The endocannabinoid system (ECS) is a remarkably complex regulatory network of neurotransmitters, receptors, and enzymes that plays an important role in maintaining homeostasis within the human body. Through its various elements the ECS modulates a wide range of physiological processes, contributing to overall health and well-being. The discovery of the ECS revolutionized our understanding of how cannabinoids, both external and internal, interact with the human body. This regulatory system influences a wide range of functions, including pain modulation, appetite, mood regulation, sleep, memory, the immune response and inflammation (1, 2).

The ECS takes its name from cannabinoids, the chemical compounds found in the cannabis plant, which interact with the system’s receptors. The ECS is not solely influenced by external cannabinoids, it also produces endogenous ligands, known as endocannabinoids (ECs). These ECs, such as N-arachidonoylethanolamine (anandamide, AEA) and 2-arachidonoylglycerol (2-AG), are synthesized on-demand in response to specific physiological signals (3). The ECs mainly act on two cannabinoid receptors, the CB1 and CB2 receptors, which were discovered in the late 20th century (4–6).

Most ECs are produced from membrane precursors in response to stimuli, although anandamide can be stored in the cell (7). The concentration of ECs in the body fluctuate greatly, and is influenced by many factors. As a consequence serum anandamide levels generally range from 1 to 5 nM, and 2-AG levels from 10 to 500 nM (8).

The ECS has significant impact on skin homeostasis. As our body’s largest organ, skin serves as a protective barrier between our internal and the external environment. Recent studies have revealed an intriguing connection between the skin and the ECS (9, 10). Beyond its role in physical defense, the skin is a dynamic and complex organ involved in numerous physiological processes, including temperature regulation, sensation, and the orchestration of immune responses. The presence of the ECS in the skin signifies its importance in maintaining the intricate balance of skin functions. Both CB1 and CB2 receptors, the primary receptors of the ECS, have been identified in various components of the skin, including epidermal keratinocytes, sebaceous glands, hair follicles, and sensory nerve fibers. The ECS is involved in the regulation of skin cell proliferation, survival and differentiation (11). This widespread distribution suggests that the ECS plays a significant role in regulating multiple aspects of skin physiology. Fine-tuning endocannabinoid tone appears to be a key factor in modulating skin growth and differentiation (12, 13).

One key area where the ECS exerts its influence is in the regulation of inflammation in the skin. Inflammatory skin conditions, such as acne, eczema, psoriasis, and allergic dermatitis, involve an immune response gone awry. Studies have shown that activation of cannabinoid receptors in the skin can help modulate immune responses, reducing excessive inflammation. This suggests that targeting the ECS may hold promise in developing novel anti-inflammatory treatments for skin disorders (9).

Local regulation of immune cells by endocannabinoid mediators is a well-documented phenomenon. The most common ECs (anandamide and 2-AG) are characterized by a dominant anti-inflammatory effect, as they reduce the production of pro-inflammatory cytokines in many immune cells, induce cell migration or even cell apoptosis (14). Anandamide at low, nanomolar concentrations (3–30 nM) reduced the production of interleukin (IL-6) and IL-8, but at higher concentrations (0.3–3 μM) inhibited the production of tumor necrosis factor alpha (TNFα), interferon γ (IFNγ), IL-4 and soluble p75 TNFα receptors (15). Anandamide has also been shown to suppress the release of IL-2, TNFα and IFNγ cytokines from activated human peripheral T lymphocytes, mainly through a CB2-dependent mechanism (16). It also effectively suppressed the in vitro induction of IL-6 and the release of nitric oxide (NO) from macrophages triggered by lipopolysaccharides (17). Primary human monocytes show increased production of NO via the CB1 receptor upon administration of increasing doses of 2-AG. Interestingly, these cells become rounded and immobile upon 2-AG treatment, which appears to be immunosuppressive in terms of decreased cytokine production and adhesion (18). 2-AG also strongly suppresses Toll-like receptor (TLR)9-mediated type I IFN production by human primary plasmacytoid dendritic cells. The effect is mediated exclusively by CB2, and the abolition of this 2-AG inhibition has been shown to be key in the pathogenesis of systemic lupus erythematosus (SLE) (19).

Although its effects on cell function may vary and even be controversial in some cases, it is generally agreed that the ECS acts as a kind of endogenous brake on both innate and adaptive immune processes, as detailed above.

LCs are professional antigen presenting cells derived from yolk sac precursors similarly to tissue resident macrophages, but functionally they share many characteristics with dendritic cells (DCs). They can be identified primarily based on their expression of CD1a and langerin/CD207 in the stratum spinosum of the epidermis and in the mucosa, where their main function is antigen presentation to T cells after their activation (20). The efficient antigen presentation of LCs to CD4⁺ and CD8⁺ T cells highlights their important role in the induction of helper T cell (Th) mediated regulatory and humoral immune responses, as well as in the induction and maintenance of cytotoxic T cell immunity (21–23). LCs that cross-present antigens are also essential for the initial reactivation of resident CD8⁺ memory T cells in the epidermis (24). LCs can migrate to surrounding lymph nodes mediated by CCR7 to activate naïve T cells by antigen presentation (25). More recent research has proposed that there are two steady-state LC subtypes in the epidermis of human skin (CD1a^highlangerin^high LC1 and CD1a^highlangerin^int LC2), with higher CD1c expression and better Birbeck granule organization in LC1 vs. LC2 (26).

LC-like cells can also be derived from myeloid precursors. Specifically, CD14⁺ monocytes from adult peripheral blood undergo differentiation into monocyte-derived LCs (moLCs) when exposed to granulocyte/macrophage colony-stimulating factor (GM-CSF), IL-4, and transforming growth factor β (TGF- β) (27–29). There is no standard protocol widely adopted among research groups for generating moLCs (27, 30–33). Our research group worked to refine protocols to ensure the accuracy and reproducibility of the moLC experiments (34). In alignment with the existing literature, our study demonstrates a striking resemblance between the model cells we employed and the previously mentioned LC2 subtype.

As detailed above both LCs and the ECS are fascinating components of the human body’s (immune) system. Nonetheless, the effect of ECs (i.e. anandamide and 2-AG) has not yet been investigated in LCs, the only professional antigen-presenting cell in the skin (35–37).

Understanding the interaction between the ECS and LCs holds promise to unveil a fascinating avenue to explore to help maintain skin health, and potentially to develop innovative therapeutic interventions targeting LCs. Furthermore, understanding the ECS’s intricate mechanisms provides a foundation for exploring the therapeutic potential of ECs in treating various inflammatory skin diseases.

Endocannabinoids in general, and anandamide specifically has generally been shown to act as an endogenous brake on multiple immune cells (3, 6, 42). We started investigating the effect of anandamide on the differentiation and activation of moLCs, while ruling out the possibility of toxicity caused by the treatment. One of the significant effects of anandamide was the downregulation of CD1a on moLCs, without a concomitant change in CD207 expression ( Figures 1B, C ). As LCs in general are characterized by both CD1a and CD207 expression, the lack of change in the latter hints that anandamide does not inhibit moLC differentiation, but selectively downregulates CD1a expression. Surprisingly anandamide increased the expression of MHCII molecules, CD86 and even CCR7 ( Figures 1E-G ). This hints that moLCs differentiated in the presence of anandamide are more effective at presenting antigen to, and activating naïve T cells, and that through their increased CCR7 expression are more likely to migrate to draining lymph nodes from the skin, further supporting the relatively selective effect on CD1a as opposed to more general inhibition. We could also rule out the possibility that anandamide was simply toxic to the cells, as we saw no increase in 7-AAD positive cells ( Figure 1A ). This was an important control experiment, as others have reported significant cell death with similar concentrations in keratinocytes [10 μM, (43)], melanoma cells [13.5 μM, (44); >5 μM, (45)], sebocytes [>10 μM, (46)], and murine DCs [>5 μM, (47)].

Further supporting the idea that anandamide in the context of moLCs is not purely anti-inflammatory we found that activation of the cells with TLR3 and TLR7/8 agonists was just as effective in cells differentiated in the presence of the endocannabinoid as in control cells ( Figures 1B-F ). Interestingly, the TLR agonists did not increase the expression of CCR7 in control cells, which corresponds to the “activated LC” phenotype established by Liu et al., which is characterized as CCR7^low as opposed to the “migratory LC” that have high CCR7 expression (26).

In contrast to anandamide, 2-AG had no significant effects on the markers typically investigated on antigen presenting cells such as LCs and DCs (34), so we excluded this treatment from subsequent experiments ( Supplementary Figure 1 ). This lack of effect from 2-AG was not completely unexpected, since although it was able to induce maturation of the DC2.4 DCs line, it did not influence either CD4⁺ or CD8⁺ T cells in the tumor microenvironment in mice (48).

CD1a is primarily expressed on the surface of cortical thymocytes, LCs in the skin and mucosal surfaces, and some DCs. The CD1a molecule has a structure similar to major histocompatibility complex (MHC) molecules, which are involved in presenting peptide antigens to T cells. CD1 molecules are specialized for presenting lipid antigens rather than peptide antigens. There are multiple possible mechanisms that can explain the decrease in CD1a, as this might occur due to increased cAMP concentration inside the cell during its differentiation (49), binding of anandamide as a lipophilic molecule to CD1a which can influence its trafficking and result in lower expression overall. In monocyte derived DC (moDC) cultures the presence of lipoproteins in culturing media causes an activation of peroxisome proliferator-activated receptor gamma, which decreases the expression of CD1a (50). This drop in CD1a could conceivably result in a decreased capability to activate naïve T cells, as there are certain T cell populations that are restricted to CD1a (51), however in our experiments we found the exact opposite effect: anandamide was capable of significantly increasing the ratio of proliferating allogenic T cells ( Figure 4 ). As CD1a-restricted T cells have been implicated in multiple inflammatory skin diseases such as psoriasis, atopic dermatitis and even contact hypersensitivity, this targeted decrease of CD1a expression might be beneficial in these diseases (52).

After investigating the effect of anandamide on the differentiation and activation of the cells, we next evaluated the function of moLCs differentiated in the presence of anandamide. Even though anandamide showed some proinflammatory effects via the markers detailed above, when investigating its effects on cytokine secretion it appeared to induce anti-inflammatory results. At high concentration (10 μM) anandamide decreased the TLR3 and TLR7/8 induced production of CXCL8, IL-6, IL-10, IL-12 ( Figures 2A-D ). Similar effects have been described by others, for example in J774 macrophages, anandamide hinders LPS-induced NO and IL-6 synthesis in a concentration-dependent manner (17). Anandamide also effectively modulates the activity of macrophages and monocytes, displaying a regulatory impact contingent on both dosage and time. Previous results on human peripheral blood mononuclear cells (PBMCs) have shown anandamide diminishes the generation of IL-6 and CXCL8 at lower concentrations, while completely suppressing the release of TNFα, IFNγ, and IL-4 at higher concentrations (15). Further in line with our findings, anandamide also suppresses TLR7/8-induced IL-12 and IL-6 production in human myeloid DCs and inhibits IFNα production in human plasmacytoid DC (53). Furthermore, anandamide also inhibits the proliferation and secretion of cytokines, such as IL-2, TNFα, and IFNγ, from activated human peripheral T lymphocytes (15, 16). These results show that our observation that anandamide dampens cytokine secretion is not unique to moLCs.

Anandamide has been shown to act on a wide range of cellular receptors, including the non-selective cation channel TRPV1 (54), the activation of which can lead to increases in intracellular calcium. As calcium signaling is an important mediator in cytokine secretion, maturation, and a host of other immune functions in both murine and human cells (55–59) we next wished to investigate the effect of anandamide on the intracellular calcium concentration of moLCs. Although anandamide increased the intracellular calcium concentration of moLCs as determined by fluorescent calcium imaging, not all cells were equally responsive to anandamide ( Figure 3 , Supplementary Videos 1 - 4 ). In fact, the calcium transient that can be observed occurs relatively late after the application of anandamide (with an approximate maximal response at 150 seconds), which stands in contrast to the response elicited by capsaicin on a similar moLC model (33). According to the kinetics of the calcium response, and the low number of responsive cells, it is unlikely that the effects of anandamide are dependent on TRPV1 activation on our moLCs. moLCs also express other TRP channels, such as TRPV2 and TRPV4 (34), but anandamide has not been described to act on these channels.

Based on the above presented results anandamide seems to induce partly antagonistic effects: it boosts antigen presentation and subsequently T cell proliferation, both of which could be considered proinflammatory responses, while on the other hand, cytokine and chemokine production is decreased by anandamide treatment. To gain deeper insight into the induced responses we performed RNASeq analysis of moLCs differentiated in the presence of anandamide, with and without activation by TLR ligands ( Figures 5 , 6 ). This analysis highlighted two important details regarding the transcriptome of moLCs. First, we found that anandamide treatment upregulated type II IFN receptor activity, which suggests that it might help polarize T cells activated by moLCs towards a Th1 phenotype. Secondly, we found that in cells activated with TLR agonists there is an enrichment of genes involved in oxidative phosphorylation ( Figure 6 , right panel), which is absent from cells differentiated in the presence of anandamide. This is accompanied by enrichment in other metabolic pathways, such as aerobic electron transport chain, ATP synthesis coupled electron transport, aerobic respiration, and others. Typically, activated myeloid cells tend to be highly glycolytic with little or no flux through oxidative phosphorylation. Activation of DCs, which share many functional characteristics with LCs, results in a switch to glycolysis to support de novo lipid biosynthesis, facilitating the expansion of the endoplasmic reticulum and Golgi apparatus and increasing the biosynthetic capacity of the cells (60). The concomitant decrease in oxidative phosphorylation typically results in a decrease in activated DC lifespan, which can be rescued by continued availability of oxidative phosphorylation to the cells, by for example mTOR inhibition (61). As we saw no decrease in either the viability or the T cell activating capability of anandamide and TLR agonist treated moLCs compared to moLCs only treated with p(I:C) and CL075, this metabolic shift is most likely insufficient in scope to decrease the longevity of the cells. Another consequence of this shift could be the accumulation of metabolic intermediaries required for the de novo synthesis of cytokines, as seen in DCs (62). As anandamide seems to interfere with this shift, this might underlie the decrease in cytokine production we observed ( Figure 2 ). It is also important to note that increased oxidative phosphorylation is a hallmark of tissue resident macrophages, which are ontogenically close relatives of LCs. Upregulation of this pathway in activated LCs would increase their longevity, which could be beneficial in the case of cells that remain in the skin (which is supported by their low CCR7 expression), but would likely not be required for migratory LCs with high CCR7 expression (63). It is important to note that these effects on the metabolism of moLCs should be investigated in more detail, as changes in the transcriptome do not necessarily translate to concomitant changes in the proteome. Further, targeted experiments are required to substantiate these preliminary findings in subsequent investigations. The effect of anandamide on cellular metabolism is not completely surprising, however. A recent manuscript (64) details a novel method of identifying cellular targets of cannabinoids, and they identify 64 new targets, many of which are associated with the mitochondria. Of these 64 novel targets 10 are directly involved in aerobic respiration pathways, which might explain our own gene set enrichment results.

In contrast to the negative effect of anandamide on chemokine/cytokine secretion, we found that moLCs treated with anandamide alone stimulated the proliferation of naïve T cells, albeit not to the level reached with combined TLR agonists ( Figure 4 ). The combination of anandamide and TLR agonists did not exhibit a notably additive effect; however, a slight increase was observed between the p(I:C) + CL075 group and the anandamide + p(I:C) + CL075 group ( Figure 4A , with representative histogram plots depicted in Figure 4B ).

Although most literature data are from mouse models, these show that LCs are able to induce Th1, Th2 and Th17 (65, 66) and most likely, regulatory T cells (67). When the immune response of moDC and moLC were investigated upon stimulation with TLR agonists, with or without proinflammatory cytokines TNFα and IL-1β minor differences could be observed in T cell polarization. Bacterial agonists selectively up-regulated CD83 and CD86 expression and induced Th1 and Th17 response in moDC, which showed higher migratory capacity compared to moLCs. MoLC activation with LPS enhanced the mature state (CD83, CD86 marker expression), and also increased IL-12p70, IL-23, and IL-6 production, and induced Th1 cytokine production by CD4⁺ T cells (68). In accordance with these findings, in our experiments, p(I:C) and CL075 stimulation up-regulated the costimulatory molecules and increased IL-12 secretion in moLCs ( Figure 2D ).

As RNASeq analysis revealed an increase in IFN gamma receptor activity in anandamide treated moLCs ( Figure 5F ), as well as a marked increase in IL-12 production in TLR agonist treated moLCs ( Figure 2D ), we next wished to investigate the Th1 polarization of activated T cells. Using ELISpot we could determine that anandamide alone can increase both the intensity of staining and the area covered by IFNγ positive T cells, indicating that there is an increase in proliferating T cells (supporting Figure 4 ), as well as their IFNγ content. The combined application of p(I:C) and CL075, as expected, resulted in a much stronger response, which was not influenced further by the addition of anandamide during the differentiation of moLCs. It is important to note that in these experiments T cells were not treated with anandamide directly, as this is known to decrease both TNFα and IFNγ production from primary human T lymphocytes at low micromolar concentrations, and can induce apoptosis at high micromolar concentrations (16, 42), but rather the endocannabinoid was only applied to moLCs during their differentiation.

Based on the results presented above we can state that moLCs join the ranks of immune cells under the regulation of the ECS. Although ECs in general, and anandamide especially are often considered anti-inflammatory, the nuanced effects on maturation, cytokine secretion, T cell proliferation and polarization presented above underline the fact that an integrated overview of cell-specific effects is required to accurately predict the role of the ECS in regulating immune responses.

As our knowledge of the ECS continues to expand, it is becoming increasingly clear that this intricate network plays a pivotal role in maintaining human health and offers promising avenues for therapeutic interventions. As CD1a signaling is implicated in common inflammatory skin disorders, the mitigating effect of anandamide on CD1a expression holds promise to selectively dampen inflammation induced by CD1a restricted T cells, without impacting systemic immune responses as antigen presentation through the MHC pathway is not impacted in a negative way. The evolving understanding of the ECS’s involvement in skin health opens up exciting opportunities for developing novel cannabinoid-based therapies for dermatological conditions and promoting overall skin well-being.

The datasets generated for this study can be found in the Gene Expression Omnibus repository hosted at the National Library of Medicine, under accession number GSE266075. https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE266075.

The studies involving humans were approved by Regional and Institutional Ethics Committee of the University of Debrecen’s Faculty of Medicine (Debrecen, Hungary; approval number: OVSZK 3572-2/2015/5200). The studies were conducted in accordance with the local legislation and institutional requirements. The human samples used in this study were acquired from a by- product of routine care or industry. Written informed consent for participation was not required from the participants or the participants’ legal guardians/next of kin in accordance with the national legislation and institutional requirements.

ZP: Conceptualization, Formal Analysis, Investigation, Methodology, Visualization, Writing – original draft, Writing – review & editing. DH: Writing – review & editing. PM: Writing – review & editing. TF: Writing – review & editing. KP: Writing – original draft, Writing – review & editing. AB: Resources, Writing – review & editing. AS: Conceptualization, Formal Analysis, Funding acquisition, Project administration, Supervision, Writing – original draft, Writing – review & editing.