracellular flow cytometry and ELISA.We also performed a bulk RNA-sequencing on SN-mel-exposed pDCs, resting or stimulated with R848.Glycolytic rate assay was performed on SN-mel-exposed pDCs using the Seahorse XFe24 Extracellular Flux Analyzer.

Introduction
The clinical success of immune checkpoint blockades (ICBs) in patients with melanoma has driven the therapeutic revolution in cancer approach and new combinations of ICBs have significantly improved the outcome of patients with metastatic melanoma (MM) (1,2).More recently, advances with ICBs have led to the approval of anti-PD1 agents and Pembrolizumab in the adjuvant setting, resulting in a higher recurrence-free survival in comparison to placebo (3)(4)(5)(6)(7).Unfortunately, half of the surgically resected patients relapses and do not show long-lasting benefit with ICBs (2), suggesting the need for a better characterization of predictive biomarkers for patient selection and for novel strategies to overcome resistance.
Features of the immune contexture of cancer predict prognosis and response to ICBs.Among immune cells involved in anti-tumor immune response, Plasmacytoid Dendritic cells (pDCs) play a crucial role bridging the innate and adaptive immunity (8).pDCs are the major type I interferon (I-IFN) producing cells upon nucleic acids sensing through Toll-like receptor (TLR) 7-and 9-dependent signaling pathways (9,10).In addition, pDCs express other Pattern Recognition Receptors (PRR), such as C-type lectin receptors (CLRs), RIG-I-like receptors (RLRs), NOD-like receptors (NLRs), and cyclic guanosine monophosphate-adenosine monophosphate synthase (cGAS) (11).Innate immune response is also promoted by free endogenous DNA (12) and pDCs are able to sense cytosolic DNA through cGAS-STING (stimulator of interferon genes) pathway eliciting a potent I-IFN production independently of TLR7/9 (13)(14)(15).interferon-a (IFN-a) production directly affects tumor growth by inducing cell-cycle arrest and apoptosis of cancer cells and inhibiting angiogenesis (16,17).Moreover, IFN-a modulates innate and adaptive immune cells involved in anti-cancer immunity (18)(19)(20), by inducing the paracrine production of pro-inflammatory chemokines (i.e.CXCL9, CXCL10, CXCL11) (21) and driving TH1 polarization of immune cells (22).In the last years, pDCs have been identified among the restricted cell subsets expressing the IFN-l functional receptor (IFNlR) other than producers of type III interferons (IFN-l or IL-28/IL-29), which play potent anti-viral activities and trigger IFN-a production during tumor progression (23).Nucleic acid-sensing mechanisms are druggable and, in combination with radiation therapy or chemotherapy, they boost the anti-tumor immune response by triggering DNA damage-induced immunogenic cell death and innate immune activation (24-27).
Preclinical data suggest that the activation of TA-pDCs by TLR-7/9 agonists administration amplifies the local and systemic antitumor immune response and promotes tumor cells killing (37).Accordingly, numerous clinical trials highlighted that TLR-7/9 agonists administration, individually or in combination with ICB therapies, has the potential to stimulate specific T cell response and subsequent tumor regression (38,39), even for patients with MM (37).More recently, the administration of STING agonists has resulted in tumor regression in mice and generated local and systemic anti-tumor immune responses, as well (40,41).Clinical trials (phase I) combining STING agonists with ICB are ongoing (Clinicaltrials.govstudy identifiers: NCT03843359, NCT04144140 and NCT03010176) for patients with advanced solid tumors or lymphomas.Finally, pDCs-based cancer vaccines have been shown to increase the frequency of circulating anti-tumor T lymphocytes, together with the induction of I-IFN signature in patients with MM (42)(43)(44).
In Cutaneous Melanoma (CM), pDCs infiltrate the primary tumor (45) and draining lymph-nodes (46,47).However, the pDC compartment decreases over disease progression and collapses in metastatic melanoma (MM) patients (34,37,46,48).Moreover, the residual blood pDC component in MM is defective in IFN-a production and displays immunosuppressive features (34, 49).These findings suggest that CM progression is, at least in part, supported by pDCs specific immune escape mechanisms (50).We have recently proposed lactic acidosis as the oncometabolite implicated in pDCs functional impairment in MM patients (49).Here, we investigated the molecular trajectory of pDCs after their exposure to CM cell lines supernatants (SN-mel).IFN-a production by TLR-7/9 and cGAS-STING activation were significantly reduced, suggesting a widespread impairment of PRR pathways in pDCs.By using RNA sequencing, we unveiled that melanoma secretome rewired the pDC transcriptomic profile towards an IFN-defective tolerogenic state, likely dependent of immunosuppressive cytokines and metabolic drift.These findings should better guide windows of intervention for novel immunotherapy approaches empowering the innate compartment of primary cutaneous melanoma (PCM).

Human tissue samples and immunohistochemistry
Tissues were represented by a cohort of 101 PCM, 60 benign nevi (NV) and 5 lupus erythematosus (LE) skin biopsies.Clinical and pathological features of the PCM cases are reported in our previous study (46).The local ethics committee provided formal approval to this project (WV-Immunocancer 2014 to WV, institutional review board code NP906).
Two-to-four micron-thick tissue sections were obtained from formalin-fixed, paraffin-embedded (FFPE) blocks and used for immunohistochemistry.For immunohistochemical staining endogenous peroxidase was blocked by incubation with methanol and hydrogen peroxide 0.03% for 20 minutes during rehydration.Immunostaining was performed using a set of primary antibodies listed in Supplementary Table S1.The reaction was revealed using Novolink Polymer (Leica Microsystems) followed by diaminobenzidine (DAB, Dako, Glostrup, Denmark) or using Mach 4 MR-AP (Biocare Medical, Concord, CA, USA), followed by StayRed/AP (Abcam).Finally, the slides were counterstained with Meyer's Haematoxylin.For double staining, after completing the first immune reaction, the second was visualized using Mach 4 MR-AP (Biocare Medical), followed by Ferangi Blue (Biocare Medical).Quantitative image analysis was performed by using AperioScanscope CS (Aperio, Nikon).

RNAscope
To localize TGF-b and IL-10 positive cells, tissues were analyzed with RNAscope assay (Advanced Cell Diagnostics, Newark) using RNAscope 2.5 HD Assay-RED kit.The Hs-IL10 probe (cat no.602051) recognizes the nt 122-1,163 of the IL-10 mRNA (reference sequence NM_000572.2),and the Hs-TGFB1 probe (cat.no.400881) recognizing the nt 170-1,649 of the TGF-b mRNA (reference sequence NM_000660.4).The sections from fixed human tissue blocks were treated following the manufacturer's instructions.Briefly, freshly cut 3-mm sections were deparaffinized in xylene (Bio-Optica, cat.no.06-1304F) and treated with the peroxidase block solution (ACD, cat.no.322335) for 10 minutes at room temperature followed by the retrieval solution for 15 minutes at 98°C and by protease plus (ACD, cat.no.322331) at 40°C for 30 minutes.Hs-POLR2a-C2 (cat.no.310451) and dapB-C2 (cat.no.310043-C2) were used as control probes.The hybridization was performed for 2 hours at 40°C.The signal was revealed using RNAscope 2.5 HD Detection Reagent and FAST RED.
In situ PLA Proximity ligation assay (PLA) was performed using NaveniBright HRP kit (Navinci Diagnostics) according to the manufacturer's protocol using the following primary antibodies: anti-cGAS (clone D1D3G, #15102, 1:100 pH9, Cell Signaling) and anti-DNA double stranded (clone AE-2, MAB1293, 1:100 pH9, Abcam).Negative controls were performed using only one primary antibody.Slides were analyzed under a Zeiss Axioscope A1 and microphotographs were collected using a Zeiss Axiocam 503 Color with the Zen 2.0 Software (Zeiss).Sections were subsequently immunostained to detect the expression of the E2.2 antigen.IHC was developed using SignalStainBoost IHC Detection rabbit (cod.#18653, Cell Signaling Technology) alkaline phosphataseconjugated produced in horse and Vulcan Fast Red as substrate chromogen.Again, the sections were analyzed and photographed using Zeiss Axioscope A1 and Zeiss Axiocam 503 Color with the Zen 2.0 Software (Zeiss).Segmented images were obtained using HALO image analysis software (v3.2.1851.229,Indica Labs).HALO image analysis software was used to quantify the PLA signals in twelve non-overlapping fields at high-power magnification (400X) and the output was expressed as "percentage positive cells".

Flow cytometry analysis
Fluorescence minus one (FMO) was used to set the marker for positive cells.A baseline fluorescence control was used as a reference to set the fluorescence thresholds for positivity.The results were expressed as the percentage of positive cells or as the mean of fluorescence intensity (MFI) of positive cells.To evaluate the IFN-a and CXCL10 intracellular production, pDCs were firstly surface labelled with anti-BDCA-2 (Miltenyi Biotec; cat.No. 130-113-192) and anti-CD123 (Miltenyi Biotec; cat.No. 130-115-270) fluorochrome-conjugated antibodies.Then, pDCs were fixed and permeabilized using the Inside Stain Kit (Miltenyi Biotec; cat.No. 130-090-477) and intracellular staining was performed using the anti-IFN-a (Miltenyi Biotec; cat.No. 130-116-874) and anti-CXCL10 (Biolegend; cat.No. 519504) fluorochrome-conjugated antibodies.Samples were processed on MACS Quant Analyzer 16 (Miltenyi Biotec).Results were analyzed by FlowJo X software (TreeStar Inc., Wilmington, NC, USA).
At 24 hours post-stimulation (as described in the Isolation, culture, and stimulation of human peripheral blood pDCs section), pDC supernatants were collected, as described above.The amounts of secreted IFN-a and CXCL10 were determined by using the Human IFN-a Matched Antibody Pair (eBioscience; cat.No. BMS216MST) and Human CXCL10/IP-10 DuoSet ® ELISA (R&D Systems; cat.No. DY266).The supernatants were stored at -20°C until use.

Western blot
pDCs were cultured as described in the Isolation, culture, and stimulation of human peripheral blood pDCs section.pDCs were stimulated with ADU-S100 (50 mg/mL; Invivogen; cat.No. TLRL-NACDA2R) for 4 hours.After stimulation, cells were lysed in RIPA lysis buffer (Thermo Fisher Scientific, cat.No. 89900) supplemented with Protease Inhibitor Cocktail (Sigma-Aldrich, cat.No. 78440) and incubated on ice for 20 minutes.Cell lysates were centrifuged at 13000xg for 15 minutes and the suspension was collected and stored at -20°C.Protein concentration was determined by Bradford assay and 20 µg of total proteins were loaded on 4

Extracellular flux analysis
Glycolytic activity of the pDCs was determined using the Seahorse XFe24 Extracellular Flux Analyzer (Agilent, Santa Clara, CA, USA).Isolated pDCs were cultured in RPMI or SN-mel for 24 hours as described above (Isolation, culture, and stimulation of human peripheral blood pDCs section) (Vescovi 2019), washed, and resuspended in Seahorse XF RPMI Medium pH 7.4 (Agilent) by adding 10 mM glucose, 2 mM L-glutamine, 1 mM sodium pyruvate and 20 ng/mL IL-3.Cells were seeded at pre-determined cell density of 200,000 cells/well in 100 µL/well on Seahorse XFe24 V7 PS Cell Culture Microplate (Agilent) pre-coated with poly-D-lysin (Sigma Aldrich, St. Louis, Missouri).Particularly, 50 µL/well of poly-Dlysin 0.1 mg/ml solution were incubated 30 minutes at 37°C, then removed and washed twice.Microplates were centrifuged once at 200× g for 1 min with the break off to aid uniform cell attachment.The plates were transferred to a 37°C incubator not supplemented with CO 2 for 30 minutes to facilitate cell adhesion.The medium volume was brought up to 500 µL in each well and the plates were return to the incubator for 15 minutes before running the assay.We applied the Seahorse XF Glycolytic Rate Assay (Agilent) to measure real-time extracellular acidification rate (ECAR) and oxygen consumption rate (OCR) of cells and determine the glycolytic proton efflux rate (glycoPER) that is the rate of protons extruded into the extracellular medium during glycolysis (discounting the effect of CO 2 -dependent acidification).The glycoPER is obtained using the following formula: [ECAR (mpH/min) x Buffer Factor (mmol/L/pH) x Geometric Volume (µL) x Volume scaling factor (Kvol)] -CO 2 Contribution Factor.Following the manufacturer's instructions, the protocol was set to measure first basal glycolysis in the absence of inhibitors/stimuli.The acute injection effects of the R848 (5 µg/mL) on glycolysis (induced glycolysis) was evaluated after 4 cycles of measurements.Subsequently, the injection of a mixture of the respiratory complex I inhibitor rotenone (Rot; 0.5 µM) and the complex III inhibitor antimycin A (AA; 0.5 µM) was performed to inhibit mitochondrial activity and therefore CO 2derived proton production to retrieve compensatory glycolysis.Finally, the glucose analogue 2-deoxy-D-glucose (2-DG; 50 mM) was injected to inhibit glycolysis by blocking glucose hexokinase and exclude other sources of extracellular acidification.
RNA isolation 2 x 10 5 pDCs (n = 4) were 24h-cultured with RPMI complete medium or SN-mel (as described in the Isolation, culture, and stimulation of human peripheral blood pDCs section), unstimulated or stimulated with R848 (5 mg/mL; Invivogen; cat.No. tlrl-r848-5) for 2 hours.Total RNA was isolated by using AllPrep DNA/RNA/ miRNA kit (Qiagen, cat.No. 80224) following manufacturer's instructions.The optional Dnase I step was included to prevent genomic DNA contamination.
Quantitation and integrity of RNA samples were determined by using the Qubit RNA HS Assay Kit (Life Technologies, cat.No. Q33230) and the RNA 6000 Nano Kit on a Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA; cat.No. 5067-1511), respectively.RNA samples with RIN > 6 were further processed for RNA-sequencing.

RNA-sequencing analysis
The TruSeq Stranded mRNA Library Prep kit (Illumina, San Diego, CA, USA; cat.No. 20020595) was used to prepare RNAseq libraries from 100ng RNA, after poly(A) capture and according to manufacturer's instructions.Quality and size of RNA-Seq libraries were assessed by capillary electrophoretic analysis with the Agilent 4200 Tape station (Agilent Technologies) and were quantified by real-time PCR against a standard curve with KAPA Library Quantification Kit (KapaBiosystems).Libraries were pooled at equimolar concentration and sequenced using 75SR reads using Illumina NextSeq 500 platform, generating ~33 million fragments on average per sample.

Data processing of the TCGA datasets and statistical analyses
Normalized RSEM counts, by the upper quartile method, for primary solid tumor samples of six TCGA projects (TCGA-BLCA, TCGA-COAD, TCGA-HNSC, TCGA-LUAD, TCGA-LUSC, TCGA-SKCM) (55) were downloaded from Xena portal and merged with metadata of TCGA samples, downloaded using TCGA biolinks R/Bioconductor package (n=2637 cases).For downstream analysis the log2(norm.count+1) gene expression was considered.The estimation of the enrichment score of 6 gene sets of interest (Supplementary Data 1) was performed by Gene Set Variation Analysis (GSVA), using the GSVA package (53, 56) and setting the following parameters: min.sz=5,max.sz=2000, kcdf="Gaussian", mx.diff=F, parallel.sz=1.In particular, PDC signature was taken from our previous work (46) and I-IFNs/III-IFNs signature was created based on the known genes encoding for interferon-a, -b and -l.The signatures of genes up-regulated in response to IFN-a protein (designated as Interferon Alpha Response) and to TGFb1 (designated as TGF-b Signaling) correspond to the Hallmark gene sets from the Molecular Signatures Database (MSigDB) collections.The signatures of Interleukin 10 Signaling and STING Mediated Induction Of Host Immune Responses were derived from the Reactome Canonical Pathway database (MSigDB).Moreover, the expression levels of single gene transcripts (IFNA1, IFNL1, IFNG, IL1A, IL1B, IL6, IL10, TNF, TGFB1, TGFB2, PTGES2, CSF1, MxA) were taken into account.
Qualitative variables were described as absolute and relative frequencies; standard descriptive statistics were used for continuous variables, expressing means, standard deviations, medians, and ranges.Comparisons were tested by Mann-Whitney U test or Kruskal-Wallis test, as appropriate.Correlations between quantitative variables were computed using Spearman correlation coefficient.Bonferroni's correction of significance level was applied for multiple testing in each analysis.Considering the enrichment scores a dimensionality reduction was performed by UMAP, setting the following parameters: n_components = 2, n_neighbors = 15, init = 'PCA', learning_rate = '0.5';min_dist = 0.01, metric = 'euclidean', random_state = 2 (57).All statistical tests were two-sided, and pvalue < 0.05 was considered statistically significant.Unpaired student's T test and one-way ANOVA were used to compare two and three or more groups, respectively.GraphPad Prism Software version 5 (GraphPad Software, San Diego, CA, USA), R (version 4.0.2) and R studio) were used for statistical analysis and rendering graphs.

Results
pDC infiltration and activation is impaired in advanced PCM pDCs infiltrate a wide set of human cancers (37), including melanoma.We extended this finding by analyzing a pan-cancer TCGA datasets (Supplementary Table S2) and measuring the pDC content using a previously identified signature (46).pDC signature resulted poorly expressed in skin cutaneous melanoma (TCGA-SKCM) as compared to primary carcinomas (TCGA-BLCA, TCGA-COAD, TCGA-HNSC, TCGA-LUAD, TCGA-LUSC; Figure 1A; Supplementary Table S3).Moreover, a high pDC content was limited to draining lymph nodes (46) and stage I primary tumors (Figure 1B; Supplementary Table S4).
We devised signatures interrogating interferons expression (referred as I-IFNs/III-IFNs signature) and response to interferons (referred as IFN-a Response signature) and STING mediated immune response activation (referred as STING signature) as well as expression of other major cytokines (Figures 1C-E; Supplementary Data 1).We found that IFNA1 and IFNL1 gene transcripts resulted negligibly expressed across the pan-cancer dataset (Figure 1E) when compared to other immunosuppressive and pro-inflammatory cytokines such as TGFB1/2, PTGES2, IL1A, IL1B, IL6 and CSF1.However, the PDC signature and IFN-a Response signature directly correlated, especially in the TCGA-SKCM dataset (Figures 1F, G).Moreover, IFN-a Response signature directly correlated with I-IFNs/ III-IFNs, MxA (human myxovirus resistant protein 1) and STING signatures (Figures 1F, G); however, no correlation was detected between pDC and I-IFNs/III-IFNs signatures (Figures 1F, G).These findings suggest that pDCs might be partly dysfunctional in type I and III IFNs production and that activation of IFN-a/STING pathway derive from different cellular sources in the CM TME.To further explore the functional impairment of PCM-associated pDCs, we assessed the expression of MxA, as surrogate of I-IFN production, on melanoma biopsies (n=99) (Figures 2A, B).Positive controls (58), represented by lupus erythematosus (LE) skin biopsies (n=5), showed a strong and diffuse MxA reactivity in all compartments, including epidermis and dermal structures (Figures 2A, B).Although heterogeneous within the PCM cohort, the MxA reactivity was significantly reduced in PCM compared to LE (1.42% vs 32.96%; p < 0.0001), but comparable between PCM and benign nevi (NV) (Figure 2B), the latter largely devoid of pDCs (46).MxA was significantly reduced along the PCM progression from T1 to T2-T4 stages (p = 0.04; Figure 2C).Moreover, linear regression analysis proved that the lack of MxA expression is not related to a poor pDC density (r 2 = 0.02; Figures 2D, E).All together, these findings propose that defective I-IFN endogenous production by pDCs occur over PCM progression.

cGAS-STING activation is impaired in PCMassociated pDCs
In PCM tissues, pDCs can be found near melanoma cells (Figure 3A).Besides TLRs, human pDCs express endosomal and cytosolic nucleic acid sensors and might sense nucleic acids released from dying tumor cells (59)(60)(61).Tumor-derived DNA leads to the activation of the cGAS-STING pathway (59).Specifically, cGAS interacts with double-stranded DNA (dsDNA) and induces conformational changes into 2',3'-cyclic GMP-AMP (cGAMP).The second messenger cGAMP then activates STING (62).The activation of dendritic cells via STING pathway is crucial to eradicating tumors in mouse models (40), resulting in a strong priming of antigen-specific T cells with clinical responses (43).
We thus explored the endogenous activation of cGAS-STING pathway (13) in human PCM samples.STING expression resulted diffuse and strong in melanoma cells, overlying keratinocytes, blood vessels and cells of the TME including pDCs (Figures 3B, C); on the contrary, no STING reactivity was observed in keratinocytes distant from melanoma nests (Figure 3B).cGAS expression resulted more limited, with positive cells represented by a variable fraction of E2.2 + pDCs (Figures 3D, E), a fraction of melanoma cells (Figure 3D) and, as expected (63), infiltrating plasma cells (Supplementary Figure S1).
To test endogenous dsDNA-cGAS activation by pDCs infiltrating PCM, we analyzed pDC-enriched PCM cases (n=5) by using proximity ligation assay (PLA) with antibodies anti-cGAS and anti-DNA double stranded.Image analysis in these cases resulted in a very limited signal, and staining with E2.2 confirmed a negligible endogenous cGAS activation in pDCs (Figure 3F upper panels and Supplementary Figure S2A).However, by exploring PCM cases (n=3) with consistent areas of spontaneous microscopic melanoma regression (64, 65) and a large number of infiltrated E2.2 + pDCs, we could detect an increased signal suggesting endogenous pDCs activation (Figure 3F lower panels and Supplementary Figure S2B).By quantification of PLA signals, we observed a significant increase of dsDNA-cGAS activation in regressive PCM (regressive PCM mean=10.800355vs PCM mean=7.58688PLA positive cells/field; Figure 3G).These findings suggest a poor cGAS-STING endogenous activation in PCM microenvironment.

Melanoma secretome impairs the production of IFN-a and CXCL10 by pDCs
In CM tissues, pDCs are exposed to secreted products of melanoma cells and cells of the microenvironment.By intracellular flow cytometry, we tested the proficiency of human circulating pDCs from healthy donors to produce IFN-a and CXCL10 after in vitro exposure to supernatant obtained from various MCL (SN-mel).The frequency of IFN-a + pDCs following R848 or CpG stimulation was reduced in every SN-mel and significantly reduced in three out of four (SN-mel146; SN-mel336; SN-mel327) compared to RPMI (Figures 4A, B).In parallel, we tested the in vitro activation of circulating pDCs by STING agonist.The frequency of IFN-a + pDCs after stimulation with ADU-S100 was significantly diminished by SN-mel (Figure 4C).It should be noted that IFNa protein resulted nearly absent (<1.5% IFN-a + pDCs; <2ng/ml IFN-a) at baseline condition; however, their exposure to SN-mel further impairs IFN-a production (Supplementary Figures S3A-C).Similarly, the production of CXCL10, a pro-inflammatory cytokines relevant for the recruitment of antigen-specific T-cells into the tumor tissues (66) and mainly dependent on IFNa-IFNAR autocrine/paracrine signaling in pDCs (67), resulted consistently reduced by SN-mel as shown by intracellular flow cytometry (Figures 4D, E).Also, MFI of both IFN-a + and CXCL10 + pDCs, confirmed a trend towards decreased amounts of expression by SN-mel conditioned cells (Supplementary Figures S3D-G).Overall, these data highlight a defective intracellular IFN-a and CXCL10 production in pDCs exposed to melanoma soluble factors.We subsequently confirmed this defect by ELISA.Specifically, upon pDC activation, the IFN-a concentration measured by ELISA was significantly diminished after exposure to supernatants of melanoma cell lines (Figures 4F-H).Similarly, the amount of CXCL10 secreted by SNmel-exposed pDCs was significantly diminished after CpG stimulation; at variance, only a slight decrease of CXCL10 concentration was measured in R848 stimulated pDCs exposed to SN-mel (Figures 4I, J).Overall, these results support a role of CMderived soluble factors in dampening the pDC proficiency to produce both IFN-a and CXCL10 upon activation via TLR7/9 and STING signaling.

Melanoma secretome imprints immunosuppressive pathways and metabolic drift in pDCs
To further dissect the molecular mechanisms driving pDC functional impairment in CM, we performed bulk RNA sequencing (RNA-seq) analysis of healthy pDCs, cultured in RPMI or in melanoma conditioned medium (SN-mel146; SN-mel336) and either left untreated or stimulated with R848 for 2 hours (Figure 5A, Supplementary Figure S4).By principal component analysis (PCA), the SN-mel exposed pDCs segregated along the first PCA axis (PC1) into TLR7 activated and non-activated sample groups (Figure 5B, Supplementary Figure S5).By combining PC2 and PC3 we could identify the subgroups of SN-mel exposed pDCs (burgundy dots) versus RPMI cultured pDCs (green dots), being the latter characterized by high PC2 and low PC3 values (Figure 5B, Supplementary Figure S5).Specifically, SN-mel146 subgroup (red dots) is characterized by higher PC3 and lower PC2 values in comparison to SN-mel 336 (orange dots) (Supplementary Figure S5C).By exploring the loadings of each PCs, type I and type III IFNs transcripts resulted strongly and directly related to PC1 and PC2 and inversely related to PC3 (Supplementary Figure S6).Moreover, genes encoding for cytokines including IL12A, IL6, and IL27 and effector/ memory T cell-attracting chemokines (i.e.CCL3, CCL4, and CCL5) are among the top-fifty PC1 loadings.A set of molecules that are known to modulate the interactions between tumor cells and microenvironment towards a pro-tumoral and immunosuppressive phenotype, such as human lectin CLEC17A (also known as Prolectin) (68, 69), the thrombin receptor FR2 (also known as proteaseactivated receptor-1, PAR-1) (70-73), the spermine oxidase (SMOX) (74), the transcription factor TOX (75-77) and type II transmembrane glycoprotein ENPP1, were among component loadings inversely related to PC2 as well as directly related to PC3, representing SN-mel conditioning (Supplementary Table S5).
As expected, GSEA-based functional enrichment showed that up-regulated genes in R848-stimulated pDCs were enriched for anti-viral response and cytokine response.Particularly, type I IFNrelated or -induced pathways were enriched.Thoroughly, TLR7 stimulation induced up-regulation of different subtypes of IFNs and IFN inducible genes, as well as cytokines encoding genes (Figure 5C; Supplementary Table S6).Additionally, R848 induced expression of co-stimulatory molecules and chemokine receptor CCR7, which is involved in pDC recruitment to the tumor site (78) (Figure 5C; Supplementary Table S6).Finally, we identified IL18RA, LILRA5, and LILRB2, as surface markers associate to I-IFN production (79) (Figure 5C; Supplementary Table S6).
We found that a relevant number of transcripts were significantly modulated by SN-mel146 and SN-mel336 as compared with RPMI (Supplementary Figure S7).We focused on the lists of differentially expressed genes (DEGs) that were upregulated in R848 activated pDCs while being down-regulated by SN-mel exposure (Figure 5D), as emphasized by volcano plots (Figures 5C, E, F).For instance, IFNA as well as IFNL genes expression was significantly down-regulated in SN-mel-exposed pDCs (Table 1; Figures 5E, F).Additionally, chemokines such as CXCL9, CXCL10 and CXCL11, were significantly down-regulated in SN-mel exposed pDCs (Table 1; Figures 5E, F).Remarkably, also mRNA expressions of CD274/PD-L1 and PDCD1/PD-1 were significantly down-regulated in pDCs exposed to melanoma secretome (Table 1; Figures 5E, F) suggesting that pDCs become hyporesponsive to TLR-mediated stimulation when exposed to CM secretome (80).On the other hand, CD86 and CXCL2 were upregulated by SN-mel (Figures 5E, F).To assign biological relevance to DEGs, we run a comprehensive GSEA-based functional analysis.SN-mel-exposed pDCs revealed negative enrichment of functional gene sets related to innate immune response and I-IFN signaling (Table 2, Supplementary Data 2, Supplementary Figure S8A,  Figures 5G, H).Among metabolic pathways, glycolysis was identified as negatively enriched only in pDCs exposed to SN-mel146 (Supplementary Figure S8B) hinting glucose deprivation (81).We further identified positive enrichment of TGF-b signaling in pDCs exposed to SN-mel146 suggesting that melanoma cells might secrete TGF-b (82) (Supplementary Figure S8C).
To better understand the molecular mechanisms responsible for the transcriptional rewiring of SN-mel-exposed pDCs, we performed upstream regulator analysis (Table 3).Interestingly, the positive upstream regulators included TAB1 that is a signaling mediator between TGF-b receptor and the kinase TAK1.On the other hand, we recognized many negative upstream regulators implicated in response to damaged-or pathogen-associated molecular patterns (DAMPs and PAMPs) and IFNs signaling (e.g.TLR7 and IRF7; Table 3).Overall, bulk transcriptomic findings indicate that soluble components produced by melanoma cells, including immunosuppressive cytokines and metabolites, significantly impair pDC activation.

Functional impairment of pDCs is mediated by immunosuppressive cytokines via IRF7 downregulation, while IRF3 activation is preserved
Tumor cells, including melanoma cells, produce and release a broad set of immunosuppressive cytokines (83-87).A diminished capacity of pDCs to produce IFN-a upon TLR7/9 stimulation has been previously reported in HNSCC (29), breast (88) and ovarian cancers (89).Transforming growth factor b (TGF-b) is abundant in these cancers and has been identified as the main soluble factor in the TME responsible for suppressing IFN-a secretion by TA-pDCs through inhibition of IRF7 signaling (83, 84, 86).By exploring pancancer TCGA dataset, we found that immunosuppressive cytokines, such as TGF-b, PGE-2 and CSF1, are highly expressed across all cancer types, including melanoma samples (Figure 1E).Moreover, PTGES2 expression showed an inverse weak correlation with PDC signature and TGF-b signature (Figures 1F, G).On the contrary, IL-10 resulted expressed in a less relevant level.To further support this data, we tested the production of IL-10 and TGF-b, as well-known immunosuppressive cytokines in melanoma microenvironment, by melanoma cell lines.We found a significantly higher concentration of IL-10 (p = 0.0019) and TGF-b2 (p < 0.0001) protein in SN-mel146 compared to the others SN-mel (Figures 6A, B), while TGF-b1 protein was detected in all SN-mel (p = 0.0508) (Figure 6C).By RNAscope analysis of a set of PCM cases, we confirmed the expression of TGF-b in PCM area infiltrated by pDCs (Figure 6D), while the presence of IL-10 was almost undetectable (Supplementary Figure S9).Moreover, we detected a considerable inhibition of IRF7 constitutive expression in pDCs after their exposure to SN-mel (Figure 6E), providing evidence that TLR/ MyD88-dependent signaling in pDCs was inhibited by melanoma secretome.In parallel, we explored STING-dependent signaling in pDCs stimulated with ADU S100 (Figure 6F).Activated STING recruits the TANK binding kinase-1 (TBK1) that is phosphorylated and in turn phosphorylates the interferon regulatory factor 3 (IRF3) and promotes its nuclear translocation (90) where IRF3 exerts its transcriptional function on immune-stimulated genes (ISG) and I-IFNs.By measuring the level of STING, TBK1 and IRF3 phosphorylation in pDCs exposed to melanoma supernatants, we could not detect evidence of molecular impairment of the STING pathway (Figure 6F).These results demonstrate that cGAS-STING signaling is not molecularly impaired in melanoma-conditioned  pDCs; however, data using STING agonist confirmed a defective pDCs activation by melanoma-derived soluble factors (Figures 4C, H).These data indicate that melanoma cells and the surrounding TME secrete immunosuppressive cytokines that might reduce the ability of PCM-associated pDCs to produce IRF7-dependent IFN-a, thus modulating pDCs anti-tumor function (91).

Melanoma secretome impairs glycolysis in TLR7-activated pDCs
Administration of TLR-7/9 agonists promptly increases glycolysis in human pDCs, as measured by extracellular acidification rate (ECAR) (92,93).Moreover, the inhibition of glycolysis by 2-deoxyglucose (2-DG) impairs type I IFN production and IFNA mRNA induction.Overall, these findings suggest that a metabolic switch towards glycolysis supports IFN-a production by activated pDCs (92,93).Glucose deprivation and lactate accumulation, as we previously reported in SN-mel, might thus limit IFN-a production (49).We tested this hypothesis on freshly purified pDCs, by measuring extracellular acidification on the Seahorse XFe24 Extracellular Analyzer.The glycolytic proton efflux rate (glycoPER) increases upon R848 administration (Figure 7A).However, SN-mel336 conditioning hinders the basal glycolysis and, consequently, the R848-induced glycolysis in pDCs, as confirmed by a significant reduction in glycoPER (Figures 7B, C).GlycoPER highly correlates with extracellular lactate accumulation, the end product of aerobic glycolysis.These results suggest that SNmel conditioned pDCs are impaired in glycolytic metabolism, independently of TLR7 activation.We could also observed a trend toward a reduced compensatory glycolysis following mitochondrial inhibition (i.e.oxidative phosphorylation inhibition by Rot/AA) in SN-mel336 exposed pDCs (Figure 7D), indicating a reduced ability of these cells to switch towards glycolysis to meet the cell energy demands.
Our results demonstrated that SN-mel negatively affects pDC glycolysis, suggesting a mechanism to explain the defective IFN-a production by SN-mel-exposed pDCs.

Discussion
PCM are divided in four major groups based on their molecular and immune landscape as well as their treatment options.Tumor immune microenvironment (TIME) in melanomas is highly complex in term of cell types, distribution and prognostic relevance (94).Our previous microscopic studies indicate that pDCs infiltrate early stages PCM, but they decrease over disease progression, and almost disappear in distant metastasis (46).Accordingly, circulating pDCs compartment collapsed in peripheral blood of CM patients with advanced disease stage, as a results of a defect of pDC differentiation from CD34 + progenitors.We and others could also detect a combined functional defect in circulating pDCs of melanoma patients in term of IFN-a and CXCL10 production, associated to poor survival (34,46,48,49).Here, based on a set of in vitro,  Frontiers in Immunology frontiersin.orgmicroscopic and in silico findings we found that pDCs infiltrating PCM are defective in IFN-a production and are hijacked in their functions, even when properly activated through the TLR7/9 or cGAS-STING signaling.Mechanistically, our findings suggest that PCM derived immunosuppressive cytokines as TGF-b and reduced glycolysis rewired fully differentiated pDCs towards a tolerogenic state (summarized by the graphical abstract in Figure 8).Properly activated pDCs drive melanoma regression, as shown by experimental model and clinical studies (95-98).In PCM, pDC infiltration is heterogeneous, occurs early, and localizes at the invasive margin, the site where pDCs interact with CD8 + T cells (46,99,100).A set of data from the present study indicate that pDCs efficiency in I-IFN production is impaired, suggesting that they are not activated or are profoundly refractory to melanoma-derived stimuli.
Based on RNA-Seq analysis, exposure to melanoma secretome rewired fully differentiated pDCs towards an IFN-defective and tolerogenic state.Specifically, transcripts for type I and type III interferons, T-cell recruiting pro-inflammatory cytokines (e.g., CXCL10) and costimulatory molecules (e.g., CD83, CD40) are significantly downregulated.Moreover, melanoma secretome induces expression of a set of transcripts encoding for protein with tumor-promoting functions including CLEC17A, FR2, SMOX, TOX and ENPP1.Interestingly, TOX drives the T-cell exhaustion within TME, with decreased cytokine production and increased expression of several inhibitory surface receptors (76,77,101); ENPP1 hydrolyzes both extracellular ATP to AMP, which is metabolized to immune-suppressive adenosine, and 2′,3′-cGAMP playing a key role as negative regulator of the STING pathway (102)(103)(104).Immunosuppressive cytokines, such as TGF-b, IL-10 and PGE-2, might play a role in promoting the pro-tumorigenic properties of pDCs (29,83,84,105).Specifically, immunosuppressive cytokines inhibit secretion of IFN-a by TLR7/9-activated pDCs and involves inhibition of IRF7 expression in pDCs, thus reducing I-IFN production (83, 84, 87, 106-108).A variable level of IL-10, TGF-b1 and TGF-b2 has been measured in SN-mel.These findings were confirmed in the TCGA dataset and, for TGF-b1, on a small institutional cohort of pDC-enriched PCM.Accordingly, TAB1, a mediator of the TGF-b pathway, resulted as activated upstream regulator in pDCs exposed to SN-mel, suggesting that TGF-b might play a relevant role as IFN-a suppressor in pDCs.
The data presented in this study hint that pDCs impairment in melanoma might result from defective TLR/MyD88-dependent signaling (i.e., IRF7 activation) or from a broad perturbation of their metabolic functions (i.e., glycolysis).Metabolic pathways regulate innate and adaptive immune responses to activation signals, including TLR agonists (109).We previously demonstrated that lactic acidosis exerted immunosuppressive function on human pDCs (46,49,110).Lactic acidosis in the TME might be induced by high glycolytic tumor metabolism.Accordingly, RNA-Seq analysis revealed a glycolytic impairment in SN-mel exposed pDCs that has been confirmed by a reduced glycolytic extracellular acidification, suggesting that glucose deprivation or lactic acidosis by melanoma cells (49) hindered glycolytic metabolism in pDCs (31,111) and consequently their innate immune functions, such as IFN-a secretion (92,93).TLR stimulation caused metabolic reprogramming in DCs, which is critical for immune activation (112).Specifically, TLR-driven activation induces glycolysis to meet the energetic demands for immediate secretion of IFN-a by pDCs (92,113).Moreover, TLR7/ 8 agonists up-regulates the oxidative phosphorylation and glutamine metabolism in pDCs, leading to a higher production of IFN-a and to increase T cell responses (114).On the contrary, inhibition of glutaminolysis and OXPHOS prevents pDC activation (114).Among PRRs responsible for pDCs functions, the activation of STING pathway by tumor-derived dsDNA (61) plays an essential role in DC recognition of dying tumor cells and potently enhanced antitumor cytotoxic T cell responses (40,59,61,115).By microscopy, using PLA, we found that endogenous activation of the cGAS-STING in PCM-infiltrating pDCs is limited to areas of microscopic regression.Longer DNA is more potent in activating cGAS by promoting liquidlike droplet formations in which cGAS and dsDNA are spatially concentrated for efficient cGAMP synthesis (116-118).The abundance of nucleases (DNases) in the extracellular milieu might explain the lack of dsDNA sensing via cGAS-STING by most pDCs (116).However, it is well known that cancer cells are often repleted with cytosolic dsDNA that might derive from the rupture of micronuclear envelopes as well as from acute genomic stressors induced by radiation, cisplatin, and intrinsic DNA damage (119).Moreover, cGAS cytosolic DNA sensors in cancer also include the mitochondrial DNA released as a consequence of oxidative stress and mitochondrial dysfunction (120).Finally, beyond the classical cGAS-cGAMP-STING axis, recent work has revealed that cGAS and STING may act independently from one another (121).Therefore, melanoma tumors lacking cGAS expression may still sustain active STING through other DNA binding proteins, such as IFI16 (121).
By stimulating blood purified pDCs with a STING agonist, we could detect pathway activation (i.e.STING, TBK1 and IRF3 phosphorylation), but reduced IFN-a production by SN-mel.In CM the expression and activation of components of the cGAS- In summary, the present study demonstrates that the pDC compartment in PCM is defective in term of type I and III interferons production.Functional impairment of melanoma-conditioned pDCs is driven by immunosuppressive microenvironment and/or metabolic drift.In clinical perspective, the data obtained from the current work will enable to define better strategies for appropriate administration of STING and TLR agonists, in combination with existing or novel immunotherapies for CM.Of relevance, based on data from this study, major clinical benefits in term of survival, would derive from the treatment of early-stage PCM in the adjuvant setting, where the pDC compartment is still preserved.In general, the complete restoration of the pDC functions could be useful in melanoma's treatment outcome.For instance, chemokine-based approaches (e.g.CXCL10) in combination with the standard BRAF and MEK inhibitors has been demonstrated to induce complete pathologic responses in mouse model (135).Among biomarkers for patient selection, responders are more likely to be identified within the CM subgroups lacking a TGF-b and IL-10 modulated TIME.It should be noted that a cross-regulatory pDCs in melanoma immune microenvironment are impaired in IFN-a production.The graphical abstract summarize data obtained from the present study on melanoma-associated pDCs.Endogenous activation of pDCs results in I-IFN and CXCL10 production, with the latter contributing to the amplification of the local T-cell infiltration.Exposure of pDCs to melanoma TME hijacks pDC functions via TGF-b and oncometabolites (i.e.increased lactic acid and reduced glucose).Specifically, the reduction of both glucose uptake and lactic acid export by pDCs impairs their proficiency to IFN-a production.Adapted from "Tumor Microenvironment", by BioRender.com(2023).Retrieved from https://app.biorender.com/biorender-templates.mechanism between I-IFN signaling pathways has been identified in pDCs.Specifically, STING-mediated pathway induced expression of negative regulator SOCS1 that, through an autocrine inhibitory loop, dampened MyD88-dependent IFN production (14,136) ruling out the combined administration of both TLR9 and STING agonists.

2 Limited 3 cGAS
FIGURE 2 Limited MxA expression in human PCM.(A) Representative sections of lupus erythematosus (LE; upper panel) and primary cutaneous melanoma (PCM; lower panel) biopsies showing high and low levels of MxA expression, respectively.Sections are stained as label and counterstained with Haematoxylin.Magnification 100X.Scale bar 100 µm.(B) Dot plots show the percentages of MxA positive signal on evaluated area in PCM, benign nevi (NV) and LE biopsies.Bars represent the mean of biological replicates.The statistical significance was calculated by One-way ANOVA (p< 0.0001) and Bonferroni multiple comparison; ***p < 0.001.(C) Dot plots show the percentage of MxA + marker area, obtained by digital microscopy analysis, in T1 versus T2-T4 stages of PCM.Bars represent the mean of biological replicates.The statistical significance was calculated by unpaired Student's T test (p= 0.0411); *p < 0.05.(D) Linear regression analysis between MxA expression and BDCA2 + pDCs density in PCM.(E) Representative images of BDCA2 (left panel) and MxA (right panel) staining in a PCM case.Magnification 100X.Scale bar 200 µm.

4 SN
FIGURE 4SN-mel impair the IFN-a and CXCL10/IP-10 production by pDCs.pDCs isolated from HD were exposed to SN-mel for 24 h.pDCs were stimulated with R848 for 2 h (A), 6 h (D) and 24 h (F, I) or with CpG-ODN 2216 or ADU S100 for 6 h (B, C, E) or 24 h (G, H, J).Brefeldin A was added 1 h and 4 h prior to 2 h and 6 h stimuli respectively.Both IFN-a and CXCL10 intracellular production (A-E) and secretion (F-J) were measured by flow cytometry and ELISA, respectively.Scatter dot plots show the percentage of IFN-a or CXCL10 positive cells evaluated on BDCA-2+/CD123+ pDCs (A-E) or the protein concentration per 1x10 5 cells measured on pDC supernatants (F-J).Bars represent the mean of biological replicates.The statistical significance was calculated by One-way ANOVA (A p= 0.048; B p= 0.0007; C p= 0.0264; D p= 0.0003; E p= 0.0001; F p< 0.0001; G p= 0.002; H p=0.009; I p= 0.3; J p= 0.0012) and Bonferroni multiple comparison; * p < 0.05; ** p < 0.01; *** p < 0.001.

5 SN
FIGURE 5 SN-mel modulates the pDC transcriptomic profile toward an immune suppressive state.(A) Graphical abstract depicts the experimental settings.Created with BioRender.com.(B) Scatter plots showing the combined projections of the first three components of a principal component analysis (PCA) run considering the whole gene expression on the SN-mel-exposed or RPMI-cultured pDCs samples resting and R848 stimulated.Samples' features are highlighted by colour code.(C) Volcano plots showing the log2 Fold Change and -log10 p-value of differential expression among RPMI-R848 stimulated vs RPMI-CTRL unstimulated pDCs.(D) Venn diagram shows the number of upregulated genes in R848 stimulated pDCs versus unstimulated pDCs (purple) and the number of downregulated genes in unstimulated pDCs exposed to SN-mel 146 (red) and SN-mel 336 (orange) as compared to unstimulated pDCs in RPMI condition.(E, F) Volcano plots showing the log2 Fold Change and -log10 p-value of differential expression among SN-mel146-CTRL vs RPMI-CTRL (E) and SN-mel336-CTRL vs RPMI-CTRL (F) pDCs.Vertical red lines highlight a | Log2FoldChange| = 1.(G, H) Enrichment plots for the set of the interferon signaling in the transcriptome of SN-mel146 (G) and SN-mel336 (H) exposed pDCs versus RPMI cultured pDCs by GSEA.

FDR
q-val ES NES FDR q-val ES NES FDR q-val ES NES

6 IL- 10
FIGURE 6 IL-10 and TGF-b production by melanoma cells and concomitant IRF7 downregulation, but preserved STING signaling, in melanoma-conditioned pDCs.IL-10 (A), TGF-b2 (B), and TGF-b1 (C) secretion by melanoma cell lines was measured by ELISA on SN-mel collected after 72 h of culture.Scatter dot plots show the concentration of cytokines per 1.5x10 6 seeded cells.Bars represent the mean value of biological replicates (n =3-5).The statistical significance was calculated by one-way ANOVA (A p= 0.0019; B p< 0.0001; C p= 0.0508) and Bonferroni's multiple comparison test.(D) Representative images of TGF-b RNAscope (left panels) and BDCA2 staining (right panels) in PCM cases.Magnification 200X.Scale bar 100 µm.(E) Immunoblots showing IRF7 expression in pDCs cultured in RPMI medium or SN-mel for 24 h, as labeled.a-tubulin was used as housekeeping control.(F) Immunoblots showing STING, TBK1, and IRF3 expression and their phosphorylation p-STING (Ser366), p-TBK1 (Ser172), and p-IRF3 (Ser396) in pDCs cultured in RPMI or SN-mel as labeled.Histone H3 and a-tubulin were used as housekeeping control.

7 SN
FIGURE 7SN-mel exposure reduced glycolysis in pDCs.pDCs were cultured in RPMI medium or SN-mel 336 with no stimuli for 24h.pDCs were then subjected to the Glycolytic Rate Assay and glycolytic real-time proton efflux rate (glycoPER) is derived over the indicated time points (A).Vertical lines indicate addition of R848, rotenone plus antimycin A (Rot/AA) and 2-deoxy-D-glucose (2-DG).Colors represent different healthy donors and dots represent the mean value of technical replicates.Basal glycolysis corresponds to the glycoPER value obtained at the time point before R848 injection (B); p = 0.013).Induced glycolysis is calculated on the average glycoPER values over the four time points after R848 treatment (C); p = 0.042).Compensatory glycolysis corresponds to glycoPER value measured at the third time point after Rot/AA injection (D); p = 0.056).Bars represent the mean ± SD of biological replicates (n = 5).The statistical significance was calculated by two-tailed Paired Student's t-test; * p < 0.05.

TABLE 1
IFNA and chemokines genes expression in pDCs.

TABLE 2
Interferon-related pathways negatively enriched in SN-mel-exposed pDCs resulting from the Gene Set Enrichment Analysis (GSEA).

TABLE 3
Upstream regulators in SN-mel-exposed pDCs predicted to be inhibited or activated based on IPA analysis.