Cannabinoids affect diverse biological processes, including functions of the immune system. With respect to the immune system, anti-inflammatory and immunosuppressive effects of cannabinoids have been reported. Cannabinoids stimulate G protein-coupled cannabinoid receptors CB1 and CB2. These receptors are found primarily on neurons. However, they are also found on dendritic cells (DC), which are recognized for their critical role in initiating and maintaining immune responses. Therefore, DC are potential targets for cannabinoids. We report in this study that cannabinoids reduced the DC surface expression of MHC class II molecules as well as their capacity to stimulate T cells. In the nervous system, CB1 receptor signaling modulates K+ and Ca2+ channels. Interestingly, cannabinoid-treated DC also showed altered voltage-gated potassium (KV) channel function. We speculate that attenuation of KV channel function via CB1 receptor signaling in DC may represent one mechanism by which cannabinoids alter DC function.
Human and animal cannabinoids belong to the group of endocannabinoids that are cell membrane-derived signaling molecules formed during the metabolism of eicosanoid fatty acids (1). The endocannabinoids, such as anandamide, can activate a group of G protein-coupled cannabinoid receptors, CB1 and CB2 (2). Plant-derived cannabinoids and synthetic structural analogs of cannabinoids referred to as exogenous cannabinoids can also function to activate the specific cannabinoid receptors, normally activated by the endocannabinoids (3). The CB1 and CB2 receptors are found primarily on neurons (4). In the nervous system, CB1 receptor signaling modulates K+ and Ca2+ channels (5, 6, 7, 8, 9). The CB1-mediated modulation of voltage-gated potassium (KV)3 channel function can be regulated by both endogenous and exogenous cannabinoids (6, 9). Cannabinoids affect not only the nervous system but also the immune system, and anti-inflammatory effects of cannabinoids have been observed in cannabinoid-based drug therapies (10). The outcome is attenuation of the symptoms and progression of neuroinflammatory disorders and inflammatory bowel diseases. These anti-inflammatory effects are mediated, at least partially, through the binding of cannabinoid receptors and correlate with decreased T cell responses and reduced production of inflammatory mediators (10, 11, 12). Although, a wealth of information indicates that cannabinoids have immune suppressive and anti-inflammatory activities, the exact mechanisms of immune modulation remain unknown. Because CB1 receptors and KV channels are present on lymphocytes, macrophages, and dendritic cells (DC) (13, 14, 15), the anti-inflammatory effects of cannabinoids might involve regulation of K+ channels via CB1 receptor signaling in immune cells, including DC.
DC are ubiquitous sentinels of the immune system and participate in regulating both innate and adaptive immunity (16). By possessing a range of cellular receptors, DC respond to microbial and inflammatory stimuli and undergo a process of cellular activation termed maturation (17). Upon maturation, DC enhance their Ag-presenting capacity by changing their surface phenotype involving redistribution of MHC class II (MHC-II) from intracellular compartments to the plasma membrane (18, 19, 20, 21, 22). Maturation is also associated with increased cell surface levels of costimulatory molecules and enhanced production of soluble inflammatory mediators (23, 24, 25, 26). Consequently, mature DC possess important properties for activating and directing functional differentiation of Ag-specific T cells (27). Regulation of the immunostimulatory capacity of DC is therefore a key step for determining the nature and effectiveness of T cell-mediated immune responses. In addition, various mechanisms may act at distinct levels in fine-tuning DC function to prevent excessive immune responses and the onset of pathophysiological conditions. Targeting such mechanisms may serve as a means for therapeutic modulation of DC function in chronic inflammatory diseases associated with, for example, autoimmunity and transplantation (28).
Attenuation of KV channel function offers a promising means to downmodulate immune responses (14, 15, 29). As previously reported, DC express functional ion channels (30, 31, 32), including KV channels (13). However, because knowledge about possible means of KV channel modulation in DC is limited, we initially set out to study whether CB1 receptor activation impacts that process. Our data revealed that KV channel function in murine DC does undergo changes in response to CB1 receptor signaling. Subsequent studies showed that DC responded to both endogenous and exogenous cannabinoids, and that this response involves attenuated KV channel-mediated outward currents. Interestingly, this cannabinoid-induced attenuation of KV channel function in DC correlated with a reduced surface expression of MHC-II on DC and rendered DC less potent activators of T cells.
Materials and Methods
C57BL/6 mice were bred at the Department of Neuroscience, Karolinska Institutet (Stockholm, Sweden). They were used at 5–8 wk of age and were housed under conventional conditions with free access to food and water. All animal care and experimental procedures were approved by the Animal Ethics Committee of Stockholm, Sweden.
Chemicals and drugs
Tetraethyl-ammonium (TEA), pertussis toxin (PTX), HEPES, EGTA, and barium chloride were all from Sigma-Aldrich. The endogenous cannabinoid, anandamide (also known as arachidonylethanolamide), the CB1 receptor agonist ACPA (arachidonylcyclopropylamide), a structural analog to anandamide, and the CB1 receptor antagonist AM251 were all from Tocris. Stock solutions of anandamide and ACPA were prepared in Tocrisolve 100 (Tocris) and stored at −20°C. Aliquots prepared in sterile water were diluted to their final concentration with extracellular solution (see below). A stock solution of AM251 was prepared in DMSO (Sigma-Aldrich) and diluted to its final concentration with extracellular solution. In all control experiments, this “vehicle” contained the appropriate concentrations of stock solution diluents, i.e., DMSO or Tocrisolve.
DC generation and isolation
DC were generated from bone marrow cells cultured as previously described (33). Bone marrow cells were obtained from the femurs and tibias of normal C57BL/6 mice and seeded at 3 × 105 cells/ml in DMEM (Invitrogen) with glutamax I (Invitrogen), 10% FCS (Invitrogen), 100 U/ml penicillin, and 100 g/ml streptomycin (Invitrogen), supplemented with recombinant murine GM-CSF (10 ng/ml; PeproTech) in the absence or presence of murine IL-4 (10 ng/ml; PeproTech). At day 3 of culture, the medium was gently removed, and fresh medium supplemented with the growth factors was added. After 6 days of culture, floating and lightly adherent cells were collected and seeded at 3 × 105 cells/ml in new tissue culture plates. On the following day, floating and lightly adherent cells were collected (34), and DC were purified using CD11c microbeads (Miltenyi Biotec). The majority of purified cells had a typical DC morphology, and more than 95% of them expressed CD11c. Purified DC were seeded in tissue culture plates at 1–5 × 105 cells/ml in medium containing GM-CSF. For the electrophysiologic recording assay, DC were cultured for 1–4 days postpurification. DC were also isolated from spleens of C57BL/6 mice. Spleens were digested in RPMI 1640 supplemented with 0.5 mg/ml collagenase and 0.5 mg/ml DNase (Sigma-Aldrich) for 10 min at 37°C. Then DC were enriched by magnetic cell sorting using CD11c microbeads and a MidiMACS separation column (both from Miltenyi Biotec) following the manufacturer’s protocol. In assays performed to study the effect of KV channel blocking on DC function, expression of MHC-II and costimulatory molecules, DC were used directly or cultured in the absence or presence of LPS for 16 h before adding the KV channel blocking reagent TEA and the cannabinoids ACPA and anandamide for 30 min.
The DC culture medium was replaced with an extracellular solution composed NaCl 140 mM, KCl 5 mM, CaCl2 1.8 mM, MgCl2 1 mM, sucrose 10 mM (all from KEBO Laboratory) and HEPES 10 mM. The pH was adjusted to 7.4 with sodium hydroxide and osmolarity to 305 mOsm with sucrose. In some experiments, calcium chloride was replaced with barium chloride BaCl2. Whole cell patch-clamp recordings (35) were then made from DC using capillary glass tubing (1.5 mm in diameter, GC150–10; Harvard Apparatus) pulled to form tips having a resistance of ∼5 MΩ. The pipettes contained an intracellular solution composed of KCl 140 mM, NaCl 4 mM, HEPES 10 mM, EGTA 5 mM, CaCl2 0.5 mM, MgCl2 1 mM. The pH was adjusted to 7.4 with potassium hydroxide and osmolarity to 300 mOsm with sucrose. Control and test solutions were exchanged using a gravity-fed perfusion system with electronically controlled valves (ValveLink 8; AutoMate Scientific). Cells chosen for recording were 20–30 μM in diameter and typically had minimal ruffling or processes; this provided the best condition for obtaining a seal with the patch pipette. An Axopatch 200B (Axon Instruments) was used for recordings. For voltage clamp recordings, series resistance was compensated by 75–85%. Computer software (pClamp8; Axon Instruments) together with a digital interface (Digidata 1320A; Axon Instruments) were used to control the amplifier and data acquisition. All experiments were done at room temperature with voltage-clamp holding potential (Vhold) of −70 mV. Voltage steps were then increased by 5-mV increments from voltage-clamp holding potential +10 or +25 mV.
Immunofluorescence and confocal analysis
For immunofluorescence labeling of the CB1 receptor, DC were grown on the culture dishes and fixed in 4% paraformaldehyde with 0.1% glutaraldehyde in PBS for 15 min and permeabilized with 0.3% Triton X-100 (Sigma-Aldrich) in PBS for 15 min. After blocking with 5% BSA for 30 min, the cells were incubated with the N-terminal Ab for the CB1 receptor (9), diluted in PBS containing 2% BSA overnight at 4°C. Specific staining was detected by incubating with Cy3-conjugated donkey anti-rabbit IgG Ab (Jackson ImmunoResearch Laboratories), diluted to 3.75 μg/ml in PBS containing 2% BSA for 2 h at room temperature. The cells were rinsed in PBS between treatments and mounted in glycerol with 2.5% diazabicyclanooctane (Sigma-Aldrich). Confocal microscopy was performed with a Zeiss LSM 510 Meta Axioplan 2 System. Projections were made from five 1-μm thick optical sections.
For immunofluorescence labeling of MHC-II, DC were spun onto glass slides by cytospin. Cells were fixed in 4% paraformaldehyde in PBS for 30 min at room temperature. After fixation, samples were blocked and permeabilized in 1.5% normal goat serum and 0.1% saponin in PBS for 45 min at room temperature. Samples were then incubated with purified Ab diluted in 1.5% normal goat serum and 0.1% saponin in PBS for 45 min at room temperature. As primary reagent, we used rat anti-mouse MHC-II mAb (clone 2G9; BD Pharmingen). After washing cells four times in 1.5% normal goat serum and 0.1% saponin in PBS, specific staining was detected by Alexa Fluor 488 conjugated goat anti-rat IgG Ab (Molecular Probes). Cover slips were mounted in anti-fade (Molecular Probes) and visualized with a confocal microscope (Leica TCS SP2 AOBS; Leica Microsystems). Images shown are single optical slices (0.8–1.0 μm).
Detection of CB1 receptor by Western immunoblotting and PCR
Control or LPS-matured DC were lysed in 250 μl of boiling lysis buffer (28 mM Tris-HCl, 22 mM Tris base, 200 mM DTT, 0.3% SDS; all from Sigma-Aldrich), precipitated in 80% methanol, and the pellets were diluted and boiled in SDS sample buffer. Electrophoresis was performed using 10% Tris-glycine gels according to Laemmli system and Hackstein and Thomson (28). Proteins were transferred to Immobilon-PSQ Transfer membranes (Millipore) in bicine transfer buffer with 20% methanol at 35 V for 4 h. Membranes were blocked in 5% BSA (Sigma-Aldrich) and incubated with the N-terminal Ab for the CB1 receptor (9), diluted in 0.5% BSA followed by the secondary peroxidase-conjugated goat anti-rabbit Ab (0.05 μg/ml; DAKOCytomation). Detection was performed with ECLplus (Amersham Biosciences).
Total RNA was extracted from triplicate cell cultures of control DC and LPS-matured DC, using the RNeasy Mini kit (Qiagen). RNA content was determined by UV spectroscopy (Ultrospec Plus; Pharmacia). RNA was treated with amplification grade DNase I and reverse transcribed using Superscript II RNase H− Reverse Transcriptase and random primers (all from Invitrogen). Oligonucleotides used for specific amplification of CB1 were 5′-TGC ACA AGC ACG CCA ATA A-3′ (sense) and 5′ACA GTG CTC TTG ATG CAG CTT TC-3′ (antisense), and for amplification of the housekeeping gene GAPDH they were 5′-CCA TGG AGA AGG CCG GGG-3′ (sense) and 5′-CAA AGT TGT CAT GGA TGA CC-3′ (antisense). Specific amplification of CB1 and GAPDH transcripts was subsequently performed in 25 μl of reaction volumes according to the following protocol. Each PCR mixture contained 1 μl of template; 100 ng of each primer; 1X Titanium TaqDNA polymerase (Clontech Laboratories); 1X Titanium TaqPCR buffer (Clontech Laboratories); and a 10 mM concentration each of dGTP, dATP, dTTP, and dCTP (Life Technologies). Amplification was performed in a GeneAmp 9700 (Applied Biosystems) with initial heat activation at 94°C for 5 min followed by 36 cycles (GAPDH) or 37 cycles (CB1) of 94°C for 30 s, 68°C for 1 min, and 72°C for 30 s. The dsDNA was subsequently electrophoresed in 3% NuSieve and 1% Seakem agarose in 40 mM Tris, 20 mM acetic acid, and 1 mM EDTA (TAE buffer). The dsDNA was visualized by staining in 1X SYBR-Gold in TAE buffer (Molecular Probes) and documented on a GelDoc 2000 System (Bio-Rad).
For flow cytometry, cells were incubated with 10 μg/ml 2.4G2 anti-Fc receptor mAb (BD Pharmingen) followed by labeling with directly conjugated mAb. Cells were labeled with FITC-conjugated anti-CD86 (GL-1), PE-conjugated anti-MHC-II (clone M5/114), biotinylated anti-CD40 (3/23), and allophycocyanin-conjugated CD11c (clone HL3) mAbs (all from BD Pharmingen). Labeling with biotinylated mAbs was visualized with PerCP-streptavidin (BD Pharmingen). Minimal background staining was observed using control FITC-conjugated, PE-conjugated, biotinylated, mouse and rat IgG2a and IgG2b Abs and allophycocyanin-conjugated hamster IgG Ab (all from BD Pharmingen). To visualize dead cells, cells were incubated with 7-aminoactinomycin D (Sigma-Aldrich). All labeling was performed on ice for 30 min in PBS containing 2% FCS, 5 mM EDTA, and 0.01% sodium azide. Flow cytometry analysis was performed with a FACSCalibur (BD Biosciences) on 50,000 cells and analyzed using CellQuest software (BD Biosciences).
Cell suspensions were made from the spleens of BALB/c mice. Splenocytes were depleted of CD11c+ cells by MACS using anti-CD11c magnetic microbeads and a MidiMACS separation column. CD4+ T cells were subsequently purified from the CD11c-depleted splenocytes by anti-CD4 magnetic microbeads and a MidiMACS separation column following the manufacturer’s protocol. Primary MLR were set up in flat-bottom 96-well plates (BD Falcon) with 1.5 × 105 responder BALB/c CD4+ T cells per well and 3 × 102, 1 × 103, 3 × 103, or 1 × 104 allogeneic stimulator cells. All stimulator cells were irradiated with 2000 rad. MLR were incubated for a total of 96 h in humidified, 5% CO2 incubators at 37°C. [3H]thymidine (0.5 μCi) was added to each well 8 h before termination. The culture medium was composed of RPMI 1640 (Invitrogen) with sodium pyruvate and supplemented with 50 μM 2-ME, 100 U/ml penicillin, 100 g/ml streptomycin, 2 mM l-glutamine, and 10% FCS (all from Invitrogen).
Clampfit software (Axon Instruments) was used for measurement of input resistance and peak K+ currents. The mean and SD or SEM were calculated for all groups of data. The results were transferred to Excel software (Microsoft) for statistical analysis. Time course data were converted into a single point area under the curve for a time period of 3–10 min, and the area under the curve was used to test for significance between groups. Data from several time points were analyzed for significance using repeated measures ANOVA with a Bonferroni posthoc test. Paired Student’s t test was used to calculate p values for all single point statistical comparisons.
KV channels are the predominant voltage-dependent ion channels in bone marrow-derived murine DC
CD11c+ DC were expanded from bone marrow with GM-CSF and initially tested with whole cell patch-clamp analysis for the presence of ion channels of different types to establish protocols for investigating their regulation and involvement in DC function. In response to voltage steps increased by 5 mV increments from Vhold −70 mV to +25 mV, mainly outward currents were observed (Fig. 1,A). Outward currents were highly attenuated by replacing sodium chloride in the extracellular solution with TEA (Fig. 1,A), which blocks KV channels. At a concentration of 145 mM TEA, outward currents were reduced by 100% (Fig. 1,A), and when used at 15 mM, outward currents were reduced by 50% (data not shown). The blocking effect of TEA was reversible as extensive washing with the control solution restored the outward currents (Fig. 1,A). Furthermore, because the intracellular solution contains a 28-fold higher K+ concentration than the extracellular solution, but approximately the same Cl− concentration, the only driving force for outward current would be K+. In 87% of the cells (n = 166), the predominant voltage-activated currents were outward and, therefore, specific for K+. The remaining cells either did not respond to voltage steps (6%) or had an inward Ca2+ current followed by an outward Ca2+-dependent current, which must also have been K+ (7%) (Fig. 1 B). Thus, K+ channels were the predominant voltage-dependent ion channels in our bone marrow-derived murine DC.
DC express the CB1 receptor
Next, we wished to determine whether DC express the CB1 receptor, a candidate target for regulating the activity of KV channels. To detect CB1 receptor expression on DC, we used confocal laser microscopy and observed immunopositive punctuate staining by a specific Ab against the N terminus (Fig. 2,A), but not against the C terminus (data not shown), of the CB1 receptor. In Western blots using the same Ab, an 87-kDa immunoreactive band corresponding to CB1 receptor protein was present and aligned with the control CB1 receptor protein from mouse brains (Fig. 2 B). Also, expression of the CB1 receptor, which was detected by using PCR, underwent no change after LPS stimulation at neither a protein nor an mRNA level (data not shown). These observations prompted us to perform a more detailed analysis of KV channel activity in DC in response to CB1 receptor activation.
CB1 receptor signaling attenuates voltage-activated outward K+ currents in DC
To test the impact of CB1 receptor activation on the KV channel function in DC, we applied the CB1 receptor agonist ACPA, a structural analog to anandamide, and recorded the effects on voltage-activated responses. ACPA significantly reduced the outward currents in DC (Fig. 3,A). The effect of ACPA on outward currents in DC compared with control DC was monitored over the test voltage step range (Fig. 3,B). Fig. 3,C shows a prolonged attenuation of outward currents in DC for up to 30 min after application of ACPA. In contrast, no such attenuation occurred in control DC when only vehicle was used (Fig. 3,C). At 5 and 10 μM concentrations of ACPA, there was a mean reduction of 30–35% of outward currents compared with initial values (Fig. 3 D). Statistical analysis (ANOVA) followed by the Bonferroni posthoc test for multiple comparisons reveled that ACPA, compared with vehicle, significantly attenuated (p < 0.01) KV channel function in DC.
To further test whether the influence of cannabinoids on voltage-activated outward currents was mediated by CB1 receptors in DC, we measured the effect of ACPA on voltage-activated outward K+ currents after DC had been treated with AM251, a specific CB1 receptor antagonist. AM251 completely abolished the attenuating action of ACPA on KV channel function (Fig. 3, E and G). As previously shown, the CB1 receptor can couple to G proteins (2), including the Gi/o subtypes (36); therefore, we used PTX, an inhibitor of Gi/o proteins, as an additional test for a G protein-mediated block of the ACPA-induced signaling in DC. Pretreatment of DC with PTX abolished the effect of ACPA on KV channel function (Fig. 3, F and G). Together, these findings support a role for the cannabinoid signaling system in the regulation of KV channel function in DC.
The endogenous cannabinoid, anandamide, attenuates voltage-activated outward K+ currents in DC
When the impact of anandamide on DC was monitored over the test voltage step range, as occurred with the exogenous cannabinoid ACPA (Fig. 3,B), we found that anandamide also reduced the voltage-activated outward currents in DC (Fig. 4, A and B). As Fig. 4,C shows, a prolonged attenuation of outward currents in DC occurred for up to 30 min after application of anandamide, similar to that with ACPA (Fig. 3,C). In contrast, outward currents were not attenuated when vehicle only was used on control DC (Fig. 4,C). Application of three different concentrations of anandamide indicated a dose-dependent reduction of outward currents in DC (Fig. 4 D). Statistical analysis (ANOVA) followed by the Bonferroni posthoc test for multiple comparisons revealed that 1 and 10 μM anandamide, compared with vehicle significantly reduced (p < 0.01) KV channel function in DC.
Attenuation of outward K+ currents by anandamide involves CB1 receptor signaling in DC
To confirm that anandamide’s blockade of KV-channel function was CB1 receptor-mediated in DC, we pretreated these DC with the specific CB1 receptor antagonist AM251, which partially inhibited the effect of anandamide on voltage-activated outward K+ currents (Fig. 5,A). In addition, pretreatment of DC with PTX before the addition of anandamide resulted in complete inhibition of outward current attenuation (Fig. 5,B). Fig. 5,C shows the mean response recorded over 10 min for each of the conditions, including controls (vehicle only). Furthermore, each of the conditions was tested for significant differences using the Bonferroni posthoc test for multiple comparisons. This assessment revealed that treatment with anandamide significantly reduced outward currents (p < 0.01) in DC compared with vehicle, whereas no significant reduction of outward currents was observed in DC treated with anandamide in combinations with PTX or the CB1 receptor antagonist. The mean responses recorded over 3–10 min for each of the conditions in addition to one higher concentration of anandamide (10 μM) are summarized in Fig. 5,D. Taken together, our findings demonstrate that both exogenous and endogenous cannabinoids can regulate KV channel function in DC by activation of the CB1 receptor. This led us to hypothesize that attenuation of K+ outward currents may be one mechanism by which cannabinoids modulate DC immune function. To approach this issue directly, we used first the ion channel blocker TEA (see Fig. 1,A) and second, the Kv channel-modulating cannabinoids ACPA and anandamide (see Figs. 3 and 4) to attenuate KV channels in DC.
Blocking of KV channels alters the immunostimulatory function of DC
Bone marrow-derived DC were stimulated with LPS followed by blocking of KV channels with TEA. LPS-matured DC are characterized by relatively high surface expression of MHC-II (Fig. 6, A and B). However, LPS-matured DC incubated with TEA showed a marked decrease in the surface expression of MHC-II (Fig. 6, A and B). The LPS-stimulated DC exposed to the KV channel blocking reagent TEA exhibited a markedly immature phenotype with a predominantly intracellular location of MHC-II (Fig. 6,B). In contrast, blocking of KV channels had no effect on cell surface expression of the costimulatory molecules CD86 and CD40. These molecules remained up-regulated on LPS-stimulated DC in response to TEA (Fig. 6, C and D). This outcome suggests that blocking of KV channels induces changes in mature DC to selectively redistribute distinct surface molecules important for T cell activation.
Bone marrow-derived DC efficiently stimulated naive T cells in MLRs, particularly after exposure to maturing agents such as LPS (Fig. 6,E). In contrast, LPS-stimulated DC treated with the KV channel blocking reagent TEA were less efficient in stimulating an MLR (Fig. 6,E). At a concentration of 15 mM TEA, the T cell-stimulatory capacity of DC was reduced by ∼35%, and when used at 45 mM, the degree of stimulation was reduced by almost 75% (Fig. 6,F). Similar to LPS-matured bone marrow-derived DC, splenic DC stimulated naive T cells in an MLR (Fig. 6,G). In addition, treatment of splenic DC with the KV channel-blocking reagent TEA reduced the T cell stimulatory capacity of DC in an MLR (Fig. 6 G). Collectively, these findings suggest that blocking KV channels, which reduced voltage-dependent outward currents, altered the Ag-presenting and T cell-stimulatory capacity of LPS-matured DC as well as splenic DC.
Attenuation of KV channels by cannabinoids affects immunostimulatory function of DC
Next, we tested whether the cannabinoids, which attenuated KV channels and reduced voltage-dependent outward currents (Figs. 3 and 4), also affect the immune stimulatory characteristics of DC. Splenic DC incubated with ACPA showed a reduction of MHC-II cell surface expression, similar to that observed with TEA. Additionally, primary splenic DC treated with the cannabinoids, ACPA and anandamide, were less efficient in stimulating an MLR as compared with control splenic DC (Fig. 7,B). Cannabinoids also reduced the capacity of LPS-matured bone marrow-derived DC to stimulate an MLR (Fig. 7,C). Treatment of splenic DC with 10 μM ACPA or 1 μM anandamide reduced the T cell stimulatory capacity of DC to the same degree as treatment with 15 mM TEA (Fig. 7,D). To exclude the possibility that treatment of DC with ACPA, anandamide, or TEA had an effect on DC survival, the proportion of dead cells in DC cultures was analyzed. Cultures of bone marrow-derived LPS-matured DC treated with ACPA, anandamide, or TEA showed no accumulation of dead cells as compared with control DC (Fig. 7 E). Similar results were observed with splenic DC (data not shown). In addition, trypan blue staining of bone marrow-derived normal or LPS-matured DC, and splenic DC treated with the different reagents revealed no difference in total number of viable cells. In summary, our findings suggest that cannabinoids that attenuate KV channel function in DC, affects DC in ways leading to impaired ability to stimulate T cells.
Because of the widespread interest in cannabinoid-based drugs that have therapeutic potential for the treatment of many human diseases, this study was initiated to extend the limited information on the mechanisms by which cannabinoids influence immunosuppression and anti-inflammatory activities. We first showed that K+ currents are the predominant outward voltage-activated currents in murine bone marrow-derived DC. We then found that selective activation of CB1 receptors in DC, using either exogenous or endogenous cannabinoids, attenuates the voltage-activated K+ currents in a time-dependent manner. This attenuation of KV channel function in response to the selective CB1 receptor agonist, ACPA, was completely abolished in the presence of a CB1 receptor antagonist. Similarly, uncoupling of G protein signaling by PTX abolished the ACPA-mediated attenuation of KV channel function. Together these findings demonstrate that cannabinoid-regulated KV channel function in DC is mediated via CB1 receptor signaling, and not by other noncannabinoid-like receptors (37, 38, 39, 40). Importantly, we demonstrated that cannabinoids attenuated KV channel function in DC, reduce the expression of MHC-II surface molecules, and decrease the capacity to induce T cell proliferation.
Treatment with cannabinoids has been shown to suppress both innate and adaptive immunity and their therapeutic potential is being evaluated on the basis of their anti-inflammatory activities (10). From those studies, cannabinoids evidently have several mechanisms of action for the attenuation of immune-mediated diseases. Studying DC biology reveals numerous steps that provide opportunities for pharmacological manipulation of immune responses. Indeed, substantial evidence indicates that many immune suppressive agents target DC (28). Furthermore, elucidation of the mechanisms underlying cannabinoid-mediated modulation of DC function is likely to facilitate the development of future pharmacotherapies. Our results, showing that both exogenous and endogenous cannabinoids modulate KV channel function in DC via a CB1 receptor-mediated pathway, depict a novel mechanism by which cannabinoids regulate DC and may contribute to cannabinoid-induced immunosuppression. Given the important role of KV channels in T cell function (15), the concept that DC are similarly regulated by KV channel modulation is compelling, particularly in light of the recent interest in DC as targets for immunosuppressive drugs (28). Cannabinoid-mediated regulation of neurons via K+ and Ca2+ channels via synaptic transmission has also been proposed (reviewed in Ref. 41).
Our initial findings revealing the presence of voltage-dependent K+ currents in murine bone marrow-derived DC is consistent with previous findings using murine splenic DC (13) as well as human monocyte-derived DC (42). Ion currents through KV channels determine the resting membrane potential that prevents depolarization, thus regulating Ca2+ influx and the many functions that involve Ca2+ signaling, including production and release of inflammatory mediators as well as cell proliferation and differentiation (43). Although, KV channel function has been implicated as a regulator of DC maturation and T cell stimulatory capacity (42), the mechanism is ill defined. Cannabinoids act at two distinct types of G protein-coupled receptors, CB1 (44) and CB2 (45). CB1 receptor is highly expressed in the CNS, but is also found in some peripheral tissues, whereas CB2 receptor is found mainly outside the CNS, particularly in association with the immune system. At an intracellular level, activation of both CB1 and CB2 receptors alters cAMP levels by inhibiting stimulus-induced adenylate cyclase (3, 46, 47). Because human DC express the cannabinoid receptors CB1 and CB2, as determined by mRNA expression (48), these receptors are potential targets for manipulation when one wishes to modulate KV channels and thereby direct DC function. Regulation of KV channels by cannabinoids is intriguing because DC function is inhibited by the same immunosuppressive and anti-inflammatory drugs, i.e., FK-506, rapamycin, and cyclosporine (28) that inhibit KV channels in lymphocytes (15). Thus, drugs that regulate KV channels can modulate DC function toward therapeutic immunosuppression. Our finding that KV channels couple through CB1 receptor is consistent with the fact that K+ and Ca2+ channels are often modulated by activation of CB1 receptor in neurons. Therefore, our observations indicate that, by modulating functions of KV channels, cannabinoids may modulate DC function in a less destructive way than the recently described NF-κB-dependent apoptosis in murine DC triggered by activation of cannabinoid receptors (49). Importantly, Do and coworkers (49) reported that a simultaneous activation of CB1 and CB2 receptors or a relatively high concentration of anandamide (20 μM) was required to induce cannabinoid-mediated apoptosis in DC.
Anandamide also acts on receptor types other than cannabinoid receptors (37). For instance, Sancho and coworkers (40) showed that anandamide can inhibit NF-κB in a cannabinoid receptor-independent manner in 5.1 cells (Jurkat T lymphocyte-derived clone) and A549 cells, which do not express the anandamide-sensitive Transient Receptor Potential Vanilloid type 1 channel (39). Furthermore, anandamide inhibits voltage-gated 1.2 K+ channels in brain slices, N-type Ca2+ channels, and G protein-coupled inwardly rectifying KV channels in mammalian neurons through a receptor-independent, PTX-insensitive mechanism (6, 38). Our finding that the attenuation of voltage-gated K+ currents by anandamide was abolished by PTX blockade of G protein signaling (Fig. 5) supports the likelihood that a G protein-coupled pathway in DC is the primary instigator of this effect on the KV channels. However, activation of the CB2 receptor had no effect on the K+ current response (data not shown), and the CB1 receptor antagonist AM251 substantially, but not completely, blocked the anandamide-induced attenuation of voltage-gated K+ currents in DC (Fig. 5). Future studies, such as screening for noncannabinoid-like receptors on DC, may identify additional target receptors responsible for the PTX-sensitive non-CB1 receptor-mediated modulation of KV channel function in DC stimulated with anandamide.
Voltage-gated K+ channels are potential targets for immunomodulation because they are present on lymphocytes, macrophages, and DC (13, 14, 15, 29, 42). In T lymphocytes, KV 1.3 channels are inactivated by hypoxia (50) and inhibited by the immunosuppressors cyclosporine, rapamycin, and FK-506 (15). Inhibition of these channels reduces T cell proliferation and activation and redirects cytolytic activity and cytokine production (51). Previously the blocking of KV channels in human monocyte-derived DC suppressed LPS-induced up-regulation of DC markers of maturation, i.e., the costimulatory molecules CD83, CD80, CD86, and the proinflammatory cytokine IL-12 (42). In the current study, we observed that blocking KV channels in LPS-matured DC reduced the cell surface expression of MHC-II molecules and decreased the capacity of DC to induce T cell proliferation. This finding along with the reported reduction of LPS-stimulated up-regulation of maturation markers in human blood-derived DC caused by blocking KV channels (42) add to the number of targets on which KV channel inhibitors, such as exogenous and endogenous cannabinoids, can exert immunosuppressive effects.
Recently, it was found that anandamide is capable of providing feedback to control activated microglia and promote neuroprotection in the CNS (52). Interestingly, peripheral neurons also express anandamide, and areas of direct communication between neurons and immune cells lie in both primary (thymus and bone marrow) and secondary (spleen, tonsils, lymph nodes and Peyer’s patches) tissues of the immune system, as well as airway epithelium and skin (53). Functionally, neurotransmitters including catecholamines and acetylcholine as well as neuropeptides including calcitonin gene-related peptide, vasoactive intestinal peptide, somatostatin, substance P, and pro-opiomelanocortin-derived peptides have been shown to modulate immune and inflammatory responses (54). Furthermore, evidence indicates that direct innervations may control immune responses (55), and the term “neuroimmunological synapse” has been proposed for contacts between neurons and APCs (56). Langerhans cells, a subtype of DC that were originally thought to originate in the nervous system because of their close contacts with nerve fibers (57), have recently been linked closely to calcitonin gene-related peptide/substance P-containing fibers in the skin (presumably nociceptive neurons) of humans (58, 59, 60), primates (58), and rodents (58, 61) as well as in the viscera (62). These same calcitonin gene-related peptide/substance P-containing primary afferent fibers express and release the immunoregulatory endocannabinoids. Specifically, Ahluwalia et al. (63) showed that stimulation of capsaicin-sensitive primary sensory neurons induces release of anandamide. Our present finding that cannabinoids can modulate the activity of voltage-dependent K+ currents in DC, therefore, tallies with the hypothesis that neuroimmunological interactions may occur at the level of the nerve fiber-immune cell interface.
In conclusion, our results show that the KV channel function in DC can be modulated by both exogenous and endogenous cannabinoids through a CB1 receptor in a PTX-sensitive manner. This discovery presents a mechanism by which cannabinoids can regulate DC and may contribute to cannabinoid-induced immunosuppression. Given the important role of KV channels in T cell function, the concept that DC are similarly regulated by KV channel modulation is interesting in the light of DC as targets for immunosuppressive drugs (28). These results also indicate a potential mechanism by which peripheral neurons may influence the function of DC through release of the endogenous cannabinoid, anandamide. The down-modulation of MHC-II on DC observed in response to blocking of KV channels may explain some of the immunosuppressive effects mediated by cannabinoids. Because CB1 receptor signaling attenuates KV channel function in DC, the CB1 receptor can be a potential target to regulate DC function, preventing DC-mediated inflammation and inducing beneficial immunosuppression.
We thank P. Löw, J. Brask, T. Harkany, and R. Wallin for technical support and advice. We are also grateful to Dr. K. Mackie (University of Washington, Seattle, WA) for providing the anti-CB1 receptor Ab.
The authors have no financial conflict of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
This work is supported by grants from the Swedish Research Council, the Swedish Foundation for Strategic Research, and the Karolinska Institutet. This work was also supported by Grants 4480 from the Swedish Research Council and FP-6-19090 from European Commission.
Abbreviations used in this paper: KV, voltage-gated potassium; DC, dendritic cell; MHC-II, MHC class II; TEA, tetraethyl-ammonium; ACPA, arachidonylcyclopropylamide; PTX, pertussis toxin.