The precise role of cannabinoid receptors (CB)1 and CB2, as well as endogenous ligands for these receptors, on immune cells remains unclear. In the current study, we examined the effect of endogenous and exogenous cannabinoids on murine bone marrow-derived dendritic cells (DCs). Addition of Δ9-tetrahydrocannabinol (THC), a major psychoactive component found in marijuana or anandamide, an endogenous cannabinoid, to DC cultures induced apoptosis in DCs. DCs expressed CB1 and CB2 receptors and the engagement of both receptors was necessary to trigger apoptosis. Treatment with THC induced caspase-2, -8, and -9 activation, cleavage of Bid, decreased mitochondrial membrane potential, and cytochrome c release, suggesting involvement of death-receptor and mitochondrial pathways. DCs from Bid-knockout mice were sensitive to THC-induced apoptosis thereby suggesting that Bid was dispensable. There was no induction of p44/p42 MAPK, p38 MAPK, or stress-activated protein/JNK pathway in THC-treated DCs. However, THC treatment induced phosphorylation of IκB-α, and enhanced the transcription of several apoptotic genes regulated by NF-κB. Moreover, inhibition of NF-κB was able to block THC-induced apoptosis in DCs. Lastly, in vivo treatment of mice with THC caused depletion of splenic DCs. Together, our study demonstrates for the first time that endogenous and exogenous cannabinoids may suppress the immune response through their ability to induce apoptosis in DCs.
Cannabinoids are compounds derived from the Cannabis sativa (marijuana) plant, as well as produced endogenously in the brain and by immune cells (1). Cannabinoids mediate their effect through cannabinoid receptors (CB),3 designated CB1 and CB2, which belong to a superfamily of G-protein-coupled receptors (1). CB1 receptors are expressed at high levels in CNS, where they regulate psychoactivity. CB1 receptors are also expressed on immune cells. In contrast, the CB2 receptors are primarily expressed on immune cells and do not contribute to the psychoactivity (1). The presence of endogenous CB-ligand systems in immune cells suggests that they may play a critical physiological role, the precise nature of which remains to be characterized.
Cannabinoids can decrease the immune response and impair the host defense against viral, bacterial, or protozoan infections as discussed in several reviews (2, 3, 4). Cannabinoids have also been widely used in the treatment of pain and inflammation (5). Moreover, preliminary studies have shown the possible use of cannabinoids in the treatment of autoimmune diseases such as multiple sclerosis (6). Although the precise mechanism of cannabinoid-induced immune dysfunction is not clear, studies have indicated that cannabinoids may cause dysregulation in cytokine production by immune cells (7). Recent studies from our lab demonstrated that Δ9-tetrahydrocannabinol (THC) can trigger apoptosis in vivo in thymocytes and splenocytes, which may account for immunosuppression (8).
DCs are the most potent APCs that play a crucial role in the presentation of self and non-self Ags to T cells (9). We demonstrate for the first time that THC and endocannabinoids such as anandamide can induce apoptosis in DCs through activation of CB1 and CB2 receptors. These studies provide the basis for understanding the mechanism by which THC triggers immunosuppression and mediates anti-inflammatory properties. Moreover, the expression of CB1 and CB2 receptors on DCs and their modulation during DC activation suggests that the endocannabinoid systems may play a critical role in the regulation DC growth and maturation.
Materials and Methods
Adult (6–8 wk of age) female C57BL/6 mice were purchased from The Jackson Laboratory (Bar Harbor, ME). Bid-knockout (KO) mice were generated as described elsewhere (10).
THC and anandamide were obtained from the National Institute on Drug Abuse (Rockville, MD) and dissolved in DMSO (Sigma-Aldrich, St. Louis, MO). CB1 antagonist, SR141716A, and CB2 antagonist, SR144528, were obtained from Sanofi-Synthélabo (Montpellier, France) and were used as described (8, 11).
Dendritic cell (DC) culture
DCs were cultured from mouse bone marrow cells with murine recombinant GM-CSF (5 ng/ml; BD Pharmingen, San Diego, CA) for 6 days (12). In brief, bone marrow was obtained from the femurs and tibias, depleted of RBCs, and placed on petri dishes with RPMI 1640 supplemented with 10% FCS, 10 mM HEPES, 1 mM glutamine, and 50 μg/ml gentamicin. The CD4+CD8+ T cells, as well as B220+ B cells, were depleted by using specific mAbs and complement. The cells (1 × 106 per well) were placed in 24-well culture plates supplemented with GM-CSF. The cultures were fed every 2 days to remove nonadherent monocytes by aspirating 75% of the medium and adding fresh medium with GM-CSF. The DCs activated with LPS (10 μg/ml; Sigma-Aldrich) were used as mature DCs (13). Phenotypic characterization of DCs revealed that they expressed CD40, CD80, CD86, and MHC class II molecules and exhibited the characteristic dendrites.
Detection of cell viability and apoptosis in DC cultures
DCs were placed in 24-well plates in serum-free RPMI 1640, devoid of any supplementation with serum or serum proteins, and in the presence or absence of various concentrations of THC for 2 h. The cells were harvested, washed twice with PBS, and analyzed for the induction of apoptosis in three ways: 1) Annexin V/propidium iodide (PI) method, 2) TUNEL assay using FITC-dUTP, and 3) 3,3′-diehexiloxadicarbocyanine iodide (DiOC6) to detect altered mitochondrial membrane potential (ΔψM). The details of these assays have been previously described (11, 14). In addition, cell viability was also determined in DCs cultured with THC for ∼3 h, using trypan blue dye exclusion.
To detect anandamide-induced apoptosis, DCs were cultured as described above. Various concentrations of anandamide were added to the cells with or without various concentrations of methyl arachidonyl fluorophosphonate (Sigma-Aldrich) as fatty acid amide hydrolase (FAAH) inhibitor and incubated for ∼4 h. The cells were harvested, washed twice with PBS, and analyzed for the induction of apoptosis using a TUNEL assay. In some experiments, various concentrations of CB1 or CB2 antagonists were added to the DCs (see below).
Effect of CB1 or CB2 receptor antagonists on survival of DC in the presence of THC
Cells were preincubated with various concentrations of CB1 (SR141716A) or CB2 (SR144528) antagonists for 30 min at 37°C and then the cells were treated with either DMSO or 5 μM THC for 2 h (11). The cell viability was determined using trypan blue dye exclusion and the induction of apoptosis was measured using annexin V/PI or TUNEL as described above.
Role of G-protein-coupled receptor in THC-induced apoptosis
DCs were preincubated with various concentrations of pertussis toxin (PTX, 50–100 ng/ml; List Biological Laboratories, Campbell, CA) for overnight at 37°C. On the next day, the cells were thoroughly washed with 10% RPMI 1640 then treated with either DMSO or 5 μM THC for 2 h and analyzed for apoptosis.
Detection of CB1 and CB2 receptor expression by semiquantitative RT-PCR
To detect CB1 and CB2 receptor gene expression, RT-PCR was conducted as described earlier (11).
Total RNA was prepared as described above and cDNA microarray analysis was conducted using a panel of 96 key genes involved in apoptosis as described by the manufacturer (Super Array, Bethesda, MD). The level of expression or induction of a gene in control and THC- or N-p-tosyl-l-phenylalanine chloromethyl ketone (TPCK)-treated cells was normalized and the changes in gene expression following different treatments were presented as a fold increase or decrease when compared with expression profiles seen in the control group.
Western blot analysis
Cells were lysed by repeated freeze/thawing,and protein concentration in the supernatant was determined using the bicinchonic acid (BCA) protein assay kit (Pierce, Rockford, IL). Total protein (15–40 μg) was separated by SDS-PAGE and transferred to a membrane. Blots were blocked with 5% (w/v) nonfat dry milk in TBS-T (0.5 M NaCl, 20 mM Tris-HCl, 0.1% (v/v) Tween 20, pH 7.6) and probed with the primary Ab. Next, blots were washed and incubated with HRP-conjugated secondary Ab and visualized with an ECL reagent (Amersham Biosciences, Piscataway, NJ).
p44/p42 MAPK, p38 MAPK, or stress-activated protein kinase (SAPK)/JNK inhibition
Cells were preincubated with various concentrations of PD98059, SB203580 (Calbiochem, San Diego, CA), or wortmanin (Sigma-Aldrich) for 30 min at 37°C and then the cells were treated with either DMSO or 5 μM THC for 2 h and analyzed for apoptosis.
Cells were preincubated with various concentrations of TPCK (Sigma-Aldrich) for 30 min at 37°C and then the cells were treated with either DMSO or 5 μM THC for 2 h and analyzed for apoptosis.
In vivo injection of THC-C57BL/6 mice were injected with a single dose of THC (1, 10, 25, or 50 mg/kg body weight, i.p.) or the vehicle. The spleens were harvested 24 h after THC injection and a single-cell suspension was prepared, as described elsewhere (8, 14). Next, Fc receptors were blocked using the Fc-blocking reagent (BD Pharmingen), the cells were double-stained with FITC-conjugated anti-CD11c and PE-conjugated anti-MHC class II mAbs (BD Pharmingen) and analyzed using a flow cytometer.
The statistical comparisons between different treatment groups were conducted using a Student’s t test and differences with p < 0.05 were considered to be significant. All experiments were repeated at least three times.
THC induces apoptosis in DCs
To study the effect of THC on DCs, various concentrations of THC such as 1, 5, 10, and 20 μM or the vehicle (DMSO) were added to bone marrow-derived DCs in serum-free medium. THC treatment changed the morphology of DCs dramatically as shown in Fig 1⇓A, where most of them lost their typical dendrites and some of them showed blebbed surface. As shown in Fig. 1⇓B, 5 μM and greater concentrations of THC caused a dramatic decrease in the viability of DCs within 3 h of incubation. Next, we used a combination of annexin V and PI which detects apoptotic (annexin+ only), late apoptotic/necrotic (annexin+PI+) and necrotic (PI+ only) cells (8). DCs treated for 2 h with 5 μM or greater concentrations of THC showed a significant increase in early apoptotic and late apoptotic/necrotic cells (Fig. 1⇓C). To further confirm THC-induced apoptosis, we used DiOC6 to detect ΔψM loss (15). The data shown in Fig. 1⇓D demonstrated that THC induced a loss in ΔψM at all concentrations tested when compared with the vehicle controls. Taken together, these data suggested that THC treatment reduced cell number by inducing apoptosis in murine bone marrow-derived DCs. It should be noted that THC was also able to induce apoptosis in DC in the presence of serum (data not shown). However, we used serum-free medium to prevent the interference from endocannabinoids present in the serum and the possible interactions between THC and serum proteins (16). When we tested DCs at 20 μM of THC in RPMI 1640 supplemented with 5% FBS, we observed >30% of apoptosis after 3 h and >60% apoptosis after 18 h, while the vehicle-treated cells showed <5% and <10% apoptosis respectively (data not shown).
The effect of CB1 and CB2 receptor antagonists on DC survival/apoptosis
THC binds to both CB1 and CB2 with similar affinities. To further address the involvement of CB1 and CB2 receptors in THC-induced apoptosis of DCs, we tested the ability of well characterized CB1 and CB2 receptor antagonists to block THC-induced apoptosis. Addition of various concentrations of CB1 (SR141716A) or CB2 (SR144528) or SR141716A + SR144528 receptor antagonists to DC cultures caused minimal or no significant apoptosis/loss of cell viability (Fig. 2⇓). Addition of 5 μM THC alone to DC cultures caused 53.8% apoptosis (Fig. 2⇓) and furthermore, addition of either SR141716A or SR144528 to DC cultures treated with THC caused a dose-dependent inhibition in THC-induced apoptosis. Moreover, the addition of a combination of SR141716A + SR144528 was more effective in blocking THC-induced apoptosis than either of these compounds independently. The fact that either SR141716A or SR144528 alone, when tested at higher concentrations, could completely reverse the apoptosis induced by THC, suggested that simultaneous ligation of both CB1 and CB2 receptors may be necessary to induce apoptosis.
Role of G proteins in THC-induced apoptosis in DCs
Because CBs are coupled to PTX-sensitive Gi/o proteins, we tested the effect of PTX on induction of apoptosis by THC. To this end, DCs were preincubated with or without 50 ng/ml PTX overnight and then incubated with 5 μM THC for 2 h and the cells were analyzed for apoptosis using annexin/PI staining. The data shown in Fig. 3⇓ demonstrated that PTX was able to partially inhibit the induction of apoptosis by THC.
Effect of an endogenous cannabinoid on DCs and the role of CB1 or CB2 receptors
To investigate whether endogenous cannabinoids such as anandamide could induce apoptosis in DCs, various concentrations of anandamide were added to DCs. It was observed that concentrations of 20 μM caused marked apoptosis while lower concentrations were not effective (Fig. 4⇓). The lesser efficacy of anandamide when compared with THC might be due to rapid hydrolysis of anandamide by endogenous enzymes such as FAAH (17). To prevent the activity of FAAH, various concentrations of a FAAH inhibitor were added to the culture in the presence or in the absence of anandamide and then the induction of apoptosis was determined (Fig. 4⇓A). The data showed that the addition of a FAAH inhibitor could increase the induction of apoptosis by anandamide in DCs significantly. Furthermore, anandamide-induced apoptosis was also inhibited by selective antagonists of CB1 (SR141716A) or CB2 (SR144528), suggesting that anandamide was acting through these receptors to induce apoptosis (Fig. 4⇓B).
Mature DCs are more resistant to THC-induced apoptosis when compared with immature DCs
DCs show distinct phenotypic as well as functional differences depending on their activation/maturation status. In this study, we investigated whether the maturation status of DCs could have a differential effect on THC-induced apoptosis. To this end, the DCs obtained from the bone marrow were either untreated or treated with LPS and cultured with 1, 5, or 10 μM of THC. After 2 h, the cells were analyzed by TUNEL assay to detect apoptosis. The data shown in Fig. 5⇓ suggested that DCs that were not treated with LPS representing immature DC, showed apoptosis at 5 and 10 μM but not at 1 μM of THC. These data corroborated the earlier findings (Fig. 1⇑) that THC can induce apoptosis in immature DCs at 5 μM or greater concentrations, although the sensitivity of the four assays used varied somewhat. Interestingly, DCs activated with LPS were found to be more resistant to apoptosis as indicated by their inability to undergo apoptosis at 5 μM THC and showed sensitivity only at 10 μM concentrations (Fig. 5⇓, A and B). RT-PCR analysis of CBs revealed that murine bone marrow-derived DCs expressed both CB1 and CB2 receptors mRNA (Fig. 5⇓C). When DCs were activated with LPS, they were shown to modestly down-regulate the expression of CB1 and to a lesser extent, CB2 receptors.
Mechanism by which THC triggers apoptosis in DCs
We next determined whether THC-induced apoptosis was mediated by caspases. Western blot analysis of DC lysates following THC treatment showed that THC treatment induced cleavage of caspase-8, -9, and -2 within 15 min of exposure (Fig. 6⇓A). Furthermore, cleavage of Bid was seen at 30 min. Also, cytochrome c was detected at all time points tested. These data suggested that THC triggers apoptosis in DC using both death receptor (DR) and mitochondrial pathways and possibly involving cross-talk. When caspase inhibitors were used to test their ability to inhibit THC-induced apoptosis, addition of pan-caspase inhibitor blocked the apoptosis completely in a dose-dependent fashion (Fig. 6⇓B). Moreover, addition of caspase-8 or -9 inhibitors caused partial inhibition of apoptosis in DCs. These data confirmed Western blotting data, suggesting the involvement of both intrinsic and extrinsic pathways in DCs following THC treatment.
To further address the cross-talk between these two pathways of apoptosis, we tested whether the cleavage of Bid seen in Fig. 6⇑A would be essential for THC-induced apoptosis in DCs. To this end, we generated bone marrow-derived DCs from Bid-KO or wild-type mice and treated them with 5 μM THC or the vehicle (DMSO) and induction of apoptosis was measured. The data showed that DCs with Bid deficiency underwent apoptosis to the same extent as the wild-type DCs, suggesting that Bid cleavage is not a critical step for induction of apoptosis in DCs (Fig 6⇑C).
Involvement of p44/p42 MAPK, p38 MAPK, or SAPK/JNK pathways in DCs following THC treatment
To further study the down-stream pathways evoked in DCs following THC treatment, Western blot analysis for MAPK or JNK pathway was conducted. As shown in Fig. 7⇓, there was no significant change in phosphorylated or nonphosphorylated form of these pathways, suggesting that these pathways were not activated in DCs following THC treatment. Such findings were further confirmed by the data that THC-induced apoptosis could not be blocked by the addition of pathway-specific inhibitors including PD98059, SB203580, and wortmanin (data not shown).
Involvement of NF-κB pathway in DCs following THC treatment
Activation of NF-κB is considered to play an important role in cell death and apoptosis. It is also known that NF-κB is critical in DC maturation. To study the effect on NF-κB in THC-induced apoptosis in DCs, first, Western blot analysis was performed (Fig. 8⇓A). With increasing time, there was a decrease of nonphosphorylated form as well as an increase of the phosphorylated form of IκB-α, which indicated the involvement of NF-κB activation following THC treatment in DCs. In addition, when the NF-κB inhibitor, TPCK, was used in the presence of THC or vehicle treatment, 5 μM and greater concentrations of TPCK almost completely inhibited THC-induced apoptosis in DCs (Fig. 8⇓B). Next, we incubated DCs with THC in the presence or in the absence of TPCK, and then conducted Western blotting for caspase-3 (Fig. 9⇓). The data showed that the cleaved form of caspase-3 was not detected in the protein extract obtained from DCs treated with THC plus TPCK at 15 min and furthermore, at 45 min, the cleaved caspase-3 was significantly less than that seen in DCs treated with THC alone. These data first, confirmed our result showing that NF-κB inhibition rescued DCs from THC-induced apoptosis, and second, they also suggested that NF-κB activation may be up-stream of caspase activation.
THC treatment up-regulates apoptotic genes regulated by NF-κB
To investigate the mechanism by which THC-induced NF-κB activation triggers apoptosis in DCs, cDNA microarray analysis was conducted. The data showed that several apoptotic genes were significantly up-regulated in THC-treated DCs when compared with control cells and furthermore, most of the up-regulated genes showed putative NF-κB binding site in their promoter, suggesting that NF-κB may act as proapoptotic molecule in THC-induced apoptosis in DCs (Table I⇓). In addition, we observed down-regulation of some of the very same genes following TPCK treatment thereby indicating that such enhanced gene expression was indeed NF-κB-dependent (Table II⇓).
In vivo effect of THC on DCs
To test whether THC would deplete DCs in vivo, mice were injected with various concentrations of THC or vehicle and 24 h later, splenic DC population was enumerated by using double-stain for CD11c and MHC-II. THC treatment led to a decrease in the total splenic cellularity at doses of 10, 25, and 50, but not at 1 mg/kg (data not shown), similar to our previous observation (8). Furthermore, THC at all concentrations tested caused a significant decrease in both the percentage and the total numbers of class II MHC-expressing DCs found in the spleen (Fig. 10⇓).
Recently, we demonstrated that both normal and malignant immune cells that express CBs undergo apoptosis when activated through these receptors (8, 11). Based on these studies, we suggested that the THC-mediated immunosuppression in vivo may result from its ability to induce apoptosis in immune cells, particularly in thymocytes and peripheral T and B cells (8). Moreover, inasmuch as transformed immune cells also expressed CBs and were susceptible to apoptosis induced by THC and other cannabinoids, we suggested that CB agonists may serve as novel anti-cancer agents (11). Furthermore, in the above studies, we had not addressed the mechanism(s) of apoptosis. Therefore, the overall objective of this investigation was to delineate the signaling pathways of apoptosis induced by THC on an important component of the immune system, the DCs. In the current study, we demonstrated that THC-mediated apoptosis in DCs occurs, at least in part, through G-protein signaling. Also, THC mediates apoptosis in DCs through activation of several caspases including caspase-2, -8, and -9 and the release of cytochrome c. These data suggested the involvement of both extrinsic and intrinsic apoptotic pathways. The sensitivity of DCs from Bid-KO mice to THC-induced apoptosis suggested that Bid did not play a critical role and that cross-talk between these pathways was not necessary to induce apoptosis. There was no involvement of MAPK or SAPK/JNK pathways in apoptosis induction; however, we noted that treatment of DCs with THC led to NF-κB activation. cDNA microarray analysis showed the induction of several apoptotic genes regulated by NF-κB in DCs following THC treatment. Moreover, preventing NF-κB activation effectively blocked the induction of apoptosis thereby suggesting that NF-κB played a critical role in this process. Treatment of mice with THC in vivo could also decrease DC population in the spleen. Although in our earlier study (8) we had not investigated the effect of anandamide on immune cells, in the current study, we found that this endocannabinoid could also induce apoptosis in DCs. Taken together, our data demonstrate for the first time that endogenous and exogenous cannabinoids can induce apoptosis in DCs.
The physiological functions of CBs expressed on immune cells and the role played by endocannabinoids produced by leukocytes in the regulation of immune response remain unclear. A recent study reported the presence of endocannabinoids, anandamide, and 2-arachidonoylglycerol, as well as CB1 and CB2 receptors, on human DCs (18). The general consensus is that endocannabinoids act as antiproliferative agents thereby down-regulating the immune response (19). In the current study, use of CB1 and CB2 antagonists showed that either antagonist when added alone was able to significantly inhibit apoptosis in DCs. These data not only suggested the involvement of both CB1 and CB2 receptors in apoptosis but also indicated that simultaneous activation through both receptors may be necessary to induce apoptosis. It should be noted that THC-induced apoptosis in tumor cells is mediated through CB-dependent (20) or -independent pathways (21).
Apoptosis is triggered primarily through two pathways: the extrinsic pathway (DR pathway) involving caspase-8 and -10, and the intrinsic pathway (mitochondrial pathway) involving caspase-2 and -9 (22). Moreover, in type II cells, there is a cross talk between these two pathways involving Bid (23). The mechanism by which THC triggers apoptosis may involve multiple pathways based on the nature of cells. THC and other cannabinoids were shown to induce apoptosis of glioma cells via ceramide generation (24). THC also promoted the mitochondrial pathway of apoptosis in cultured cortical neurons as evidenced by cytochrome c release and activation of caspase-3 (25). Others have suggested that THC caused sphingomyelin hydrolysis in C6.9 glioma cells resulting in apoptosis (26). In addition, THC treatment of splenocytes was shown to cause a decrease in Bcl-2 mRNA and protein expression (27). In the current study, we demonstrated that THC-induced apoptosis in DCs involved both the DR and mitochondrial pathways. The cleavage of Bid suggested the possible cross-talk between the two pathways. However, the susceptibility of DCs from Bid-KO mice to THC-induced apoptosis suggested that Bid was dispensable in THC-induced apoptosis.
The CB1 and CB2 are G-protein-coupled receptors that are linked to a variety of signaling pathways including the cAMP pathway (28) and MAPK pathway (29, 30) in a PTX-sensitive manner. In addition, it was shown that THC-induced apoptosis in malignant glioma induce sustained ceramide accumulation and Raf/extracellular signal regulated kinase activation (21). However, in the current study, our data showed that THC-induced apoptosis in DC did not induce MAPK nor JNK pathways by Western blot analysis. Moreover, specific inhibitors for each pathway could not block apoptosis in DCs after THC treatment. Also, addition of PTX only partially inhibited THC-induced apoptosis in DCs. Previous studies on apoptosis induced in neurons suggested that G proteins may exert both proapoptotic (31) and anti-apoptotic effects (32). Other studies have suggested that PTX-sensitive G proteins may act as facilitators of THC-induced apoptosis in neurons, with potential effectors being members of the Bcl family of mitochondria-associated proteins (25). Although the precise signaling mechanisms involved in G-protein-mediated apoptosis in the cannabinoid system are not clear, our data are consistent with the above studies and demonstrate that PTX-sensitive G proteins may at least play a partial role in THC-induced apoptosis in DC.
NF-κB is a transcription factor that regulates inflammation, immunity, cell death, or survival (33). In general, NF-κB exists as an inactive form in cytoplasm, where it binds to IκB inhibitory proteins. Extracellular signals induce a phosphorylation of IκB, leading to its degradation by proteasome. This results in the rapid translocation of NF-κB to the nucleus where NF-κB can stimulate transcription of various NF-κB-responsive target genes (33). In general, NF-κB functions as an anti-apoptotic transcription factor, but conversely, it also has been proposed as a proapoptotic protein depending on cell type and the nature of apoptotic inducers (34). The anti-apoptotic function of NF-κB has been strongly supported by large body of literature although most of the data were generated using knockout mice (35, 36). Interestingly, a few recent reports suggest that NF-κB may contribute to induction of proapoptotic molecules such as DR6, DR4, DR5, or Fas/FasL (37, 38, 39, 40). Moreover, it was shown that NF-κB is activated and promotes cell death in focal cerebral ischemia (41). Regarding the induction of NF-κB following THC treatment, a previous study showed that cannabinol treatment inhibits IL-2 secretion by activated thymocytes, probably due to inhibition of NF-κB and CREB (42). Some studies have suggested that the immunosuppressive function of cannabinoid might be related to the inhibition of NF-κB (43). In another study, the activity of NF-κB was directly measured by the NF-κB-responsive reporter gene, which was integrated into PC12 cells. Treatment with 6-hydroxydopamine or the CB1-agonist, CP55,940, increased NF-κB activity, but it was not an integral part of the apoptotic cascade in PC12 cells because attenuated NF-κB activity was not related to the reduction in TUNEL-positive cells (44). On the contrary, in the current study, we found a potential link between the THC-induced apoptosis in DCs and the activation of NF-κB. Increased phosphorylation of IκB-α was detected following THC treatment in a time-dependent manner. Moreover, when the activity of NF-κB was inhibited by TPCK, DCs were almost completely prevented from undergoing apoptosis. It is also worthy to point out that ubiquitination-mediated proteolysis of IκB-α by the 26S proteasome is considered to be a critical step in NF-κB activation (45). The exact mechanism by which attenuated NF-κB activity was effective to prevent DCs from THC-induced apoptosis remains unclear and further study is necessary. Interestingly, it is well known that NF-κB also plays an important role in DC development as well as functional maturation (46, 47). Three general models by which NF-κB transcription factors may regulate apoptosis have been proposed, which include first, the direct regulation of genes involving anti- or proapoptosis, and second, the regulation of cell cycle, and lastly, the protein-protein interactions (34). Our cDNA microarray data showed the induction of several apoptotic genes, which may be regulated by NF-κB, in DCs following THC treatment. Moreover, the induction of some of these genes was down-regulated following TPCK, thereby suggesting that direct gene regulation by NF-κB may play a role in promoting apoptosis in DCs. Taken together, this study provides evidence for the unique role of NF-κB in DC death.
In the current study, we used GM-CSF to derive DC from the murine bone marrow cells as originally described by Inaba et al. (48). Subsequently, several reports have been published on the differentiation of DCs from human peripheral blood monocytes or CD34+ hemopoietic progenitors, using combinations of GM-CSF, IL-4, and TNF-α (49, 50). It is not known whether the DCs generated using different cytokines would exhibit differential sensitivity to THC-induced apoptosis. Cytokines such as IL-4 and TNF-α have been shown to activate NF-κB (51, 52) and inasmuch as we demonstrated that THC-induced activation of NF-κB regulates apoptosis in DCs, it would be interesting to further investigate the effect THC on DCs generated using GM-CSF and IL-4 or TNF-α. Such reports also suggest that our studies may have limitations in extrapolating the results to human DCs.
The dose of THC that induced apoptosis in vitro in the current study was found to be 5 μM or greater. The relevance of this dose to the recreational use of marijuana remains to be established. It is noteworthy that Azorlosa et al. (53) showed that marijuana smoking could lead to plasma levels as high as 1 μM in humans. Moreover, THC can be concentrated 15- to 20-fold in some tissues (54). Therefore, it may be possible to reach levels as high as 20 μM in some tissues after recreational use. It is not clear whether THC in lymphoid organs will increase by 20-fold, and additional studies are necessary in this regard. However, it should be noted that in our study, as low as 1 mg/kg of body weight of THC could deplete DCs in vivo. Moreover, 1 μM THC was able to cause significant ΔψM loss in DCs in vitro, although we did not study whether this effect on mitochondria alters DC functions. In an earlier study, rats injected with 50 mg/kg of THC were shown to exhibit 10 μM THC in the serum within 10 h of administration (55). Also in these studies, rats were given THC at 50 mg/kg 5 times a week for 2 years while mice were given as much as 500 mg/kg for a similar duration. Interestingly, despite such high doses, the rats survived for a longer time than the controls and both the rats and mice had decreased tumor incidence. Thus, additional studies are necessary to address the consequences of THC-induced apoptosis on immune functions and its relevance to marijuana abuse and pharmacological use in cancer and AIDS patients.
In the current study, we injected 1, 10, 25, and 50 mg/kg of body weight of THC into mice and found that the total spleen cellularity was decreased at 10 mg/kg or higher doses but not at 1 mg/kg, similar to our earlier studies (8). In the previous study, we also noted that 10 mg/kg of THC did not cause a significant decrease in the percentage of T and B cells and macrophages, although this dose caused a significant decrease in the total numbers of each of these subpopulations in the spleen due to an overall decrease in spleen cellularity (8). Although, in the previous study, we had not investigated the effect of THC on DCs, our data suggested that T and B cells and macrophages may exhibit similar susceptibility to THC-induced apoptosis at 10 mg/kg (8). Interestingly, in the current study, we noted that THC at all concentrations tested, including 1 mg/kg, caused a significant decrease in the percentage of DCs as well as a decrease in the total numbers of DCs in the spleen. These data suggested that DCs may be more sensitive to THC-induced apoptosis when compared with other immune cells. Clearly, additional studies are necessary to evaluate this possibility, including the mechanisms of differential susceptibility to THC. It should be noted that DCs represent a small proportion of the splenocytes and thus it is not clear whether a decrease in their number would result in suppression of an Ag-specific immune response. Such studies are necessary to understand the consequences of THC-induced depletion of DCs on immune response. In our previous study, while we had shown that anandamide could induce apoptosis in transformed immune cells (11), we had not investigated whether a similar action is demonstrable in normal immune cells (11). In the current study, we noted that anandamide was able to induce apoptosis in DCs only at 20 μM but not at lower doses. This can be explained by the fact that anandamide is rapidly hydrolyzed by enzymes such as FAAH (17). Therefore, when we used a FAAH inhibitor, even lower concentrations of anandamide (5 μM) could induce significant levels of apoptosis in DCs.
Many studies have suggested the use of THC or related cannabinoids in the treatment of autoimmune diseases. THC administered to mice at 150 mg/kg for 11 days suppressed streptozotocin-induced autoimmune diabetes in CD1 mice (56). Dexanabinol (HU-211), a synthetic nonpsychotropic cannabinoid, significantly reduced clinical symptoms of experimental allergic encephalomyelitis and suppressed the inflammation (57). There is also clinical evidence to suggest that THC-treatment may benefit multiple sclerosis patients (58). Because DCs actively participate in presenting self-Ag to trigger autoimmunity, whether the effectiveness of THC and other cannabinoids against autoimmune diseases results from their ability to induce apoptosis in DCs remains a possibility that needs further investigation.
We thank Dr. Michael Fisher for his help in the analysis of NF-κB binding sites on the promoter of various genes using MatInspector.
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.
↵2 Address correspondence and reprint requests to Dr. Prakash Nagarkatti, Department of Pharmacology and Toxicology, Medical College of Virginia Campus, Virginia Commonwealth University, Richmond, VA 23298. E-mail address:
↵3 Abbreviations used in this paper: CB, cannabinoid receptor; THC, Δ9-tetrahydrocannabinol; KO, knockout; DC, dendritic cell; PTX, pertussis toxin; PI, propidium iodide; ΔψM, mitochondrial membrane potential; SAPK, stress-activated protein kinase; TPCK, N-p-tosyl-L-phenylalanine chloromethyl ketone; FAAH, fatty acid amide hydrolase; DR, death receptor; DiOC6, 3,3[prime]-diehexiloxadicarbocyanine iodide.
- Received December 16, 2003.
- Accepted May 24, 2004.
- Copyright © 2004 by The American Association of Immunologists