MCT1 lactate transporter blockade re-invigorates anti-tumor immunity through metabolic rewiring of dendritic cells in melanoma

Niveau, Camille; Cettour-Cave, Mélanie; Mouret, Stéphane; Sosa Cuevas, Eleonora; Pezet, Mylene; Roubinet, Benoît; Gil, Hugo; De Fraipont, Florence; Landemarre, Ludovic; Charles, Julie; Saas, Philippe; Aspord, Caroline

doi:10.1038/s41467-025-56392-x

Download PDF

Article
Open access
Published: 27 January 2025

MCT1 lactate transporter blockade re-invigorates anti-tumor immunity through metabolic rewiring of dendritic cells in melanoma

Nature Communications volume 16, Article number: 1083 (2025) Cite this article

8695 Accesses
14 Citations
9 Altmetric
Metrics details

Subjects

Abstract

Dendritic cells (DC) are key players in antitumor immune responses. Tumors exploit their plasticity to escape immune control; their aberrant surface carbohydrate patterns (e.g., glycans) shape immune responses through lectin binding, and manipulate the metabolism of immune cells, including DCs to alter their function and escape immune surveillance. DC metabolic reprogramming could induce immune subversion and tumor immune escape. Here we explore metabolic features of human DC subsets (cDC2s, cDC1s, pDCs) in melanoma, at single cell level, using the flow cytometry-based SCENITH (Single-Cell ENergetIc metabolism by profiling Translation inHibition) method. We demonstrate that circulating and tumor-infiltrating DC subsets from melanoma patients are characterized by altered metabolism, which is linked to their activation status and profile of immune checkpoint expression. This altered metabolism influences their function and affects patient clinical outcome. Notably, melanoma tumor cells directly remodel the metabolic profile of DC subsets, in a glycan-dependent manner. Strikingly, modulation of the mTOR/AMPK-dependent metabolic pathways and/or the MCT1 lactate transporter rescue cDC2s and cDC1s from skewing by tumor-derived glycans, Sialyl-Tn antigen and Fucose, and restore anti-tumor T-cell fitness. Our findings thus open the way for appropriate tuning of metabolic pathways to rescue DCs from tumor hijacking and restore potent antitumor responses.

Immuno-metabolic dendritic cell vaccine signatures associate with overall survival in vaccinated melanoma patients

Article Open access 08 November 2023

Cytokine-armed dendritic cell progenitors for antigen-agnostic cancer immunotherapy

Article Open access 23 November 2023

CD4⁺ helper T cells endow cDC1 with cancer-impeding functions in the human tumor micro-environment

Article Open access 13 January 2023

Introduction

Dendritic cells (DC) are active players in shaping antitumor immune responses^1,2. DCs are composed of different subsets harboring complementary functions^3,4. While conventional type 1 BDCA3⁺ DCs (cDC1s) excel in the cross-priming of tumor antigens to CD8⁺ T cells⁵, conventional type 2 BDCA1⁺ DCs (cDC2s) activate CD4⁺ T cells and drive Th1/Th17 responses⁶. BDCA2/4⁺ plasmacytoid DCs (pDCs) participate in NK cell and CTL recruitment, display direct cytotoxicity toward tumor cells and potentiate the function of other cells⁷. Recognition of pathogen- or/ damage-associated molecular patterns (PAMPs or DAMPs) by toll-like receptors (TLR) triggers IL-12p70, type I (IFN-α, IFN-β) or III (IFN-λ1) IFN production by cDC2s, pDCs and cDC1s respectively together with secretion of chemokines attracting effector cells, thus favoring antitumor responses.

However, the plasticity of DCs is used by tumors to escape immune control. Many ways of DC hijacking have been already identified^8,9. In melanoma patients, cDC1s were reported to be functional and associated with a favorable outcome, whereas cDC2s and pDCs display altered features^10,11,12,13 and exhibit altered crosstalk with antitumor immune effector cells driving pro-tumor regulatory responses^12,14, ultimately leading to poor clinical outcome. Such hijacking of DCs can be mediated by tumor-derived immunosuppressive factors (e.g., gangliosides, IL-6, IL-10, TGF-β, PGE2) and immunomodulatory molecules (e.g., TARC/CCL17, MDC/CCL22)¹². In addition, interactions between carbohydrate patterns on tumor cells and C-type lectin receptors (CLRs) on DCs are emerging pathways of immune subversion¹⁵. We recently demonstrated that melanoma tumor cells harbored aberrant glycan motifs on their surface that influence DC function¹⁶ and that circulating and tumor-infiltrating DCs from melanoma patients displayed major perturbations in their CLR profiles¹⁷. We further demonstrated that melanoma-derived glycan motifs can bind to human DC subsets through CLRs and induce DC dysfunction by influencing their expression of activation markers and immune checkpoints and reshaping their cytokine/chemokine secretion pattern^18,19.

Besides, immuno-metabolism emerges as critical in the regulation of immune cell functions in cancer^20,21,22,23. Indeed, the energetic profile and metabolic pathways determined cell activation, differentiation and functions^24,25. Cells use two main pathways –oxidative phosphorylation (OXPHOS) and aerobic glycolysis – to produce ATP, mostly governed by AMPK and mTOR signaling pathways, respectively. Upon activation, immune cells undergo a metabolic switch from OXPHOS to glycolysis, which provides essential resources to support new protein synthesis for cell proliferation, activation and function. Notably, metabolic pathways crucially orchestrate DC activation and function²⁶. A switch toward enhanced glycolysis occurs rapidly upon mouse bone marrow (BM)-derived DC activation mediated by HIF1α or PI3K-AKT pathway, and is crucial for appropriate mouse DC maturation, functions, and cytokine secretion²⁷. Importantly, the metabolic flexibility of human monocyte-derived DCs can be tuned by the microenvironment to further guide the elicitation of inflammatory or tolerogenic DCs, which is determinant to subsequently drive immunity or tolerance²⁸. Immunogenicity versus tolerance of DCs are crucially determined by the orientation of metabolic pathways that govern their functions^28,29. Overall, metabolism governs immune responses.

Recent studies suggest modulations of the metabolism during immune dysfunction, especially in cancer^30,31. Tumor cells undergo a metabolic switch from OXPHOS to glycolysis to support active proliferation, creating an environment hostile to infiltrating immune cells, with competition for limiting nutrients, oxygen changes and accumulation of toxic metabolites in the tumor microenvironment (TME) that impair immune responses^20,21,22. Indeed, tumors create a suppressive metabolic microenvironment that supports the metabolism of tumor-promoting cells (regulatory CD4⁺ T [Treg] cells, myeloid-derived suppressive cells [MDSC], tolerogenic DC) and dampens immune cell functions, affecting immune cell proliferation, survival, and functionality. This contributes to ineffective immune responses and cancer progression²⁰. Such metabolic stress interferes with DC differentiation, recruitment, maturation and function³². Dysregulation of tumor-associated DC metabolism is associated with their reduced immunostimulatory capacities and functions³³. Thus, tumors, through a metabolic hostile microenvironment, manipulate the metabolism of immune cells to hijack their function and escape immune surveillance.

The two emerging hallmarks of cancer, metabolic reprogramming and immune evasion, are crucially linked²³. Immunometabolism is of growing interest and considered as an interesting therapeutic target to orientate and fine-tune immune responses. As metabolism crucially regulates DC function, their metabolic reprogramming could be a pathway of immune subversion and tumor immune escape. Yet, currently, metabolic features and perturbations of immune cells in melanoma together with their functional impact are unknown. This prompted us to explore the metabolic features of human DC subsets in melanoma. We used the flow cytometry-based SCENITH method (Single-Cell ENergetIc metabolism by profiling Translation inHibition), developed by R. Arguello et al. to determine the metabolic profiles of DC subsets at a single cell level. SCENITH is based on measuring the level of protein synthesis (PS) directly within cells by flow cytometry³⁴ using puromycin - a drug able to inhibit translation but whose incorporation is a reliable method for measuring PS level. Thus, PS reflects the global metabolic activity of a cell, and changes in PS under the action of metabolic inhibitors allows depicting the metabolic profiles of the cells. We deciphered the metabolic profiles of circulating and tumor-infiltrating DC subsets from melanoma patients, their link with the phenotypic and functional features of DCs, as well as their impact on patient clinical outcome. We then explored the direct impact of tumor cells and tumor-derived glycans on the metabolic profiles of DC subsets.

Our study reveals major metabolic disturbances of circulating and tumor-infiltrating DC subsets in melanoma with functional and clinical impacts. We further reveal that melanoma tumor cells remodel DC metabolism and features in melanoma patients through sialylated and fucosylated glycan motifs. Strikingly, modulation of the mTOR/AMPK-dependent metabolic pathways and/or the MCT1 lactate transporter rescued DCs from skewing by the tumor-derived glycans sialyl-Tn antigen and fucose while restoring their functions. Our findings uniquely bring insights into the link between metabolism, tumor glycode and immune evasion in melanoma, and paves the way for innovative therapeutic options exploiting metabolic pathways to overcome immune subversion and restore immune function, ultimately improving the success of immunotherapies.

Results

Circulating and tumor-infiltrating DC subsets from melanoma patients harbor an altered metabolism when compared with circulating DC subsets from healthy donors

To decipher the energetic metabolism of DC subsets in melanoma, we explored the metabolic profiles of circulating and tumor-infiltrating DC subsets in patients compared to healthy donors (HDs) using the SCENITH method (Fig. 1A). DC subsets, e.g.,; cDC2s, cDC1s and pDCs, were first discriminated using a multiparametric flow cytometry approach (Supplementary Fig. 1A), and their incorporation of puromycin was measured in presence of inhibitors (2-Deoxy-Glucose [DG] and Oligomycin [O]) that abrogate the two major metabolic pathways for energy production in cell, glycolysis and OXPHOS respectively (Supplementary Fig. 1B). These measurements allow us assessing the global energetic status (in absence of inhibitors, e.g., direct MFI of puromycin) and calculating the contribution of each pathway to the global metabolism, called dependencies and capacities (Supplementary Fig. 1C). Dependencies represent the direct contribution of each pathway to the general metabolism, while capacities depict the ability of the cell to use a pathway when the other is inhibited. We explored DC metabolic profiles, both at basal state and upon TLR stimulation, to decipher their intrinsic metabolic features and their potential perturbations associated with TLR pathways.

**Fig. 1: Circulating and tumor-infiltrating DC subsets from melanoma patients harbor a perturbed energetic metabolism in comparison with DC subsets from healthy donors.**

Heat maps displaying the overviewed metabolic parameters of DC subsets highlighted noteworthy differences in the metabolic patterns of DCs from melanoma patient blood and tumor in comparison with blood DCs from HDs (Fig. 1B). The raw puromycin incorporation data in the presence or not of inhibitors used for the calculation of dependencies and capacities are shown in Supplementary Fig. 1D. At basal state, we found that blood and tumor-infiltrating cDC2s, cDC1s, and pDCs from melanoma patients exhibited significantly higher levels of global metabolism compared with DCs from HD (Fig. 1C). This is associated with metabolic switches within cDC1s and pDCs. Indeed, tumor-infiltrating cDC1s and pDCs displayed a decreased glucose-dependent metabolism concomitant to an increased fatty acid and/or amino acid oxidation (FAO/AAO) capacity compared to circulating DCs from HDs and/or patients (Fig. 1D). Tumor-infiltrating pDCs displayed both increased capacities and decreased dependencies, highlighting their high metabolic plasticity. Furthermore, circulating cDC1s from melanoma patients strongly increased their mitochondrial dependency while lowering their glycolytic capacity compared with HDs, whereas tumor-derived pDCs underwent a decrease of mitochondrial dependency and an enhancement of their glycolytic capacities (Fig. 1D). Despite no difference in global metabolism between blood and tumor of patients (Fig. 1C), metabolic switches appeared to be divergent in circulating and tumor-infiltrating cDC1s and pDCs (Fig. 1D). We further analyzed the metabolic profiles of DC subsets after TLR stimulation. TLR-activated circulating cDC2s, cDC1s and pDCs from melanoma patients all exhibited a higher level of global metabolism compared to HDs, while tumor-infiltrating DCs did not show significant change in this parameter (Fig. 1E). Strikingly, all DC subsets displayed metabolic switches upon TLR pathway engagement compared to HDs, characterized by a reduced glycolytic dependency concomitant to an enhanced FAO/AAO capacity, in both blood and tumor (Fig. 1F). The clinical settings of patients (i.e., disease stage and/or previous treatment before sampling) did not influence the metabolic features of DCs, as no differences according to disease stage for blood samples or prior treatment for tumor samples, both a basal state (Supplementary Fig. 2A) and upon TLR stimulation (Supplementary Fig. 2B) were noticed.

These findings prompted us to deeply explore the metabolism within different recently described DC subsets (that share common markers with pDCs or cDC2s)^{35,36,37,38,39}, and depict metabolic features according to heterogeneous DC populations. We thus assessed the metabolic profiles of circulating and tumor-infiltrating DC2s (including CD5⁺ and CD5^– subsets), DC3s (including CD14⁺ and CD14^– subsets), AS DCs and “pure” pDCs in patients compared to HDs using the SCENITH method (Supplementary Figs 3 and 4). Among Lin^– (including the CD14 marker) HLA-DR⁺ cells, pure pDCs were identified as CD16^– BDCA3^– CD123⁺ CD11c^– AXL^– SIGLEC6^–, and AS DCs defined as CD16^– BDCA3^– CD123⁺ CD11c^–/lo AXL⁺ SIGLEC6⁺ (Supplementary Fig. 3, panel A). Among Lin (CD3/19/20/56)^– (without CD14) HLA-DR⁺ cells, DC2s and DC3s were depicted as CD88^– BDCA1⁺ CD11c⁺ BDCA3^–, and further distinguished as CD163^– CD14^– CD5^+/– DC2s and CD163⁺ CD14^+/– DC3s (Supplementary Fig. 3, panel B). At basal state, blood and tumor-infiltrating pure pDCs and AS DCs from melanoma patients exhibited higher levels of global metabolism compared with HDs, whereas within cDC2s, only tumor-infiltrating CD14^+/– DC3s (both CD14^+/– subsets) but not DC2s displayed increased global metabolism compared with HDs (Supplementary Fig. 4A). Metabolic switches were further observed in melanoma patients for pure pDCs, which displayed decreased mitochondrial dependency in tumor, and for AS DCs, which exhibited decreased glucose dependency in blood compared to HDs, whereas no major modulations occurred for DC2s and DC3s (Supplementary Fig. 4B). Upon TLR stimulation, global metabolism of DC subpopulations were similar in patients and HDs (Supplementary Fig. 4C), while circulating and tumor-infiltrating DC2s and DC3s from patients displayed a strong drop in mitochondrial dependency concomitant to an enhanced FAO/AAO capacity (Supplementary Fig. 4D). Thus, metabolism of DC subsets related to pDCs or cDC2s was also perturbed in melanoma patients.

To directly assess the glycolytic aptitude of DC subpopulations, we measured glucose uptake by DCs using 2-NBDG, a fluorescent analog of glucose (Supplementary Fig. 5). This was done at basal state (Fig. 2) and upon TLR stimulation (Supplementary Fig. 6). At basal state, we observed a decreased glucose uptake by tumor-infiltrating cDC1s in patients compared to HDs (Fig. 2), which is in accordance with their drop in glucose dependency and glycolytic capacity observed using the SCENITH method (Fig. 1D). In contrast, tumor-infiltrating pDCs, DC2s and DC3s exhibited an increased glucose uptake compared to HDs (Fig. 2). This is in agreement with their significant increased glycolytic capacity (for pDCs) or tendency toward an improved glycolytic capacity (for DC2 and DC3s) observed using the SCENITH method (Fig. 1D, Supplementary Fig. 4A). In the setting used, no differences in glucose uptake capacity of DCs were observed between patients and HDs upon TLR stimulation of the cells (Supplementary Fig. 6).

**Fig. 2: DC subsets exhibit distinct capacities of glucose uptake in melanoma patients at basal state.**

Overall, these results reveal significant alterations of metabolic profiles and metabolic switches of circulating and tumor-infiltrating DC subsets in melanoma patients, suggesting a role for immunometabolism in the subversion of DCs by melanoma.

The activation status and profile of immune checkpoints by DC subsets are linked with their metabolism

To explore the importance of metabolism on DC subset features, we investigated simultaneously metabolic profiles of DCs and expression of activation markers (CD40) as well as immune checkpoints (PD-L1 and TIM-3), both at basal state and upon TLR stimulation. Percentages of expression of these markers by DCs from HD blood, melanoma patient blood and tumor were measured by flow cytometry (Supplementary Fig. 7A). We exploited conditions harboring negative and positive populations for the corresponding marker among each DC subset. Heat maps displaying an overview of metabolic parameters of each DC subset according to their expression (+) or not (-) of CD40, PD-L1 or TIM-3, at basal state or upon TLR stimulation, revealed distinct metabolic profiles according to the activation status (i.e., expression of CD40) and immune checkpoint expression of DCs (Fig. 3A). At basal state, expression of CD40 by cDC2s and cDC1s from HD blood was associated with a lower level of global metabolism together with an enhanced mitochondrial dependency concomitant to a decreased glycolytic capacity (Supplementary Fig. 7B), and similar pattern was observed for CD40-expressing cDC2s (Fig. 3B) and cDC1s (Fig. 3C) from melanoma patient blood. In contrast, CD40 expression by pDCs was linked with a decreased glucose dependency and an increased FAO/AAO capacity in melanoma patient blood, while associated with a higher level of metabolism in melanoma tumor (Fig. 3D). Besides, changes in metabolic profile according to PD-L1 expression was more heterogeneous among DC subsets. TLR-stimulated cDC2s from HDs and tumor-infiltrating cDC2s associated PD-L1 expression with a higher level of metabolism, together with an increased glucose dependency for HDs (Supplementary Fig. 7C and 7D). TLR-stimulated PD-L1⁺ pDCs from HDs also exhibited a higher level of metabolism, but associated with a decreased mitochondrial dependency (Supplementary Fig. 7C). In contrast, PD-L1 expression by cDC1s from HDs after TLR-L stimulation (Supplementary Fig. 7C) and from melanoma tumors at basal state (Fig. 3C) was linked with a lower level of global metabolism, and correlated with an enhanced mitochondrial dependency in stimulated circulating cDC1s from patients (Supplementary Fig. 7D). The link between TIM-3 expression and metabolic patterns of DCs was homogenous among DC subsets and within groups. Indeed, TIM3-expressing circulating cDC2s and pDCs as well as tumor-infiltrating cDC2s, cDC1s and pDCs displayed an increased level of global metabolism at both basal state and upon stimulation (Fig. 3B–D and Supplementary Fig. 7B–D). TIM-3 expression was strongly associated with a decreased mitochondrial dependency concomitant to a drastic enhancement in glycolytic capacity, for all DC subsets and within all groups (Fig. 3B, D and Supplementary Fig. 7B, 7C, 7D). These observations underlined that DC subsets display distinct metabolic profiles depending on their activation status and immune checkpoint profile, underlining critical connections between DC features and metabolic pattern.

**Fig. 3: DC subsets display distinct metabolic profiles depending on their activation status and immune checkpoint profile.**

The metabolism of DC subsets influences their function and affects patient clinical outcome

As DC metabolism is altered in melanoma patients and influenced their phenotypic features, we further investigated the potential correlation between DC metabolic patterns and functional features. We simultaneously assessed in response to TLR stimulation the metabolic profile of DC subsets and their pattern of secretion of a large panel of cytokines/chemokines involved in immunity, tolerance or inflammation (Supplementary Fig. 8). In order to explore the link between metabolic parameters and the secretome of DCs, we then performed Spearman correlations between these variables (Fig. 4A). First, in HDs, we observed that some mediators were highly glucose-dependent, such as MDC, and potentially TARC and MIP-1α, some others were mostly mitochondrial-dependent, such as IL-10 and MCP-1, and some others were negatively linked with both glycolysis and OXPHOS while being positively associated with global metabolism, such as IFN-α, IFN-λ1, RANTES and TNF-α (Fig. 4A). A strong correlation was observed between the production of MDC and the glucose dependency of cDC2s from HDs (Fig. 4A). In melanoma patients, these patterns were disturbed for some mediators. In melanoma patient blood, secretion of IFN-α and IFN-λ1 became mostly dependent on glycolysis of cDC2s and pDCs, whereas production of IL-12, IL-10 and IL-23 exhibited cDC glucose- and mitochondrial-dependency (Fig. 4A). Moreover, global metabolism of circulating cDC1s of patients looked positively correlated with several mediators of immunity (such as I-TAC, IFN-α, IFN-β, IL-12, IFN-λ1, IP10, IL-23) whereas this pattern was reversed for tumor-infiltrating cDC1s (Fig. 4A). Importantly, the pattern observed for DC subsets from melanoma tumors revealed a high glucose dependency, but also a global increase in mitochondrial dependency whatever the cytokine/chemokine (Fig. 4A). This might reflect a high metabolic plasticity of tumor-infiltrating DCs. These observations suggest a link between the metabolic status of DC subsets and the composition of their secretome in accordance with their microenvironment.

**Fig. 4: DC subset metabolism is linked with their function and influences the clinical outcome of patients.**

To further decipher whether the disturbed DC metabolism affects patient clinical outcome, we compared overall survival (OS) and progression free survival (PFS) of patients according to the metabolic profiles of their circulating and tumor-infiltrating DC subsets (Supplementary Table 4). Interestingly, the metabolism of circulating pDCs was pivotal for clinical outcome in melanoma. Indeed, a high level of global metabolism and a high mitochondrial dependency of stimulated circulating pDCs were associated with a better prognosis, whereas a high glycolytic metabolism was linked with shorter survival (Fig. 4B). In contrast, in tumor, higher levels of metabolism harbored by TLR-L-stimulated cDC1s and pDCs were associated with a bad prognosis (Fig. 4C). Altogether, these data proved that the metabolism of DC subsets is connected with their functional orientation and further influences the clinical outcome of melanoma patients.

Melanoma tumor cells directly hijack metabolic properties of DC subsets

To deepen the understanding of DC metabolic alterations in melanoma, we explored whether tumor cells could directly trigger such metabolic perturbations within DCs. We settled-up an experimental model where DCs purified from HDs were co-cultured with primary tumor cell lines derived from melanoma patients. We evaluated the impact of melanoma tumor cells on the metabolism of the three DC subsets directly or upon TLR stimulation (Fig. 5A). Heat maps displayed metabolic parameters of DC subsets upon culture or not with melanoma tumor cells and after stimulation or not with TLR-L (Fig. 5B). In unstimulated conditions, tumor cells drove a slight but significant increase in global metabolism of pDCs, contrary to cDC2s and cDC1s, which remained unaffected (Fig. 5C). In addition, although cDC1s and pDCs exhibited no change in the type of metabolism, cDC2s underwent a decrease in mitochondrial dependency associated with an enhanced glycolytic capacity upon contact with tumor cells (Fig. 5D, Supplementary Fig. 9A left). After TLR stimulation, pDCs experienced a more marked increase in global metabolism after co-culture with tumor cells, while cDC2s displayed a decreased energetic profile (Fig. 5E). These disruptions were accompanied by an increased mitochondrial dependency concomitant to a decreased glycolytic capacity for pDCs, while an inverse metabolic shift occurred for cDC2s (Fig. 5F, Supplementary Fig. 9A right). Interestingly, cDC1s were the only subset for which metabolism remained unaffected after contact with melanoma tumor cells.

**Fig. 5: Melanoma tumor cells directly alter DC subset metabolism, which dictate DC function.**

As metabolic features of circulating and tumor-infiltrating DC subsets influenced the clinical outcome of melanoma patients (Fig. 4B, C), we investigated the potential link between the metabolism of DC subsets from HDs after co-culture with melanoma tumor cells, and the survival of patients from whom the cell lines were derived (Supplementary Table 4, right part). We observed that cDC2s displaying a higher global metabolism and higher mitochondrial dependency in unstimulated conditions (Fig. 5G), or cDC1s harboring a higher global metabolism upon stimulation (Fig. 5G and Supplementary Fig. 9B) were triggered by tumor cells derived from patients with a better PFS and/or OS.

We further investigated whether disrupted DC metabolic patterns induced by tumor cells in vitro correlated with subverted DC functions. We simultaneously assessed in the presence or not of tumor cells and in response to TLR stimulation the metabolic profile of DC subsets and their pattern of secretion of a large panel of cytokines/chemokines involved in immunity, tolerance or inflammation (Supplementary Fig. 9C). In order to explore the link between metabolic parameters and secretome of DCs in this context, we then performed Spearman correlations between these variables. Cytokine/chemokine quantification revealed few differences in the secretome of DCs according to the presence of tumor cells, such as decreased levels of IL-12, MIG and TNF-α, and increased levels of MDC (Supplementary Fig. 9C, D). We studied correlations between DC secretome and metabolism either independently for the two experimental groups (with tumor and without tumor) (Supplementary Fig. 9E) or together (Fig. 5H). The pattern of correlation matrix was quite similar in absence or presence of tumor cells for glucose- and mitochondrial dependencies. Interestingly, tumor cells triggered a switch from negative to positive link between the global metabolism of cDC1s and the secretome (Supplementary Fig. 9E). In addition, we highlighted a strong positive relationship between levels of IFN-β and IFN-λ1 with a high global metabolism and a mitochondrial dependency of pDCs (Fig. 5H). Notably, MDC level was significantly negatively linked to the glucose dependency of cDC1s (Fig. 5H). Thus, melanoma tumor cells directly altered DC subset metabolism, driving modulations in global metabolism associated with metabolic switches, which in turn dictated subversion of DC function and patient outcome.

The glycocode of tumor cells dictates DC metabolism

To get further insights into the mechanisms underlying metabolic modifications of DCs triggered by melanoma tumor cells, we explored the potential involvement of the glycocode (i.e., the glycan motifs expressed at their surface) that we previously found to be aberrantly expressed by melanoma tumor cells and to influence DC function¹⁶. We thus studied interrelations between the metabolism of DC subsets in co-culture with primary melanoma tumor cells from patients, their function, and the glycocode of corresponding cell lines (Fig. 6A, Supplementary Fig. 10). We previously identified two groups of tumor cell lines based on their different glycocode composition and their negative or boosting influence on the capacity of cDC1s and pDCs to produce IFN-λ1 or IFN-α, respectively¹⁶. Here we analyzed and compared metabolic profiles of DC subsets triggered by tumors exhibiting a “negative” or “positive” influence on DC function (Fig. 6B). In the presence of “positive” tumors, cDC1s and pDCs tended to display a higher level of general metabolism (Fig. 6B, left). A strong switch in the metabolic balance of cDC1s was observed between the two groups of tumors with “positive” tumors triggering a diminished glucose dependency associated with an enhanced FAO/AAO capacity (Fig. 6B, right). To examine whether tumor-derived soluble factors or metabolites could influence DC metabolism and function, we assessed whether tumor-conditioned medium could affect DC metabolism as tumor cells did (Supplementary Fig. 11). At basal state, tumor-conditioned medium displayed similar effects on cDC2s than the corresponding tumor cells, driving a decreased global metabolism (Supplementary Fig. 11A) associated with a drop in mitochondrial dependency concomitant to an improved glycolytic capacity (Supplementary Fig. 11B), as previously observed with the 23 melanoma cell lines tested (Fig. 5D). However, upon TLR stimulation, the decreased global metabolism and metabolic switches of cDC2s induced by the tumor cells (previously observed in Fig. 5E, F) did not occur with the corresponding tumor-conditioned medium (Supplementary Fig. 11C and 11D). For pDCs, tumor-conditioned mediums did not reproduce the metabolic modulations induced by tumor cells (Supplementary Fig. 11). Thus, interactions between tumor cells and pDCs were required for metabolism modulation, while metabolism modulation of cDC2s at basal levels was dependent of soluble factors or metabolites released by melanoma cells.

As tumor-conditioned medium influenced cDC2 metabolism, we further assessed soluble factors and metabolites potentially involved in such effect. Tumor supernatants contained significant amounts of immunomodulatory factors, such as IL-6, IL-8, MDC, MCP-1 MIP-1β and TGF-β (Supplementary Fig. 11E). These factors are known to modulate DC function. However, no major differences were found between tumor lines exhibiting a positive or negative impact on DC function, whereas a differential impact was observed on DC metabolism (Fig. 6B). This excludes a modulatory role of these factors on DC metabolism. We then evaluated glucose and lactate amounts in tumor supernatants, as their level could influence immune cell metabolic activity. We observed heterogeneous levels of glucose and lactate in tumor supernatants, which were strongly negatively correlated between them (Supplementary Fig. 12A). Correlations between glucose/lactate ratios found in tumor supernatants and metabolism of DC subsets cocultured with the corresponding cell lines did not allow to link these metabolites with the metabolic orientation of DC subsets (Supplementary Fig. 12B). In addition, glucose and lactate levels found in tumor supernatants were similar between tumor cells triggering a positive or negative impact on cDC1s or pDCs’ function (Supplementary Fig. 12C). However, glucose level correlated positively with MIP-1α and negatively with IL-29/IFN-λ1 ratio (w/wo tumor) secretions while lactate level correlated negatively with MIP-1α and MCP1 secretions by DCs cocultured with the corresponding tumor cells (Supplementary Fig. 12D), sustaining a link between glucose/lactate levels and DC function.

We further examined potential correlations between the metabolism of DC subsets upon co-culture with tumor cells and the level of expression of specific glycan motifs by the different tumor cell lines analyzed by lectin binding, as previously done in Sosa Cuevas et al.¹⁶ (Fig. 6C). Interestingly, the energetic metabolism of cDC2s and their glucose dependency correlated positively (r = 0.59) and negatively (r = -0.64) with the binding of HPA by tumor cells (revealing GalNAc motifs), respectively (Fig. 6C). The mitochondrial dependency of cDC2s upon TLR stimulation negatively correlated with the binding of MAA (revealing NeuAc motifs) (r = -0.57) (Fig. 6C). In addition, pDC global metabolism was negatively correlated with PSA fixation to tumor cells (revealing Man/Glc expression), and their glucose dependency upon TLR stimulation was positively linked to the HPA binding (GalNAc motifs). To deeply explore the involvement of glycans in the modulation of DC metabolism and exclude the possibility that the observed effects could be due to HLA mismatches between tumor cells and DCs, we deglycosylated the tumor cells prior to contact with DCs by removing either sialylated and fucosylated motifs (by treatment with neuraminidase and fucosidase) or all N-linked and O-linked oligosaccharides motifs (using a mixture of N-Acetylgalactosaminidase, Galactosidase, Glucosaminidase, N-Glycosidase F and Neuraminidase) (Fig. 7, Supplementary Fig. 13). At basal state for cDC2s and pDCs, the drop of mitochondrial dependency and increased glycolytic capacity triggered by tumor cells were dampened upon sTn/Fuc removal or the whole deglycosylation of tumor cells (Fig. 7A). In addition, the removal of surface glycans on tumor cells triggered a strong increase in glucose dependency and an abrogation of FAO&AAO capacity of both cDC2s (Fig. 7A) and cDC1s (Supplementary Fig. 13A) at basal state, as well as robust metabolic rewiring toward increased dependencies and decreased capacities of cDC2s and pDCs upon TLR stimulation (Supplementary Fig. 13B). These data demonstrated that tumor-associated glycans can directly elicit metabolic modulations of DCs. The removal of tumor-associated glycans strongly potentiated DC function, as illustrated by increased secretions of IFN-β, IL-29/IFN-λ1, IL-18, IP-10, I-TAC while dampening productions of MCP1, MIP-1β (Fig. 7B, C, Supplementary Fig. 13C, 13D). This further emphasize the link between glycans and functional orientation of DCs. Overall, these data demonstrated that the tumor glycocode, especially sialylated and fucosylated motifs, is a crucial parameter that drives DC metabolic and functional alterations.

**Fig. 7: Tumor deglycosylation affects DC metabolism and function.**

Tumor-derived glycans induce changes in conventional DC metabolism with Sialyl-Tn antigen, GlcNAc and Fucose affecting both cDC2 and cDC1 metabolism

As the tumor glycocode crucially revamped DC metabolic patterns, we further examined the impact of tumor-derived glycans on metabolic features of the three DC subsets. DCs from HDs were incubated with NeoGlycoproteins (NeoGPs), consisting of a BSA backbone functionalized with multiple identical glycan motifs: Galactose (Gal), Mannose (Man), N-Acetyl Galactosamine (GalNAc), Sialyl-Tn antigen (s-Tn), Fucose (Fuc), and N-Acetyl Glucosamine (GlcNAc). Afterwards, the metabolism of DC subsets was evaluated with the SCENITH method (Supplementary Fig. 14A). In absence of stimulation, s-Tn, Fuc and/or GlcNAc NeoGPs triggered a decrease in the global metabolism of cDC2s and cDC1s (Supplementary Fig. 14B), whereas pDCs remained unaffected by these glycans (Supplementary Fig. 14B). cDC2s also experienced an increase in mitochondrial dependency in the presence of Man and Fuc (Supplementary Fig. 14C). On the contrary, no changes in metabolic profiles could be noticed for cDC1s and pDCs (Supplementary Fig. 14C). This phenomenon could be due to a decrease in all metabolic pathways in the same proportions, decreasing the global metabolism while keeping the same balance between glycolysis and OXPHOS pathways. Under TLR stimulation, cDC2s and cDC1s were even more affected by s-Tn, Fuc and GlcNAc, as we observed a strong drop in their global metabolism (Supplementary Fig. 14D). Notably, Fuc, which we found to drastically inhibit cDC2 function¹⁸ drove a sharp increase in the mitochondrial dependency of cDC2s, and GlcNAc increased their glucose dependency (Supplementary Fig. 14E). No metabolic switches could be observed for cDC1s, and pDCs that remained completely unaffected by the exposure to glycans (Supplementary Fig. 14D and 14E). These data unveiled that, among the glycocode, sTn, GlcNAc and Fuc motifs strongly reshaped cDC2s metabolism and affected cDC1s global metabolism.

Modulation of mTOR/AMPK-dependent metabolic pathways and/or MCT1 transporter rescue cDC2s and cDC1s from skewing by tumor-derived glycans Sialyl-Tn antigen and Fucose

We previously established that glycans altered DC cytokine/chemokine secretion profile¹⁸, and further demonstrated here that they also modified DC metabolic features. The next challenge consisted in investigating whether the metabolic and functional skewing of cDCs induced by tumor-derived glycans could be reverted by acting on different components of metabolic pathways. Reversion of metabolic disturbances can be achieved by acting on key signaling pathways regulating metabolic networks (mTOR/AMPK), on the availability of nutrients (nutrient deprivation/overdose), or on regulating nutrient transporters. We assessed all these tracks by using Oligomycin (Oli), an inhibitor of ATP synthase (mitochondrial metabolism); Rapamycin (Rap), an inhibitor of mTOR; Metformin (Met) and Dorsomorphin (Dor), respectively activator and inhibitor of AMPK; BAY-8002 (Bay), an inhibitor of the lactate transporter MCT1; and succinate (Suc), an input component of the Krebs cycle. We used s-Tn and Fuc as candidate glycans, as they were the only studied glycans to strongly affect both the metabolism and the function of cDCs¹⁸.

Heat map illustration –representing metabolic parameters of DC subsets upon incubation with BSA (control), s-Tn or Fuc motifs in presence of the different metabolic regulators– highlighted the significant modulations triggered by these metabolic regulators (Fig. 8A). As previously observed (Fig. 7), sTn and Fuc decreased the global metabolism of cDC2s and cDC1s (Fig. 8B, left, white boxes). Global metabolism of cDC2s was downregulated by Oli and Rap, whatever the presence or not of glycans, while this decrease was seen only in absence of glycans under Suc (Fig. 8B, left). As expected for an OXPHOS inhibitor, Oli strongly increased cDC2 glucose dependency and dampened mitochondrial metabolism regardless of the glycan (Fig. 8B, middle and right). Mitochondrial dependency of cDC2s is upregulated by Dor without glycan, and by BAY-8002 in the presence of Fuc (Fig. 8B).

**Fig. 8: Modulation of mTOR/AMPK-dependent metabolic pathways and/or MCT1 transporter rescues cDCs from skewing by the tumor-derived glycans Sialyl-Tn antigen and Fucose.**

Conventional DC1s underwent similar changes with Oli and Rap inhibiting their global metabolism in the presence of glycans, whereas only Rap and BAY-8002 have this effect in absence of glycans (Fig. 8B). As observed for cDC2s, cDC1 glycolytic metabolism was highly upregulated by Oli with or without glycans, while Rap slightly increased it when exposed only to s-Tn. Moreover, cDC1 mitochondrial dependency was decreased by Oli in the presence of Fuc (Fig. 8B).

Plasmacytoid DCs –although not directly metabolically and functionally impacted by glycans– were sensitive to metabolic modulators in a similar way than cDCs probably through cell crosstalk (Supplementary Fig. 15A and 15B). Thus, the metabolic regulators used provoked changes in global metabolism and metabolic switches in DCs, without restoring the level of metabolism of cDCs in absence of glycans.

Moreover, MCT1 inhibition modulated the expression by DC subsets of C-type lectin receptors (CLRs) potentially involved in the sensing of tumor-derived glycans (Supplementary Fig. 15C). It triggered a downregulation of DCIR, Dectin1, DC-SIGN, Langerin and/or CD206 as well as an increase of Dectin2 by cDC1s and cDC2s. These observations suggest a link between glycan-stimulated signaling and metabolic remodeling.

Knowing that glycans modify the production of cytokines and chemokines by DCs¹⁸, we tested the impact of the previously used metabolic regulators on the secretion of these factors that are crucial for the orientation of immune responses. The heat map—displaying an overview of the secretion of cytokines and chemokines under the different conditions– unveiled critical positive or negative influences of the metabolic regulators on DC secretome (Fig. 8C).

In the absence of glycans (condition BSA), we observed that inhibition of AMPK by Dor or ATP synthase by Oli strongly dampened secretion of antitumor factors by DCs while Dor increased the pro-tumor mediator MDC. In contrast, inhibition of the lactate transporter MCT1 by BAY enhanced IFN-β and IL-12 levels while dampening MIG, IL-10 and TARC production. Succinate increased antitumor mediators, such as IFN-α, IFN-β, IFN-λ1 and TNF-α, but also IL-10 (Fig. 8C). Such observations demonstrate that metabolic regulators are crucially able to dictate DC functional orientation.

As expected, sTn and Fuc glycans inhibited IL-12 and TNF-α productions while enhanced TARC secretion compared to control BSA (Fig. 8D, white bars). BAY-8002 completely reverted the perturbations triggered by s-Tn and Fuc, as demonstrated by the strong boosting in IL-12 and TNF-α productions concomitant to the annihilation of TARC secretion (Fig. 8D). It is noteworthy that BAY-8002 was the only metabolic regulator amongst those tested that was able to totally reverse DC dysfunction triggered by glycans. The inflammatory chemokine MIP-1β was not impacted by glycans. However, interestingly, despite being inhibited by Dor, Oli and Rap, it was positively stimulated by BAY-8002 (Fig. 8D).

Taken together, these findings demonstrate that the metabolic and functional skewing of DCs induced by tumor-derived glycans s-Tn and Fuc could be rescued by blocking the lactate transporter MCT1, without significantly altering DC metabolism. It also unveiled that metabolic regulators are crucially able to dictate DC functional orientation through remodeling of DC metabolism and function.

Lactate regulates the metabolic and functional orientations of DCs

To better elucidate the mechanism underlying the regulation of lactate metabolism in DCs by tumor-derived glycans, we investigated whether the modulation of lactate metabolism through lactate deshydrogense (LDH) inhibition or the blockade of lactate transport through MCT1 inhibition would affect glycan-dependent production of metabolites and DC function. We measured glucose and lactate levels, as well as cytokine/chemokine production in the culture supernatants of DCs pre-treated with the LDH-A inhibitor FX11 [inhibitor of the conversion of pyruvate into lactate], LDH-B inhibitor AXKO [inhibitor of the conversion of lactate into pyruvate] or BAY8002 [MCT1 inhibitor], further incubated with NeoGPs functionalized with s-Tn or αFuc motifs, and stimulated or not with TLR-L (Supplementary Fig. 16). In the setting used, s-Tn and Fuc did not directly modulate glucose and lactate consumption/production by DCs (Supplementary Fig. 16A, white bars). At basal state, MCT1 inhibition decreased lactate production by s-Tn-treated DCs, while LDH-A inhibition tended to decrease lactate production specifically by Fuc-treated DCs (Supplementary Fig. 16A left panels). This suggests that glycans can influence lactate metabolism or transport. Upon TLR stimulation, all inhibitors dampened lactate production independently of the glycans (Supplementary Fig. 16A right panels). We observed distinct modulation of DC function when blocking lactate transport or lactate metabolism. This was illustrated by dampened secretions of RANTES, MIP-1α and eotaxin specifically with LDH-A/B inhibitors, and the boosting of IL-18 production by sTn/Fuc treated DCs following LDH-B inhibition (Supplementary Fig. 16B). This shows a connection between tumor-associated glycans, lactate metabolism and DC functional profile.

To gain further insights into the role of lactate in regulating DC metabolism and function, we treated DCs with rising concentration of lactate and assessed the impact on DC metabolic and functional profiles (Fig. 9, Supplementary Fig. 17). Lactate dampened the global metabolism of cDC2s, cDC1s and pDCs (Fig. 9A) and triggered increased mitochondrial dependency associated with decreased glycolytic capacity of cDC2s and pDCs (Fig. 9B). Lactate also profoundly affected DC function by dampening immune and inflammatory mediators such as IFN-α, IFN-β, IL-12, IL-29/IFN-λ1, IL-18, IP-10, I-TAC, MIG, MCP-1, TNF-α, MIP-1β, and eotaxin-3 (Fig. 9C, Supplementary Fig. 17) while promoting regulatory factors such as IL-10 and IL-23 (Fig. 9D). Altogether, these observations demonstrate that lactate directly drives metabolic switches and dampens DC function. This supports a connection between the tumor-derived glycans s-Tn and Fuc, lactate metabolism and DC functional profile.

**Fig. 9: Lactate directly modulates DC metabolism and dampens DC function.**

The blockade of the MCT1 transporter influences cDC-driven naïve T-cell functional and metabolic orientation

To decipher the global immune impact of DC metabolic disturbances in melanoma, we further explored the consequences of glycan-driven perturbations of DC metabolism on the subsequent metabolic and functional orientation of T-cell responses. We thus cocultured purified FACS-sorted cDC2s or cDC1s pre-incubated with NeoGPs functionalized with s-Tn or αFuc motifs with naïve CD4 or CD8 T cells purified from human cord blood for 5 days and assessed the phenotypic, functional and metabolic modulations of T cells using the SCENITH method combined to multiparametric flow cytometry. Interestingly, Fuc-subverted cDC2s dampened global protein synthesis in CD4 T cells, and tended to increase glucose dependency of CD8 T cells while reducing their FAO&AAO capacity (Supplementary Fig. 18). This highlights an impact of glycan-subverted cDC2s on the metabolic fitness of T cells.

To ultimately explore whether lactate modulation would affect glycan-subverted cDC-driven naïve T-cell functional and metabolic orientation, we additionally blocked MCT1 on cDCs using BAY-8002 before performing the cocultures of purified FACS-sorted cDC2s or cDC1s pre-incubated with s-Tn or Fuc glycans with naïve CD4 or CD8 T cells. The phenotypic (i.e., activation status and immune checkpoint profile) and functional orientation (appreciated by transcription factor expression and cytokine secretion) of T cells, as well as their metabolic modulations were then assessed. The SCENITH method combined with surface and intranuclear staining was used to determine the activation status (CD69, CD25), the immune checkpoint profile (GITR, TIGIT, PD1, OX40, TIM3, 41BB, CTLA4, LAG3, ICOS) and transcription factor expression (T-bet [Th1], GATA-3 [Th2], PU.1 [Th9], RORγt [Th17], AhR [Th22], FoxP3 [Treg], Bcl6 [TFH]) (Fig. 10, Supplementary Fig. 19, Supplementary Fig. 20). T-cell cytokine secretion was assessed by Luminex (Fig. 10, Supplementary Fig. 19, Supplementary Fig. 20). MCT1 inhibition triggered deep changes in the activation status, immune checkpoint profiles and transcription factor expression by naïve T cells. Specifically upon coculture with sTn- and/or Fuc-treated cDC2s, we observed an increased expression of GITR and OX40 by CD4 T cells and a decreased expression of CD25 and PD1 by CD8 T cells (Fig. 10A, Supplementary Fig. 19A). Specifically upon coculture with sTn- and/or Fuc-treated cDC1s, we observed an increased expression of GITR, OX40 and 41BB and a decreased expression of TIM3 and PD1 by CD4 T cells (Supplementary Fig. 20A). Some changes driven by MCT1 inhibition in cDCs occurred independently of glycans, such as increased 41BB and decreased CD25, TIM3, LAG3, ICOS levels on CD4 T cells, and increased GITR and decreased TIM3, LAG3, ICOS levels on CD8 T cells (Fig. 10A, Supplementary Fig. 19A, Supplementary Fig. 20A). Concerning transcription factor expression, MCT1 inhibition promoted FoxP3 in CD4 T cells and Bcl-6 in CD4 and CD8 T cells (Fig. 10B, Supplementary Fig. 19B). This inhibition tended to support T-bet in CD8 T cells, while dampened GATA-3 and AhR in CD8 T cells upon culture with cDC2s (Fig. 10B, Supplementary Fig. 19B), while it decreased GATA3 and RORγ expression and increased Bcl6 upon culture with cDC1s (Supplementary Fig. 20B). Furthermore, MCT1 inhibition strongly enhanced global protein synthesis in CD4 T cells following coculture with Fuc-subverted cDC2s, and dampened glucose dependency while promoted FAO&AAO capacity of CD4 T cells (Fig. 10C) without affecting the metabolism of cDC2-driven CD8 T cells (Supplementary Fig. 19C) or cDC1-driven CD4 T cells (Supplementary Fig. 20C). Moreover, MCT1 inhibition affected the cytokine secretion profile of cDC2- and cDC1-driven CD4 and CD8 T cells also specifically in the presence of glycans. Thus, T cells cocultured with sTn- and Fuc-conditioned cDC2s demonstrated an increase of IL-1β and a strong decrease of IL-10 upon BAY compared to control conditions (Fig. 10D). T cells cocultured with sTn- and/or Fuc-conditioned cDC1s showed an increase of IL-21 and a decrease of IL-5 and IL-10 (Supplementary Fig. 20D), while coculture with Fuc-conditioned cDC2s tended to decrease TGF-β (Fig. 10E) upon BAY compared to control conditions. This impact on cytokine secretions was shown both at basal state (Fig. 10D, Supplementary Fig. 19D, Supplementary Fig. 20D) and upon restimulation of T cells (Supplementary Fig. 19E, Supplementary Fig. 20E).

**Fig. 10: Blockade of the MCT1 transporter influences DC-driven naïve T-cell functional and metabolic orientation.**

Altogether, these findings demonstrate that the blockade of the MCT1 transporter influences cDC-driven naïve T-cell functional and metabolic orientation. Inhibition of MCT1 triggers a strong reshaping of T-cell activation status, immune checkpoint profile, functional orientation and metabolic fitness. Importantly, MCT1 inhibition display effects on T cells specifically upon coculture with glycan-subverted DCs but not with control BSA-treated DCs. Lactate metabolism is therefore crucially linked with glycan-driven disturbances in DCs, and subsequently profoundly affect DC-mediated crosstalk with effector T cells.

Discussion

By investigating metabolic profiles of circulating and tumor-infiltrating DC subsets in melanoma, we identify major disturbances in the metabolism of the three main human DC subsets (i.e., cDC1s, cDC2s, and pDCs) and related DC subsets that share some critical markers (i.e., CD5^+/– DC2s, CD14^+/– DC3s and AS DCs). In comparison to circulating DCs of HDs, an increase of protein synthesis that reflects global metabolism is measured in blood and tumor-infiltrating DCs of melanoma patients. Some of these disturbances are linked with DC phenotypic alteration and dysfunction (including their capacity to activate and polarize naïve T cells), as well as affect patient clinical outcome. For instance, melanoma-infiltrating cDC1s and pDCs exhibiting a high global metabolism after TLR stimulation are associated with a bad OS. Moreover, analysis of blood pDCs from melanoma patients could also predict patient outcome, since TLR-L-stimulated circulating pDCs with a high global metabolism are associated with a better PFS, whereas a high glycolytic metabolism was linked with shorter OS. This highlights that metabolic reprogramming could trigger DC subversion in melanoma. We further demonstrate that metabolic interventions (e.g., modulation of the mTOR/AMPK pathways) can both rescue DCs from altered metabolic reprogramming and restore their functions (see above). This opens the way to target DC metabolism to reestablish DC function and antitumor responses.

Currently, several methodologies are available to study cell metabolism⁴⁰. Mass Spectrometry identifies and quantifies metabolites produced by the different metabolic pathways. Analysis using the Seahorse metabolic flux assay determines cell capacities and dependencies indirectly through the measurement of oxygen consumption rate and extracellular acidification rate of live cells under the action of metabolic inhibitors⁴¹. Mass cytometry or cytometry by time of flight (CyTOF®) is a technique similar to flow cytometry using heavy metal-conjugated antibodies and based on the measurement of metabolic enzymes or transporters at single cell level. The SCENITH method reveals global metabolic functions in link with immune phenotypes and functions at a single cell resolution on multiple cell subsets in parallel³⁴. Its main advantages are the single-cell analysis, the possibility to analyze large cytometry panels including immune checkpoints, activation markers or intracellular cytokines. This SCENITH method is a potent way to explore the functional metabolic heterogeneity at single cell level. Using this method, we show a strong positive relationship between the levels of IFN-β and IFN-λ1 with a high global metabolism and a mitochondrial dependency of pDCs. MDC levels are negatively correlated with the glucose dependency of cDC1s.

We highlight that human DC subsets display distinct metabolic profiles, cDC1s and cDC2s being highly glucose-dependent, whereas pDCs exhibiting both glycolytic and FAO/AAO capacities. This is in line with the study by Arguello et al. ³⁴. They observe that in human, pDCs were dependent on glucose and mitochondrial respiration, whereas cDC1s and cDC2s displayed high glycolytic capacity together with moderate OXPHOS capacity. These findings emphasize that distinct metabolic programs shape the functional specialization of different DC subsets²⁶. We also extend this observation to newly characterized human DC subsets, which share gating markers with pDCs (i.e., AS DC) or cDC2s (i.e., CD5^+/– DC2s and CD14^+/– DC3s)^{35,36,37,38,39}. Under resting conditions, mouse cDC1s retain higher levels of both glycolysis and mitochondrial metabolism than cDC2s, and inhibition of one of these pathways impedes cDC1-dependent priming of CD8⁺ T cells⁴². In murine cDCs, glycolytic reprogramming is mediated by HIF1α whereas it is driven by inducible nitric oxide synthase in GM-CSF-derived BM-DCs²⁶. Here, tumor-associated glycans (i.e., Fuc. or s-Tn antigen) alter cDC metabolism.

We highlighted a strong elevation of global metabolism of DCs in melanoma patients compared to HDs, for all the three main DC subsets (i.e., pDCs, cDC1s and cDC2s), in both blood and tumor. The in situ activation of DC subsets previously observed in the context of melanoma¹⁰ combined with the required metabolic adaptation to meet the energy demand associated with phenotypic, morphologic and functional changes that are required to elicit antitumor immune responses could explain such phenomenon. An increase of the global metabolism was also observed for all these DC subsets under TLR stimulation, which is consistent with the study of Arguello et al. ³⁴. Yet, the function of different DC subsets may rely on distinct metabolic programs. Indeed, cDC1s and pDCs are able to increase their ability to use OXPHOS in response to specific TLR stimulation, demonstrating a shift in the metabolic profile of these DCs upon TLR stimulation. Metabolic adaptation of human blood-derived cDC1s, cDC2s and pDCs upon TLR triggering has been deciphered using the Seahorse metabolic flux assay, RNA sequencing and nuclear magnetic resonance (NMR)-based analysis of metabolites^43,44. TLR stimulation increases glutaminolysis (which can fuel OXPHOS) and the OXPHOS pathway in human pDCs⁴⁴. In mouse pDCs, TLR-induced OXPHOS occurs through type I IFN and is necessary for pDC activation and subsequent T-cell priming⁴⁵. Glycolysis plays also a role in type I IFN production and costimulatory molecule upregulation by both human and mouse pDCs in response to TLR activation^45,46. Kinetics and type I IFN dependency may differ between mouse and human pDCs regarding their metabolic pathways induced by TLR activation as discussed in Saas et al. ⁴⁷. TLR stimulation of human CD1c⁺ cDC2s promotes OXPHOS activity and induces glycolysis⁴⁴. Regarding cDC1s, in contrast to our results and the study by Du et al. ⁴², others report that TLR stimulation of human cDC1s reduced the expression of genes regulating glycolysis while supporting OXPHOS⁴³. These discrepancies may come from the source of DCs (species origin, natural or in vitro differentiated), the time of analysis, or methods used to depict metabolic profiles (analysis of enzymes, metabolites or activities of the metabolic pathways). Nevertheless, all studies show that, changes in metabolism are required for DC activation even though each DC subset uses different metabolic pathways (mostly glycolysis in cDCs and OXPHOS in pDCs). Melanoma, by affecting DC metabolism, may alter their activation.

In cancer, recent studies highlight the pivotal role of metabolic pathways to regulate DC activation and functions, especially to trigger proper antitumor immunity^26,31,33. Indeed, glycolysis drives STING signaling in DCs, contributing to DC-mediated antitumor immune responses as demonstrated with murine DCs and tumor-infiltrating DCs from patients with non-small cell lung cancer⁴⁸. Yet, immunogenicity and tolerance of DCs are crucially determined by orientation of the metabolic pathways that govern their functions^28,29. Regarding melanoma, studies have shown that TME creates metabolic defects leading to T cell dysfunction⁴⁹, whereas the impact of metabolic disturbances on antigen-presenting cells remains unexplored. We demonstrate here that metabolic reprogramming of DC subsets within the melanoma microenvironment could participate in immune subversion and tumor escape. We uncover how melanoma perturbed metabolic pathways of circulating and tumor-infiltrating DC subsets to subsequently hijack their function. It appears that in both blood and tumor, protein synthesis in DCs increased as compared to HD cells not exposed to a subversive TME. Such changes were associated with drastic metabolic switches in both circulating and tumor-infiltrating DC subsets. Tumor-derived pDCs increased both glycolytic and FAO/AAO capacities while circulating and tumor-infiltrating cDC1s exhibited dampened glycolytic capacity and enhanced oxidative metabolism. These results are consistent with previous reports in different cancer types for human cDC1s³⁴. However, unlike here in melanoma, these authors find metabolic alterations in human cDC2s in several other types of cancers³⁴. This could be explained by the specific metabolic disturbances according to the tumor type and/or the anatomical origin of metastasis. Based on available studies, metabolic profiles of DC subsets in tumors seem heterogeneous. A high glycolytic capacity of human cDC1s and cDC2s has been pointed out in kidney tumors, whereas a high respiratory metabolism profile has been described in human brain tumors³⁴, suggesting that the anatomical origin of tumor could influence the metabolism of infiltrating immune cell subsets.

Dysregulation of DC metabolic pathways in cancer, including the proper induction of glycolysis within TME could rely on several mechanisms. As conventional DCs are critically dependent on glycolysis to fulfil their functions, glucose competition within TME strongly participates to inhibition of DC activation and function, limiting their capacity to trigger appropriate antitumor responses. Lactic acid/lactate, a byproduct of glycolysis, can skew human monocyte-derived DC differentiation into a tolerogenic state (i.e., IL-10 production after TLR stimulation)⁵⁰. Besides, DAMP signals/alarmins may be insufficient to promote glycolysis in tumor-associated DCs (TADCs) and may affect mTOR/AMPK pathways. Alternatively, chronic exposure to DAMPs may lead to a state of glycolysis-induced exhaustion within DCs. IL-37 inhibits cancer immune surveillance by affecting the AMPK/Akt signaling axis resulting in inhibition of glycolysis in mouse CD103⁺ DCs, which dampens their antitumor function⁵¹. AMPK activation has been shown to prevent the ability of human DCs to elicit antigen-specific CD8⁺ T cell activation⁵² and to drive elevated OXPHOS, which has been associated with tolerogenic features of human TADCs⁵³. Modulation of these mTOR/AMPK pathways restores inflammatory cytokine (e.g., IL-12 and TNF-α) production previously inhibited by tumor-associated glycans -Fuc and s-Tn antigen.

We also unveiled the critical interplay between metabolic pathways and DC features and functions. We underlined that DC subsets display distinct metabolic profiles depending on their activation status and immune checkpoint profile, supporting critical connections between DC features and metabolic pattern. The metabolism of DC subsets was also connected with their functional polarization, as illustrated by correlations between MDC secretion levels and metabolic parameters. Thus, metabolism appears to dictate DC phenotypic and functional features.

Our work further reveals mechanisms of glycan-induced DC dysfunction. Glycan motifs expressed by tumor cells could directly abrogate the glycolytic activity of human monocyte-derived DCs (moDCs)⁵⁴. Indeed, induction of Treg cells by moDCs upon O-glycan Tn antigen/MGL interactions is the consequence of a reduced glycolytic activity in DCs due to the modulated expression of genes coding for enzymes involved in glycolysis and OXPHOS pathways⁵⁴. Furthermore, tumor-derived α-fetoprotein (AFP) - a protein secreted by hepatocellular tumor cells and associated with poor outcome in patients with hepatocellular carcinoma (HCC) - promotes immune suppression by skewing human moDC metabolism toward glycolysis⁵⁵. In addition, AFP suppresses OXPHOS in human moDCs and this may explain CD1c⁺ cDC2 dysfunction in HCC⁵⁶. This suggests that tumor cells can directly affect DC metabolism, as observed in our study with melanoma tumor cell lines. The switch to OXPHOS could illustrate an adaptation of DCs to the glucose depletion observed in TME. However, this switch may induce cell dysfunction as glycolysis is crucial for cell proliferation and cytokine production²³. This is consistent with the observation that a high OXPHOS profile within circulating and tumor-infiltrating T cells from melanoma patients is predictive of resistance to immunotherapy⁵⁷. Interestingly, MCT-1 inhibition modulated the expression by DC subsets of C-type lectin receptors (CLRs) potentially involved in the sensing of tumor-derived glycans. This suggests a link between glycan-stimulated signaling and metabolic remodeling. CLRs represents potential targets of upstream glycan-stimulated signaling, able to prevent metabolic remodeling in DCs.

The link between metabolism and immune function supports that targeting the metabolism emerges as a potent mean to regulate DC function and control immune responses⁵⁸. Therapeutic metabolic reprogramming of TADCs could restore immune cell function and appropriate anticancer responses²². The PI3K/AKT and mTOR pathways are responsible for glycolysis induction and are counterbalanced by the AMPK signaling pathway that drives OXPHOS. These pathways represent appropriate targets. Indeed, mTOR and AMPK/PI3K are key regulators of metabolic reprogramming, thus of immune responses, especially in DCs^59,60. Hence, mTOR-deficient mouse cDC1 have reduced ability to prime CD8⁺ T cell proliferation in vitro⁴², while AMPK activation by promoting OXPHOS, blocks mouse DC maturation²⁷. Reversion of metabolic disturbances can be achieved by acting on these key signaling pathways, but also on the availability of nutrients (nutrient deprivation/overdose), or by regulating nutrient transporters. We demonstrate here that blocking the MCT1 lactate transporter (using BAY-8002) can rescue cDCs from glycan-induced DC functional hijacking, and further restore anti-tumor T-cell fitness. Notably, Adamik et al. recently demonstrated that MCT1 expression increased on monocyte-derived DCs from melanoma patients, and could be crucial to regulate their metabolism and their immunostimulatory potential upon vaccination⁶¹. Current strategies of metabolic reprogramming include ex vivo manipulation of DC metabolism before therapeutic vaccination or in situ targeting of metabolic pathways to reprogram TADCs. Inhibition of mTOR by rapamycin in cancer vaccines –either by treating DCs ex vivo before injection⁶² or by direct in vivo administration⁶³– improves CD8⁺ T cell activation, and leads to better antitumor efficacy. Moreover, pharmacological inhibition of fatty acid synthesis restores TADC-dependent antitumor immunity in mice⁶⁴. Pathogen-recognition receptor agonists as well as nanoparticles-based approaches allowing specific DC targeting may also affect TADC metabolism and improve their function³¹. Thus, inducing DC metabolic reprogramming gives TADC a competitive advantage over tumor cells. In contrast, limiting glucose availability during the production of DC vaccines helps sustaining costimulatory molecule expression and IL-12 production⁶⁵. Supporting OXPHOS and limiting nitric oxide levels improves DC survival and function⁶⁶. In addition, mouse cDC1s lacking the Hippo pathway kinases Mst1 and Mst2 have deregulated mitochondrial dynamics. Thus, manipulation of the mitochondrial dynamic in TADCs - using mitochondrial fission inhibitors or fusion activators - improves mouse cDC1-dependent tumor antigen-specific CD8⁺ T-cell activation and remains a promising perspective⁴². Altogether, these results support the development of new therapeutic drugs to restore the metabolism of DCs infiltrating melanoma tumors. Controlling immune subversion at the metabolic level could represent a complementary therapeutic approach in addition to current therapies targeting melanoma signaling pathways (e.g., BRAF-targeted agents) or effector immune cell dysfunctions (e.g., immune checkpoint blockers)⁶⁷. Combined therapies associating also the metabolic reprogramming of immune cells should improve the success of targeted and immune therapies. Several studies in mouse models have already demonstrated that targeting tumor metabolism restores a TME favorable for immune cell function, fostering immune responses⁶⁸ and sensitizing tumors to immune checkpoint inhibitors⁶⁹. Thus, metabolic checkpoints emerge as promising targets to restore antitumor immune responses and strengthen the success of targeted and immunotherapies.

Altogether, our work reveals that melanoma evades the immune system partly by reprogramming DC energetic metabolism, potentially through glycan motifs. Further insights into the metabolic regulation of DC subset function and dysregulation in cancer would allow a better understanding of immune subversion by tumors. This paves the way for appropriate tuning of metabolic pathways to rescue DC from tumor hijacking, in order to restore potent antitumor responses, and enhance the efficacy of current targeted and immunotherapies. As metabolism crucially dictates DC functional orientation, integration of metabolic regulation into the optimized development of DC vaccines and cancer immunotherapies holds great promise to achieve better clinical success.

Methods

Melanoma patients and healthy donor (HD) samples

All procedures were approved by the Ethics committee of Grenoble University Hospital (CHUGA) and the French Blood Agency’s Institutional Review Board Committee (IRB), and accredited by the Ministry of Education and Research under the reference #AC-2020–3959 (EFS) and collection #AC-2014-2094 (CHUGA). Mandatory written informed consent was obtained from all donors prior to their participation in this study (including indirect identifiers such as sex and gender) and samples were analyzed anonymously. Blood samples were obtained from healthy donors (HD, n = 87) and stage I-IV melanoma patients (n = 19). Lymph node or cutaneous metastatic tumors (n = 44) were obtained from melanoma patients. Clinical features of melanoma patients are stated in Supplementary Table 1 and Table 2.

PBMC isolation and PanDC purification

Peripheral blood mononuclear cells (PBMC) were obtained by Ficoll density gradient separation (Eurobio) on the blood of HD and melanoma patients. Cells were frozen and stored at –150 °C until use. For some experiments, negative selection was performed on thawed PBMC to sort PanDCs (mixture of cDC1s, cDC2s and pDCs), using the EasySep PanDC pre-enrichment Kit (StemCell).

Tumor-infiltrating cell isolation and generation of primary tumor cells and tumor-conditioned medium

Tumor samples were cut into small fragments using a sterile scalpel and enzymatically digested with collagenase D (final concentration 2.5 mg/ml) and DNase I (final concentration 100 μg/ml) during 30 min at 37 °C with 5% of CO₂ followed by mechanical disruption. Cell suspensions were obtained by filtering the mixture through a 50μm sieve with RPMI 10% FBS, frozen and stored at –150 °C before use. After thawing, tumor-infiltrating cells were first cleared of tumor cells by an adhesion step in a 75 cm² flask and further separated from residual tumor cells using CD45⁺ positive selection isolation kit (StemCell) according to the manufacturer instructions. Adherent cells (corresponding to primary melanoma tumor cells) were maintained in culture to generate primary tumor lines and cultured in RPMI 1640 GLUTAMAX I supplemented with gentamicin (20 µg/mL), non-essential amino acids (MEM 1X) (Invitrogen), sodium pyruvate (1 mM) (Sigma) (complete RPMI) and 10% heat-inactivated FCS in a humidified incubator maintained at 37 °C with 5% CO₂ atmosphere. Tumor supernatants were collected on tumor cell cultures reaching 80–90% confluence, having changed the medium 24 h prior, and stored at –80 °C.

Glycoprofiling of melanoma tumor cells

The profiles of glycans expressed by melanoma tumor cells were assessed by performing the GLYcoPROFILE™ with the LEctPROFILE® plates from GLYcoDiag (Orléans, France)¹⁶. Briefly, when reaching 80–90% confluence in 75 cm² culture flask, tumor cells were harvested with a Trypsin/EDTA solution, washed, suspended in PBS and labeled with carboxyfluorescein diacetate succinimidylester (CFDA-SE, Sigma-Aldrich, St. Louis, MO, USA) in PBS. Next, 100 μL of labeled cells (2 × 105 cells) were added in each well of the LEctPROFILE® plates (containing 16 lectins with distinct glycan specificity) and incubated 2 h at room temperature under gentle agitation. After washing with PBS, fluorescence intensity was measured using a microplate reader (λex = 485 nm, λem = 530 nm, Fluostar OPTIMA, BMG LABTECH, France). In parallel, a calibration curve was achieved with the labeled cells solution to determine the number of cells stayed in interactions with lectins.

Coculture of PanDCs with tumor cells, tumor-conditioned medium or deglycosylated tumor cells

Primary tumor cells from melanoma patients were plated and cultured in 48-well plates until cell confluence was reached. Purified human panDCs were co-cultured with or without confluent tumor cells or in 50% tumor-conditioned medium at 1.10⁶/ml in 48-well flat bottom plates for 20 h at 37 °C with 5% CO₂. In some experiments, sialylated and fucosylated motifs or all N-linked and O-linked oligosaccharides motifs were removed from tumor cells prior to addition of PanDCs using either the α(2 → 3,6,8,9) Neuraminidase (sialidase) and the α-1 → (2,3,4)-Fucosidase (SIGMA), or the Glycoprotein deglycosylation kit (SIGMA) [containing the endo-α-N-Acetylgalactosaminidase, β1,4-Galactosidase, Glucosaminidase, N-Glycosidase F, Neuraminidase], respectively. Briefly, adherent tumor cells were incubated with the glycosidases for 1 h at 37 °C 5%CO₂ according to manufacturer’s instructions, washed 3 times with RPMI 10% FCS before adding PanDCs.

Coculture of PanDCs with lactate

Purified human panDCs were co-cultured with rising concentrations of sodium L-lactate (ThermoFisher) at 10, 25 and 50 mM for 45 min at 37 °C with 5% CO₂ in complete RPMI 10% FCS. Metabolic, phenotypic and functional assays were then performed as mentioned below.

PanDC exposure to soluble glycans and inhibition of metabolic pathways

Purified panDC placed in 48-well plates (at 1.10⁶ cells per mL) were first incubated with soluble glycans (e.g., Neoglycoproteins (NeoGPs)) functionalized with αGal (NeoGa), αMan (NeoM), αGalNAc (NeoGaN), s-Tn (NeoSTn), αFuc (NeoF) and βGlcNAc (NeoGN) motifs (GlycoDiag, Orleans) at a concentration of 40 µg/mL in complete RPMI for 1 h, and further cultured for 20 h at 37 °C with 5% CO₂ in complete RPMI complemented with 10% FBS. In some experiments, purified PanDCs were pre-incubated for 1 h at 37 °C with the following metabolic inhibitors before addition of the NeoGPs: Oligomycin [1 µM, OXPHOS inhibitor], Rapamycin [100 nM, mTOR inhibitor], Metformin [100 nM, AMPK activator], Dorsomorphin [3.75 µM, AMPK inhibitor], BAY-8002 [80 µM, lactate transporter MCT1 inhibitor], Succinate [200 µM, intermediate of Krebs cycle]. Lactate metabolism was also inhibited by pre-incubating PanDCs for 1 h with the lactate dehydrogenase (LDH) inhibitors FX11 [20 µM, LDH-A inhibitor] and AXKO [10 µM, LDH-B inhibitor] (MedChemExpress). LDH-A has a higher affinity for pyruvate than lactate, conducing the conversion of pyruvate into lactate, while LDH-B has a higher affinity for lactate than pyruvate, therefore driving the conversion of lactate into pyruvate. The phenotype, metabolism and functions of PanDCs were then analyzed as described below.

Cell activation

After 20 h of DC co-culture with or without tumor cells or NeoGPs, TLR stimulation was performed with including polyinosinic-polycytidylic acid (PolyI:C, 30 μg/mL, Invivogen), Resiquimod (R848, 1 μg/mL, Invivogen), Class-A CpG oligonucleotide ODN-2336 (CpG_A, 1.5 μM, Invivogen), or the mixture of PolyI:C, R848 and CpG_A TLR-ligands for 4 h.

Metabolism profiling using the SCENITH method

Up to 2.10⁶ PBMC, CD45⁺-tumor infiltrating cells or up to 2.10⁵ purified PanDCs per well were seeded in 96 well plates in RPMI 10% FBS for studying cell metabolism. At basal state or after activation, cells were treated during 30 min with DMSO (Control), 2-Deoxy-D-Glucose (DG, final concentration 100 mM, Sigma), Oligomycin (Oligo, final concentration 1 mM, Ozyme), or a combination of the drugs at the final concentrations before mentioned. As inhibition positive control, the translation elongation inhibitor Cycloheximide was used (CHX, 20 μg/mL, Sigma). Puromycin (Puro, final concentration 10 mg/mL, Cayla) was added during 15 min after the metabolic inhibitors treatment. Cells were then washed in cold PBS, and stained with the fluorescent cell viability marker and primary conjugated antibodies against surface markers during 20 min at 4 °C in PBS. After washing, cells were fixed and permeabilized using FOXP3 fixation and permeabilization buffer (ThermoFisher) following manufacturer instructions. Intracellular staining of puromycin was performed by incubating cells during 30 min at 4 °C in permeabilization buffer with anti-Puro-AF488 monoclonal antibody (Merck). Percentages and absolute numbers of DC subsets analyzed from HDs, patient blood and patient tumors are displayed in Supplementary Table 3.

Assessment of activation status, immune checkpoint expression and CLR profile of DC subsets by flow cytometry

In parallel to puromycin labeling, cells were stained in order to define DC subsets [cDC1s, pDCs, CD5^+/– DC2s, CD14^+/– DC3s and AS DCs], their activation status and immune checkpoint expression. To depict cDC1s, pDCs and total cDC2s, the following antibodies were used: PerCP-Cy5.5-anti-CD11c (BD), PE-Cy7-anti-BDCA1 (BioLegend), APC-anti-BDCA2 (Miltenyi), APC-anti-BDCA3 (Miltenyi), APC-H7-anti-HLA-DR (BD), BV570-anti-CD45 (BioLegend), Live and Dead-PE-Texas Red (ThermoFisher) and BV510-coupled mixed antibodies against lineage markers including CD14 (BioLegend), as well as anti-CD40 (Beckman), -PDL1, -TIM3 antibodies (BD). Among PanDCs, defined as alive CD45⁺ Lin^- HLA-DR⁺ cells, cDC1s were depicted as CD11c^int BDCA3^high, pDCs defined as CD11c^neg BDCA2⁺ and total cDC2s as CD11c^high BDCA1⁺. To depict AS DCs, we additionally used PE-Cy7 anti-AXL, BV786 anti-SIGLEC6, PE anti-CD123, BV421 anti-CD16, and defined them as alive CD45⁺ Lin^– HLA-DR⁺ CD16^– BDCA3^– CD123⁺ CD11c^–/lo AXL⁺ SIGLEC6⁺. To define CD5^+/–DC2s and CD14^+/– DC3s, we used PE-Cy5-coupled mixed anti-CD3/19/20/56 (Lin without CD14), PE anti-CD88, PerCP-Cy5.5-anti-CD11c, PE-Cy7-anti-BDCA1, APC-anti-BDCA3, APC-H7-anti-HLA-DR, BV421 anti-CD163, BV510 anti-CD5, BV786 anti-CD14, BV570-anti-CD45, Live&Dead-PE-Texas Red. DC2s and DC3s were depicted as alive CD45⁺ Lin^– HLA-DR⁺ CD88^– BDCA1⁺ CD11c⁺ BDCA3^–, and further distinguished as CD163^– CD14^– CD5^+/– for DC2s and CD163⁺ CD14^+/– for DC3s. In some experiments, CLR expression profile by DC subsets was assessed using anti-DCIR, -Dectin1, -Dectin2, -DC-SIGN, -Langerin, and anti-CD206 antibodies. After washing, cells were finally resuspended in BD FACS lysing solution before analysis by flow cytometry. Data were acquired with BD LSRII and analyzed with BD FACS DIVA 9 software (version 9.0.1).

Impact of DC metabolism on naïve T-cell functional and metabolic orientation

Co-cultures of FACS-sorted cDCs with naïve CD4/CD8 T cells were performed during 5 days and the phenotypic, functional and metabolic modulations of T cells were then assessed. PanDCs were first enriched from PBMC derived from HDs using the EasySep PanDC pre-enrichment Kit (StemCell), further labeled with PE-Cy7 anti-BDCA1, APC anti-BDCA3, PerCP anti-HLA-DR and BV510 anti-Lin, and submitted to FACS sorting (BD Aria) to isolate cDC2s (Lin^– HLA-DR⁺ BDCA1⁺ BDCA3^–) and cDC1s (Lin^– HLA-DR⁺ BDCA1^– BDCA3⁺) (purity >99,8%). In parallel, naïve CD4 and CD8 T cells were purified from human cord blood using the Human naïve CD4 or CD8 T-cell enrichment kits (StemCell) (purity >97%). Sorted cDC2s or cDC1s (100.000/mL) were pre-incubated or not with BAY-8002 [80 µM, lactate transporter MCT1 inhibitor] for 1 h, followed by culture with NeoGPs functionalized with s-Tn or αFuc motifs (40 µg/mL, GlycoDiag) in complete RPMI for 1 h, complemented with 10% FBS for 20 h at 37 °C 5% CO₂, and then further co-cultured with naïve CD4 or CD8 T cells (1.10⁶/mL) in a 1:10 ratio for 5 days. The metabolism of T cells was then depicted with the SCENITH method as previously described. The activation status and immune checkpoint profile of T cells were analyzed by surface staining using PE-Cy5 anti-CD69, APC anti-CD25, PE anti-CD27, PE-Cy7 anti-GITR, AF700 anti-TIGIT, BV650 anti-PD1, FITC anti-OX40, PE anti-TIM3, PE-Cy5 anti-41BB, PE-Cy7 anti-CTLA4, APC anti-LAG3, BV650 anti-ICOS together with PE-TxRed Live&Dead, BV570 anti-CD45, APC-H7 anti-CD3 and BV786 anti-CD4 or CD8. Transcription factor expressions were analyzed by performing a surface staining of CD45, CD3 and CD4 or CD8 followed by an intranuclear staining using BV650 anti-T-bet (Th1), PE-Cy5 anti-GATA-3 (Th2), FITC anti-PU.1 (Th9), PE anti-RORγt (Th17), PE-Cy7 anti-AhR (Th22), AF700 anti-FoxP3 (Treg), and BV421 anti-Bcl6 (T_FH) transcription factors using the eBioscience™ Foxp3/Transcription Factor Staining Buffer Set. Samples were analyzed by flow cytometry using BD LSRII. T cells were also restimulated for 16 h with 50 ng/mL phorbol 12-myristate 13-acetate (PMA) and 1 mg/mL ionomycin (Sigma). Culture supernatants were collected at day 5 and after 16 h of restimulation to assess cytokine/chemokine secretion profiles by Luminex.

Quantification of secreted cytokine/chemokine by Luminex

Culture supernatants were collected and frozen at –20 °C until use. Cytokine/chemokine dosage was performed with the Human ProcartaPlex Mix&Match kit (ThermoFisher) and measured using the Luminex instrument MagPix200 (BioRad) with xPONENT software. For DC analysis, the concentrations of I-TAC (CXCL11), MIG (CXCL9), IFN-α, IP-10 (CXCL10), MCP-1 (CCL2), MIP-1α (CCL3), MIP-1β (CCL4), RANTES (CCL5), IL-10, IL-23, TNF-α, TARC (CCL17), MDC (CCL22), IL-29/IFN-λ1, IFN-β, IL-12p70, LAG3, eotaxin-1 (CCL11), eotaxin-3 (CCL26), IDO, and IL-18 were analyzed. For supernatants of DC/naïve T-cell co-cultures, we used the ProcartaPlex™ Human Th1/Th2/Th9/Th17 Cytokine Panel 18plex to assess the concentrations of GM-CSF, IFN-gamma, IL-1 beta, IL-2, IL-4, IL-5, IL-6, IL-12p70, IL-13, IL-18, TNF-alpha, IL-9, IL-10, IL-17A (CTLA-8), IL-21, IL-22, IL-23, IL-27. For tumor cell supernatant analysis, the concentrations of IL-1β, IL-6, IL-8, IL-10, MDC, MCP-1, MIP-1α, MIP-1β and TGF-β were analyzed. Data analysis was done with the ThermoFisher ProcartaPlex App.

Extracellular glucose and lactate concentration measurements

Glucose and lactate levels were analyzed in PanDCs culture supernatants and tumor supernatants using the StatStrip Xpress2 Glucose/Ketone Meter System and the StatStrip Xpress2 Lactate Meter System (NOVA BIOMEDICAL).

Glucose uptake by DC subsets

To measure glucose uptake by DC subsets, PBMCs or CD45⁺-tumor infiltrating cells, pre-stimulated or not with the mixture of PolyI:C (30 μg/mL), R848 (1 μg/mL) and CpG_A (1.5 µM) for 4 h, were incubated in glucose-free RPMI (ThermoFisher) supplemented (or not) with 20 μM 2-NBDG (2-[N-(7-nitrobenz-2-oxa-1, 3-diaxOL-4-yl) Amino]-2-deoxyglucose) (ThermoFisher) for 30 min at 37 ^°C under 5% CO2. Cells were then washed in cold PBS, and flow cytometry analysis was performed after surface staining of DC subsets (cDC1s, pDCs, CD5^+/– DC2s, CD14^+/– DC3s and AS DCs) as previously mentioned.

Statistical analysis

Statistical analyses were performed with the GraphPad Prism Software (version 9.3.1), by using the non-parametric one way ANOVA matched Friedman test followed by Bonferroni multiple comparison post-test, one way ANOVA unmatched Kruskal–Wallis test followed by Dunn’s multiple comparison post-test, the non parametric 2-way ANOVA or the mixed-effect model analysis test followed by Tuckey’s multiple comparison post-test, the non-parametric 2-way repeated measures (RM) ANOVA test or the mixed-effect model followed by Sidak’s multiple comparison post-test or the Geisser–Greenhouse correction, and the non-parametric Wilcoxon test. The significance threshold was placed at p < 0.05. Survival analyses (Kaplan-Meier), correlations and heat maps were performed using the clipr (0.8.0), dplyr (1.0.8), survival (3.3.1), survminer (0.4.9), forestmodel (4.0.5) corrplot (0.95), gplots (3.1.3), ggbiplot (0.6.2), and RColorBrewer (1.1.3) packages using RStudio software R version 4.2.1.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

Source data are provided with this paper.

References

Hargadon, K. M. Strategies to improve the efficacy of dendritic cell-based immunotherapy for melanoma. Front. Immunol. 8, 1594 (2017).
Article PubMed PubMed Central MATH Google Scholar
Del Prete, A. et al. Functional role of dendritic cell subsets in cancer progression and clinical implications. Int. J. Mol. Sci. 21, 3930 (2020).
Article PubMed PubMed Central MATH Google Scholar
Kvedaraite, E. & Ginhoux, F. Human dendritic cells in cancer. Sci. Immunol. 7, eabm9409 (2022).
Article CAS PubMed Google Scholar
Geginat, J. et al. Immunity to pathogens taught by specialized human dendritic cell subsets. Front. Immunol. 6, 527 (2015).
Article PubMed PubMed Central MATH Google Scholar
Cancel, J. C., Crozat, K., Dalod, M. & Mattiuz, R. Are conventional type 1 dendritic cells critical for protective antitumor immunity and how? Front. Immunol. 10, 9 (2019).
Article CAS PubMed PubMed Central Google Scholar
Saito, Y., Komori, S., Kotani, T., Murata, Y. & Matozaki, T. The role of type-2 conventional dendritic cells in the regulation of tumor immunity. Cancers 14, 1976 (2022).
Article CAS PubMed PubMed Central MATH Google Scholar
Monti, M., Consoli, F., Vescovi, R., Bugatti, M. & Vermi, W. Human plasmacytoid dendritic cells and cutaneous melanoma. Cells 9, 417 (2020).
Article CAS PubMed PubMed Central Google Scholar
Tucci, M. et al. Immune system evasion as hallmark of melanoma progression: the role of dendritic cells. Front. Oncol. 9, 1148 (2019).
Article PubMed PubMed Central MATH Google Scholar
Hargadon, K. M. Tumor-altered dendritic cell function: implications for anti-tumor immunity. Front. Immunol. 4, 192 (2013).
Article PubMed PubMed Central Google Scholar
Sosa Cuevas, E. et al. BDCA1(+) cDC2s, BDCA2(+) pDCs and BDCA3(+) cDC1s reveal distinct pathophysiologic features and impact on clinical outcomes in melanoma patients. Clin. Transl. Immunol. 9, e1190 (2020).
Article CAS MATH Google Scholar
Sosa Cuevas, E. et al. Diversification of circulating and tumor-infiltrating plasmacytoid DCs towards the P3 (CD80(+) PDL1(-))-pDC subset negatively correlated with clinical outcomes in melanoma patients. Clin. Transl. Immunol. 11, e1382 (2022).
Article CAS MATH Google Scholar
Aspord, C., Leccia, M. T., Charles, J. & Plumas, J. Plasmacytoid dendritic cells support melanoma progression by promoting Th2 and regulatory immunity through OX40L and ICOSL. Cancer Immunol. Res. 1, 402–415 (2013).
Article CAS PubMed Google Scholar
Aspord, C., Leccia, M. T., Charles, J. & Plumas, J. Melanoma hijacks plasmacytoid dendritic cells to promote its own progression. Oncoimmunology 3, e27402 (2014).
Article PubMed PubMed Central Google Scholar
Girard, P. et al. Dysfunctional BTN3A together with deregulated immune checkpoints and type I/II IFN dictate defective interplay between pDCs and gammadelta T cells in melanoma patients, which impacts clinical outcomes. Clin. Transl. Immunol. 10, e1329 (2021).
Article CAS Google Scholar
RodrIguez, E., Schetters, S. T. T. & van Kooyk, Y. The tumour glyco-code as a novel immune checkpoint for immunotherapy. Nat. Rev. Immunol. 18, 204–211 (2018).
Article CAS PubMed Google Scholar
Sosa Cuevas, E. et al. The melanoma tumor glyco-code impacts human dendritic cells’ functionality and dictates clinical outcomes. Front. Immunol. 14, 1120434 (2023).
Article PubMed PubMed Central Google Scholar
Sosa Cuevas, E. et al. Unique CLR expression patterns on circulating and tumor-infiltrating DC subsets correlated with clinical outcome in melanoma patients. Front. Immunol. 13, 1040600 (2022).
Article PubMed PubMed Central Google Scholar
Niveau, C. et al. Melanoma tumour-derived glycans hijack dendritic cell subsets through C-type lectin receptor binding. Immunology 171, 286–311 (2024).
Article CAS PubMed Google Scholar
Niveau, C., Sosa Cuevas, E., Saas, P. & Aspord, C. Glycans in melanoma: drivers of tumour progression but sweet targets to exploit for immunotherapy. Immunology 173, 33–52 (2024).
Shevchenko, I. & Bazhin, A. V. Metabolic checkpoints: novel avenues for immunotherapy of cancer. Front. Immunol. 9, 1816 (2018).
Article PubMed PubMed Central MATH Google Scholar
Leone, R. D. & Powell, J. D. Metabolism of immune cells in cancer. Nat. Rev. Cancer 20, 516–531 (2020).
Article CAS PubMed PubMed Central MATH Google Scholar
O’Sullivan, D., Sanin, D. E., Pearce, E. J. & Pearce, E. L. Metabolic interventions in the immune response to cancer. Nat. Rev. Immunol. 19, 324–335 (2019).
Article PubMed MATH Google Scholar
Kareva, I. & Hahnfeldt, P. The emerging “hallmarks” of metabolic reprogramming and immune evasion: distinct or linked? Cancer Res. 73, 2737–2742 (2013).
Article CAS PubMed MATH Google Scholar
O’Neill, L. A., Kishton, R. J. & Rathmell, J. A guide to immunometabolism for immunologists. Nat. Rev. Immunol. 16, 553–565 (2016).
Article PubMed PubMed Central MATH Google Scholar
Loftus, R. M. & Finlay, D. K. Immunometabolism: cellular metabolism turns immune regulator. J. Biol. Chem. 291, 1–10 (2016).
Article CAS PubMed Google Scholar
Du, X., Chapman, N. M. & Chi, H. Emerging roles of cellular metabolism in regulating dendritic cell subsets and function. Front. Cell Dev. Biol. 6, 152 (2018).
Article ADS PubMed PubMed Central MATH Google Scholar
Krawczyk, C. M. et al. Toll-like receptor-induced changes in glycolytic metabolism regulate dendritic cell activation. Blood 115, 4742–4749 (2010).
Article CAS PubMed PubMed Central MATH Google Scholar
Adamik, J. et al. Distinct metabolic states guide maturation of inflammatory and tolerogenic dendritic cells. Nat. Commun. 13, 5184 (2022).
Article ADS CAS PubMed PubMed Central MATH Google Scholar
O’Neill, L. A. & Pearce, E. J. Immunometabolism governs dendritic cell and macrophage function. J. Exp. Med. 213, 15–23 (2016).
Article PubMed PubMed Central Google Scholar
Chang, C. H. et al. Metabolic competition in the tumor microenvironment is a driver of cancer progression. Cell 162, 1229–1241 (2015).
Article CAS PubMed PubMed Central MATH Google Scholar
Giovanelli, P., Sandoval, T. A. & Cubillos-Ruiz, J. R. Dendritic cell metabolism and function in tumors. Trends Immunol. 40, 699–718 (2019).
Article CAS PubMed MATH Google Scholar
Hargadon, K. M. Tumor microenvironmental influences on dendritic cell and T cell function: a focus on clinically relevant immunologic and metabolic checkpoints. Clin. Transl. Med 10, 374–411 (2020).
Article PubMed PubMed Central Google Scholar
Dong, H. & Bullock, T. N. Metabolic influences that regulate dendritic cell function in tumors. Front. Immunol. 5, 24 (2014).
Article PubMed PubMed Central MATH Google Scholar
Arguello, R. J. et al. SCENITH: a flow cytometry-based method to functionally profile energy metabolism with single-cell resolution. Cell Metab. 32, 1063–75.e7 (2020).
Article CAS PubMed PubMed Central MATH Google Scholar
Villani, A. C. et al. Single-cell RNA-seq reveals new types of human blood dendritic cells, monocytes, and progenitors. Science 356, eaah4573 (2017).
Article PubMed PubMed Central MATH Google Scholar
Collin, M. & Bigley, V. Human dendritic cell subsets: an update. Immunology 154, 3–20 (2018).
Article CAS PubMed PubMed Central MATH Google Scholar
Bourdely, P. et al. Transcriptional and functional analysis of CD1c(+) human dendritic cells identifies a CD163(+) subset priming CD8(+)CD103(+) T cells. Immunity 53, 335–52.e8 (2020).
Article CAS PubMed PubMed Central Google Scholar
Dutertre, C. A. et al. Single-cell analysis of human mononuclear phagocytes reveals subset-defining markers and identifies circulating inflammatory dendritic cells. Immunity 51, 573–89.e8 (2019).
Article CAS PubMed MATH Google Scholar
See, P. et al. Mapping the human DC lineage through the integration of high-dimensional techniques. Science 356, eaag3009 (2017).
Article PubMed PubMed Central MATH Google Scholar
Voss, K. et al. A guide to interrogating immunometabolism. Nat. Rev. Immunol. 21, 637–652 (2021).
Article CAS PubMed PubMed Central MATH Google Scholar
Artyomov, M. N. & Van den Bossche, J. Immunometabolism in the single-cell era. Cell Metab. 32, 710–725 (2020).
Article CAS PubMed PubMed Central Google Scholar
Du, X. et al. Hippo/Mst signalling couples metabolic state and immune function of CD8alpha(+) dendritic cells. Nature 558, 141–145 (2018).
Article ADS CAS PubMed PubMed Central MATH Google Scholar
Basit, F. et al. Metabolomic and lipidomic signatures associated with activation of human cDC1 (BDCA3(+) /CD141(+)) dendritic cells. Immunology 165, 99–109 (2022).
Article CAS PubMed MATH Google Scholar
Basit, F., Mathan, T., Sancho, D. & de Vries, I. J. M. Human dendritic cell subsets undergo distinct metabolic reprogramming for immune response. Front. Immunol. 9, 2489 (2018).
Article PubMed PubMed Central Google Scholar
Wu, D. et al. Type 1 interferons induce changes in core metabolism that are critical for immune function. Immunity 44, 1325–1336 (2016).
Article CAS PubMed PubMed Central MATH Google Scholar
Bajwa, G. et al. Cutting edge: critical role of glycolysis in human plasmacytoid dendritic cell antiviral responses. J. Immunol. 196, 2004–2009 (2016).
Article CAS PubMed MATH Google Scholar
Saas, P., Varin, A., Perruche, S. & Ceroi, A. Recent insights into the implications of metabolism in plasmacytoid dendritic cell innate functions: Potential ways to control these functions. F1000Res 6, 456 (2017).
Article PubMed PubMed Central Google Scholar
Hu, Z. et al. Glycolysis drives STING signaling to facilitate dendritic cell antitumor function. J. Clin. Investig. 133, e166031 (2023).
Article CAS PubMed PubMed Central Google Scholar
Le Bourgeois, T. et al. Targeting T cell metabolism for improvement of cancer immunotherapy. Front. Oncol. 8, 237 (2018).
Article PubMed PubMed Central MATH Google Scholar
Nasi, A. et al. Dendritic cell reprogramming by endogenously produced lactic acid. J. Immunol. 191, 3090–3099 (2013).
Article CAS PubMed MATH Google Scholar
Zeng, F. L. et al. Interleukin-37 promotes DMBA/TPA skin cancer through SIGIRR-mediated inhibition of glycolysis in CD103(+)DC cells. MedComm 4, e229 (2023).
Article CAS PubMed PubMed Central Google Scholar
Challier, J., Bruniquel, D., Sewell, A. K. & Laugel, B. Adenosine and cAMP signalling skew human dendritic cell differentiation towards a tolerogenic phenotype with defective CD8(+) T-cell priming capacity. Immunology 138, 402–410 (2013).
Article CAS PubMed PubMed Central Google Scholar
Malinarich, F. et al. High mitochondrial respiration and glycolytic capacity represent a metabolic phenotype of human tolerogenic dendritic cells. J. Immunol. 194, 5174–5186 (2015).
Article CAS PubMed MATH Google Scholar
Zaal, A. et al. Activation of the C-type lectin MGL by terminal GalNAc ligands reduces the glycolytic activity of human dendritic cells. Front. Immunol. 11, 305 (2020).
Article CAS PubMed PubMed Central MATH Google Scholar
Munson, P. V. et al. Polyunsaturated fatty acid-bound alpha-fetoprotein promotes immune suppression by altering human dendritic cell metabolism. Cancer Res. 83, 1543–1557 (2023).
Article CAS PubMed PubMed Central MATH Google Scholar
Santos, P. M. et al. Tumor-derived alpha-fetoprotein suppresses fatty acid metabolism and oxidative phosphorylation in dendritic cells. Cancer Immunol. Res. 7, 1001–1012 (2019).
Article CAS PubMed Google Scholar
Li, C. et al. A high OXPHOS CD8 T cell subset is predictive of immunotherapy resistance in melanoma patients. J. Exp. Med. 219, e20202084 (2022).
Article CAS PubMed Google Scholar
Patel, C. H., Leone, R. D., Horton, M. R. & Powell, J. D. Targeting metabolism to regulate immune responses in autoimmunity and cancer. Nat. Rev. Drug Discov. 18, 669–688 (2019).
Article CAS PubMed MATH Google Scholar
Snyder, J. P. & Amiel, E. Regulation of dendritic cell immune function and metabolism by cellular nutrient sensor mammalian target of rapamycin (mTOR). Front. Immunol. 9, 3145 (2018).
Article CAS PubMed MATH Google Scholar
Saravia, J., Raynor, J. L., Chapman, N. M., Lim, S. A. & Chi, H. Signaling networks in immunometabolism. Cell Res. 30, 328–342 (2020).
Article CAS PubMed PubMed Central Google Scholar
Adamik, J. et al. Immuno-metabolic dendritic cell vaccine signatures associate with overall survival in vaccinated melanoma patients. Nat. Commun. 14, 7211 (2023).
Article ADS CAS PubMed PubMed Central MATH Google Scholar
Amiel, E. et al. Inhibition of mechanistic target of rapamycin promotes dendritic cell activation and enhances therapeutic autologous vaccination in mice. J. Immunol. 189, 2151–2158 (2012).
Article CAS PubMed MATH Google Scholar
Li, Q. et al. Regulating mammalian target of rapamycin to tune vaccination-induced CD8(+) T cell responses for tumor immunity. J. Immunol. 188, 3080–3087 (2012).
Article CAS PubMed MATH Google Scholar
Jiang, L., Fang, X., Wang, H., Li, D. & Wang, X. Ovarian cancer-intrinsic fatty acid synthase prevents anti-tumor immunity by disrupting tumor-infiltrating dendritic cells. Front. Immunol. 9, 2927 (2018).
Article CAS PubMed PubMed Central MATH Google Scholar
Lawless, S. J. et al. Glucose represses dendritic cell-induced T cell responses. Nat. Commun. 8, 15620 (2017).
Article ADS CAS PubMed PubMed Central Google Scholar
Thwe, P. M. & Amiel, E. The role of nitric oxide in metabolic regulation of dendritic cell immune function. Cancer Lett. 412, 236–242 (2018).
Article CAS PubMed MATH Google Scholar
Guo, W., Wang, H. & Li, C. Signal pathways of melanoma and targeted therapy. Signal Transduct. Target Ther. 6, 424 (2021).
Article PubMed PubMed Central MATH Google Scholar
Leone, R. D. et al. Glutamine blockade induces divergent metabolic programs to overcome tumor immune evasion. Science 366, 1013–1021 (2019).
Article ADS CAS PubMed PubMed Central MATH Google Scholar
Sharma, N. S. et al. Targeting tumor-intrinsic hexosamine biosynthesis sensitizes pancreatic cancer to anti-PD1 therapy. J. Clin. Investig. 130, 451–465 (2020).
Article CAS PubMed MATH Google Scholar

Download references

Acknowledgements

We thank Dr D. Legrand, Dr Y. Merieux, Y. Bouvier at EFS Auvergne Rhone-Alpes for providing healthy volunteers’ blood samples. We acknowledge the surgeons and anatomo-pathology Department from CHU Grenoble Alpes for providing tumor samples especially Pr Nathalie Sturm. We are grateful to Dr A. Grichine and Dr P. Marche for access to the cytometry platform; and all the volunteers and patients who agreed to participate in this study. We are grateful to Pr B. Toussaint, Dr A. Legouellec and C. Trocme from TIMC-IMAG-UMR-5525, CHU-Grenoble for access to the Luminex platform. We thank Dr N. Bendriss-Vermare and Pr F. Fieschi for stimulating discussions and precious advices. We are grateful to Dr R. Arguello for sharing the SCENITH technology and helpful advices.

This work was supported by the Etablissement Français du Sang (EFS) AuRA, Ligue contre le Cancer, the Société Française de Dermatologie (SFD), GEFLUC, Fondation BMS, and Université Grenoble Alpes (UGA).

Author information

Authors and Affiliations

Institute for Advanced Biosciences, Team: Epigenetics, Immunity, Metabolism, Cell Signaling & Cancer, Inserm U 1209, CNRS UMR 5309, Univ. Grenoble Alpes, Grenoble, France
Camille Niveau, Mélanie Cettour-Cave, Eleonora Sosa Cuevas, Philippe Saas & Caroline Aspord
Etablissement Français du Sang Auvergne-Rhône-Alpes, R&D Laboratory, Grenoble, France
Camille Niveau, Mélanie Cettour-Cave, Eleonora Sosa Cuevas, Philippe Saas & Caroline Aspord
Dermatology, Allergology & Photobiology Department, Univ. Grenoble Alpes, CHU Grenoble Alpes, Grenoble, France
Stéphane Mouret & Julie Charles
Optical Microscopy and Flow Cytometry (MicroCell), Institute for Advanced Biosciences, Inserm U 1209, CNRS UMR 5309, Univ. Grenoble Alpes, Grenoble, France
Mylene Pezet
GLYcoDiag, Orléans, France
Benoît Roubinet & Ludovic Landemarre
Department of Anatomopathology, Grenoble Alpes University Hospital Center, Grenoble, France
Hugo Gil
Medical Unit of Molecular genetic (hereditary diseases and oncology), Grenoble University Hospital, Grenoble, France
Florence De Fraipont

Authors

Camille Niveau
View author publications
Search author on:PubMed Google Scholar
Mélanie Cettour-Cave
View author publications
Search author on:PubMed Google Scholar
Stéphane Mouret
View author publications
Search author on:PubMed Google Scholar
Eleonora Sosa Cuevas
View author publications
Search author on:PubMed Google Scholar
Mylene Pezet
View author publications
Search author on:PubMed Google Scholar
Benoît Roubinet
View author publications
Search author on:PubMed Google Scholar
Hugo Gil
View author publications
Search author on:PubMed Google Scholar
Florence De Fraipont
View author publications
Search author on:PubMed Google Scholar
Ludovic Landemarre
View author publications
Search author on:PubMed Google Scholar
Julie Charles
View author publications
Search author on:PubMed Google Scholar
Philippe Saas
View author publications
Search author on:PubMed Google Scholar
Caroline Aspord
View author publications
Search author on:PubMed Google Scholar

Contributions

C.A. conceived the project and supervised the study; C.A., C.N. designed the experiments; C.A., C.N., M.C.C., E.S., B.R., L.L. M.P. performed the experiments and analyzed the data; S.M., J.C., H.G., F.F. provided patient material and clinical data, and expertise with clinical interpretation of the data; B.R., L.L., P.S. provided research input and contributed to data interpretation; C.A., C.N. wrote the manuscript; S.M., E.S., B.R., L.L., P.S. contributed to manuscript editing.

Corresponding author

Correspondence to Caroline Aspord.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks Alpaslan Tasdogan and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Reporting Summary

Transparent Peer Review file

Source data

Source Data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Cite this article

Niveau, C., Cettour-Cave, M., Mouret, S. et al. MCT1 lactate transporter blockade re-invigorates anti-tumor immunity through metabolic rewiring of dendritic cells in melanoma. Nat Commun 16, 1083 (2025). https://doi.org/10.1038/s41467-025-56392-x

Download citation

Received: 20 October 2023
Accepted: 16 January 2025
Published: 27 January 2025
DOI: https://doi.org/10.1038/s41467-025-56392-x

This article is cited by

Dendritic cells: understanding ontogeny, subsets, functions, and their clinical applications
- Wenhao Li
- Chenyu Yu
- Yanfeng Wu
Molecular Biomedicine (2025)
A new era in melanoma immunotherapy: focus on DCs metabolic reprogramming
- Mina Afrashteh Nour
- Mansour Rajabivahid
- Farid Ghorbaninezhad
Cancer Cell International (2025)
Metabolic reprogramming in melanoma therapy
- Dongliang Shen
- Lu Zhang
- Liling Tang
Cell Death Discovery (2025)
The Emerging Role of Lactate and Lactylation Modifications in the Pathophysiology of Atherosclerotic Cardiovascular Diseases
- Xianqiong Lu
- Qiyuan Luo
- Jiyu Chen
Cardiovascular Drugs and Therapy (2025)