Role of cannabinoid receptors in inhibiting macrophage costimulatory activity

International Immunopharmacology 4 (2004) 265 – 278 www.elsevier.com/locate/intimp

Role of cannabinoid receptors in inhibiting macrophage costimulatory activity Siriporn Chuchawankul a,1, Mika Shima a, Nancy E. Buckley b, Constance B. Hartmann a, Kathleen L. McCoy a,* a

Department of Microbiology and Immunology, MCV Station, Box 980678, Medical College of Virginia/Virginia Commonwealth University, Richmond, VA 23298, USA b Biological Sciences Department, California State Polytechnic University, Pomona, CA 91768, USA Received 3 April 2003; received in revised form 23 April 2003; accepted 16 December 2003

Abstract Delta9-tetrahydrocannabinol (THC) inhibits several immunologic functions of macrophages. THC’s impact on peritoneal macrophages to deliver costimulatory signals to a helper T cell hybridoma was investigated by T cell interleukin-2 production stimulated with immobilized anti-CD3 antibody. The drug’s inhibition of costimulatory activity depended on the macrophages. THC decreased costimulation provided by peritoneal cells elicited with polystyrene beads and thioglycollate, but the drug had no influence with macrophages elicited with thioglycollate alone. Bead administration induced CB2 mRNA expression in macrophages, while CB1 mRNA was not detected. Although inhibition was associated with functional heat-stable antigen, a costimulatory molecule, on macrophages, THC exposure did not alter cell surface heat-stable antigen expression. Inhibition by THC and anti-heat-stable antigen antibody was not additive suggesting the inhibitory mechanisms may overlap. Cannabinoid suppression was stereoselective; low affinity synthetic isomer CP56,667 did not diminish the T cell response. CB1-selective antagonist SR141716A completely reversed, and CB2-selective antagonist SR144528 partially blocked THC’s inhibition. Both antagonists appeared to behave as inverse agonists in a receptor-selective manner. Although T cells expressed a low level of CB2 mRNA, neither THC nor SR141716A affected T cell activation in a system independent of macrophages, while SR144528 was inhibitory. High affinity synthetic agonist CP55,940, but not partial agonist THC, impaired costimulation by macrophages from mice lacking CB2 receptor. Although CB1 mRNA was not detected in CB2 null macrophages, CP55,940 reversed the inverse agonist activity of SR141716A. Hence, CB2 and possibly another receptor subtype may be involved in mediating cannabinoid suppression of macrophage costimulation. D 2004 Elsevier B.V. All rights reserved. Keywords: Cannabinoids; CB receptors; Costimulation; Immunosuppression; Macrophages

* Corresponding author. Tel.: +1-804-828-2305; fax: +1-804828-9946. E-mail address: [email protected] (K.L. McCoy). 1 Present address: Chulalongkorn University, Bangkok, Thailand. 1567-5769/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.intimp.2003.12.011

1. Introduction Numerous reports have documented that cannabinoids modulate the immune system, and their influ-

266

S. Chuchawankul et al. / International Immunopharmacology 4 (2004) 265–278

ence is not limited to any particular leukocyte lineage [1]. Delta9-tetrahydrocannabinol (THC) is an extensively studied cannabinoid and is the major psychoactive component of marijuana. Investigations regarding cannabinoid-induced immunomodulation have emphasized alterations in cytokine production. For example, interleukin-2 (IL-2) secretion by T cells induced by mitogenic stimuli is diminished by THC [2,3]. Furthermore, THC depresses production of nitric oxide [4,5] and tumor necrosis factor-a [6,7] by activated macrophages. However, not all immune functions are compromised by drug exposure; some functions are actually enhanced. THC augments secretion of interleukin-1 by endotoxin-stimulated macrophages [8]. The proliferative response of B cells to mitogens is also enhanced in the presence of THC [9]. One hypothesis is that cannabinoids suppress cytokines required for cell-mediated immunity, but augment those for humoral immunity [10,11]. The diverse consequences of drug exposure may reflect multiple mechanisms responsible for immune modulation. With the discovery of two cannabinoid receptor subtypes, the mode of action of cannabinoids is being elucidated. The central CB1 receptor is predominantly expressed in the central nervous system [12], whereas the peripheral CB2 receptor is highly expressed in lymphoid tissues [13]. Transcripts of the CB1 receptor have been detected in leukocytes [14,15], indicating expression of both receptors within the immune system. Despite this, cannabinoids can induce physiological changes in leukocytes by a receptor-independent mechanism [16,17] perhaps due to their highly lipophilic nature, which perturbs membranes. On the other hand, receptor binding and subsequent biological activity exhibit stereoselectivity [18,19]. Selective cannabinoid receptor antagonists demonstrate that enhanced B cell proliferation [20], inflammatory cytokine induction by a promyelocyte cell line [21], and suppressed tumor necrosis factor-a production by human mononuclear cells [22] are mediated through the CB2 receptor. Because the CB2 receptor is apparently expressed at higher levels in leukocytes than the CB1 receptor [1,13 – 15], the CB2 receptor is viewed as being mainly responsible for cannabinoid-induced immune modulation. Contrary to this notion, selective receptor antagonists implicate the CB1, not the CB2, receptor as mediating cannabinoids’ effect during

endotoxin shock [23] and suppression of macrophage killing of Brucella [24]. As a further complication, both receptors seem to be involved in cannabinoids’ anti-inflammatory property [25], and decreased host resistance to Legionella [26]. Hence, the relative contribution of CB1 and CB2 receptors in immune modulation remains unclear. Helper CD4+ T cells are important regulatory cells that control virtually all aspects of immunity by cytokine secretion. Activation of these T cells requires their physical interaction with antigen-presenting cells or accessory cells. This requirement is partially due to the specificity of the T cell receptor, because its ligand is generated by antigen-presenting cells [27,28]. However, the signal transduced through the T cell receptor is usually insufficient to stimulate T cell proliferation and cytokine secretion maximally [29,30]. Independent of the T cell receptor, a second critical signal, termed costimulatory activity, is delivered by costimulatory molecules expressed on antigen-presenting cells [29,30], and several costimulatory molecules have been identified. The lack of costimulatory signals during T cell receptor occupancy results in longlasting T cell unresponsiveness or anergy [31,32]. We previously investigated whether THC influences costimulatory activity of macrophages. THC exposure of a macrophage cell line suppresses their costimulatory activity and decreases expression of heat-stable antigen (HSA), but not B7 molecules [33,34], which are costimulatory molecules. In addition, THC impairs costimulatory activity of peritoneal macrophages from wild-type mice, while cells from mice genetically deficient in the CB2 receptor (CB2 / ) are refractory to THC inhibition [35], indicating a CB2 receptormediated effect, but these findings do not exclude CB1 receptor involvement for wild-type peritoneal macrophages. In the present study, we investigated the role of cannabinoid receptors in diminished costimulatory activity of murine peritoneal macrophages. The partial agonist THC, but not low affinity stereoisomer CP56,667, interfered with costimulation provided by wild-type macrophages. Antagonists selective for both cannabinoid receptors reversed, at least partially, THC-induced inhibition, although only CB2 mRNA was detected in the macrophages. High affinity agonist CP55,940, but not THC, impaired the ability of macrophages from CB2 / mice to costimulate the T

S. Chuchawankul et al. / International Immunopharmacology 4 (2004) 265–278

cells. CP55,940 reversed the inverse agonist activity of CB1-selective antagonist using CB2 / macrophages, while CB1 transcripts were not detected in these macrophages. These findings suggest that CB2 receptors and possibly another receptor subtype may mediate cannabinoids’ inhibitory effect on macrophage costimulation.

267

2. Materials and methods

2C11 (anti-CD3) antibodies were obtained from American Type Culture Collection. Purified J11d (anti-HSA), 1G10 (anti-B7-1), GL1 (anti-B7-2) and biotinylated M1/70 (anti-Mac-1) antibodies were purchased from BD PharMingen (San Diego, CA). Culture supernatant of 20C9 (anti-HSA) [36] was kindly provided by Dr. Yang Liu (Ohio State University, Columbus, OH). Protein A affinity-purified MAR 18.5 was conjugated with fluorescein isothiocyanate isomer I (Molecular Probes, Eugene, OR).

2.1. Mice

2.4. T cell costimulation assays

Female C3D2F1/J mice were purchased from Jackson Laboratory (Bar Harbor, ME). Generation and description of CB2 / mice were previously reported [35]. Genotyping to identify CB2 / and wild-type mice was performed by Southern blot hybridization and then polymerase chain reaction analysis of tail DNA as described [35]. Mice were housed four to a cage under specific pathogen-free conditions and fed Purina diet and acidified water ad libitum.

Costimulation assays with PEC were performed as described [33]. Briefly, wells of flat bottom microtiter plates were coated with culture supernatant containing monoclonal anti-CD3 antibody or complete medium overnight at 4 jC. PEC at 1.25  105/ml in complete medium were cultured with 1.5  105/ml T cell 2B4.11 in replicates at 37 jC for 24 h. Before culture, PEC were irradiated with 3000 rad from a Cs source to prevent the cells from depleting cytokines secreted by 2B4.11 cells. In some experiments, culture supernatants containing M1/69 or 20C9, monoclonal anti-HSA antibodies, at 1:10 dilution were added at culture initiation. Cell-free culture supernatants were assayed for IL-2 by an antibody-sandwich enzyme-linked immunosorbent assay (Pierce Endogen, Rockford, IL). Absorbance at 450 nm was measured with a SpectraMax 250 microplate reader (Molecular Devices, Sunnyvale, CA). IL-2 quantities in culture supernatants were calculated from standard curves of recombinant IL-2 (R&D Systems, Minneapolis, MN). Another costimulation assay utilized 2.5 nM phorbol 12,13-dibutyrate (Sigma) instead of PEC. Wells were coated with anti-CD3 antibody and 2B4.11 cells were added to the wells as above. IL-2 in culture supernatants was measured as above.

2.2. Cells Mice at least 9 weeks of age were administered intraperitoneally saline or 4 mg polystyrene beads with 1.0-Am diameter (Sigma, St Louis, MO), followed by 2.0 ml of 3% Brewer’s thioglycollate broth (Difco Laboratories, Detroit, MI), 24 h later [35]. Peritoneal exudate cells (PEC) from at least 2 mice were harvested by lavage after another 4 days and used as accessory cells. Cell purity was assessed by immunofluorescence staining and flow cytometry. Fewer than 5% of peritoneal cells were B and T lymphocytes, and >75% were Mac-1+. Murine helper T cell hybridoma 2B4.11 was provided by Dr. Ronald Schwartz (National Institutes of Health, Bethesda, MD). Murine macrophage cell line J774A.1 was obtained from American Type Culture Collection (Manassas, VA). Complete medium for cell culture was previously described [33]. 2.3. Monoclonal antibodies Cell hybridomas producing monoclonal M1/69 (anti-HSA), MAR 18.5 (anti-rat n L chain), and 145-

2.5. Drug exposure THC, CP56,667 and CP55,940 were provided by Center for Drug Abuse Research (Dr. Bill Martin, Virginia Commonwealth University, Richmond, VA). SR141716A and SR144528 were supplied by Sanofi Recherche (Montpellier, France). All drugs were prepared in ethanol as described [37] and used from 0.1

268

S. Chuchawankul et al. / International Immunopharmacology 4 (2004) 265–278

to 1000 nM diluted in complete medium. Vehicle consisted of 0.1% ethanol and did not affect cell viability. Cells were exposed to cannabinoids as described [37]. Briefly, PEC were preincubated with vehicle or agonists for 4 h. T cells were added, and the concentrations of agonists and vehicle were maintained. Expression of HSA on PEC was measured by two-color immunofluorescence staining and flow cytometric analysis as described below. For antibody blocking experiments, PEC were preincubated with vehicle or 100 nM THC for 4 h in the absence or presence monoclonal antibodies. Anti-HSA antibody 20C9 was added as described above. Anti-B7-1/2 antibodies were added at 10 Ag/ml. For reversal of THC inhibition, wild-type PEC were preincubated with 1 AM SR141716A or SR144528, or vehicle for 4 h, and then with 100 nM THC or vehicle for an additional 4 h [38]. For reversal of enhanced response, CB2 / PEC were preincubated with 0.1 nM CP55,940 or vehicle for 4 h, and then with 10 nM SR141716A, 10 nM SR144528 or vehicle for an additional 4 h. T cell costimulation assays were conducted as described above. For costimulatory assays utilizing phorbol 12,13-dibutyrate instead of PEC, CP55,940, SR141716A, SR144528 or vehicle were added at culture initiation.

and cell clumps. Data were collected on 20,000 or 50,000 cells for two-color analysis with logarithmic amplification. Fluorescence intensity is expressed in arbitrary fluorescence units.

2.6. Immunofluorescence staining and flow cytometric analysis PEC before and after culture were stained for HSA as described [34]. Briefly, cells were coincubated with 25 Ag normal mouse IgG to block Fc receptors. To detect Mac-1 expression on macrophages, cells were incubated with biotinylated M1/70 and then phycoerythrin-conjugated streptavidin (Life Technologies, GIBCO BRL Products, Rockville, MD). To detect HSA, cells were incubated with J11d followed by fluoresceinated MAR 18.5. Controls for nonspecific fluorescence consisted of cells incubated with irrelevant isotype-matched antibody (purified myeloma protein MOPC 104E; Cappel Organon Teknika, Durham, NC), fluoresceinated MAR 18.5, and phycoerythrin-conjugated streptavidin. Fluorescence intensity of cells was measured by a Becton Dickinson FACScan (San Jose, CA) equipped with a 15 mW 488 nm argon laser and appropriate excitation filters. Forwardangle side scatter gates were set to exclude dead cells

Fig. 1. THC impairs macrophage costimulatory activity. Four days after thioglycollate administration, PEC were used as accessory cells and incubated with T cell hybridoma 2B4.11 in monoclonal anti-CD3 antibody-coated wells. PEC were preincubated with 0.1% ethanol (vehicle) or various THC concentrations for 4 h. After another 24 h, IL-2 secretion by 2B4.11 cells was measured. Values are mean IL-2 in pg/ml F S.D. from triplicate cultures. Each experiment shown is representative of three. (A) PEC from mice intraperitoneally administered saline followed by thioglycollate, 24 h later. Cultures lacking anti-CD3 antibody contained 81.0 F 15.3 pg/ml IL-2. Cultures containing anti-CD3 antibody but lacking PEC had 82.2 F 11.3 pg/ml IL-2. (B) PEC from mice intraperitoneally administered polystyrene beads followed by thioglycollate, 24 h later. Cultures lacking anti-CD3 antibody contained 36.3 F 1.5 pg/ ml IL-2. Cultures containing anti-CD3 antibody but lacking PEC had 34.0 F 3.5 pg/ml IL-2. *Significantly different from vehicle control ( p < 0.01).

S. Chuchawankul et al. / International Immunopharmacology 4 (2004) 265–278

2.7. Reverse transcriptase polymerase chain reaction (RT-PCR) As described [39], total RNA was isolated from cells using RNA STAT-60 (Tel-Test ‘‘B’’, Friendswood, TX). RNA at 1 Ag/ml was treated with amplification grade DNase I (Invitrogen, Carlsbad, CA) for 20 min at room temperature to remove genomic DNA. Yield was determined by absorbance at 260 nm, and purity was verified by 260:280 and 260:230 nm absorbance ratios. RNA was reversed transcribed for 1 h using Superscript II RNase H kit (Invitrogen) using random hexamer primers according to manufacturer’s directions. Controls omitted reverse transcriptase to verify that genomic DNA was not amplified by PCR. Products were amplified by PCR using Taq DNA polymerase (Promega, Madison, WI). Forward primer for amino terminus of CB2 receptor was 5V-AAA TGC TTG ATT GCT GTC AGC TCT C3V, and reverse primer was 5V-CCA CAG AGG CTG TGA AGG TCA-3V. Forward primer for the carboxy terminus of CB2 receptor was 5V-CGG CTA GAC GTG AGG TTG GC-3V, and reverse primer was 5VGGC TCC TAG GTG GTT TTC AC-3V. Forward and reverse primers for CB1 receptor were 5V-CTG GAA CTG CAA GAA GCT-3V and 5V-GCC CCA GCA GAT GAT CAA CAC C-3V, respectively. Forward and reverse primers for h-actin were 5V-ATT GCC GAT AGT GAT GAC CT-3V and 5V-CGT GAA AAG ATG ACC CAG AT-3V, respectively. Following initial denaturation at 95 jC for 2 min, samples were denatured Fig. 2. No alteration in HSA expression on macrophages. Profiles display cell surface expression of HSA on Mac-1+ PEC as assessed by two-color immunofluorescence staining and flow cytometry. (A) HSA staining on macrophages from mice administered thioglycollate (dashed line). HSA staining on macrophages from mice administered polystyrene beads followed by thioglycollate (dotted line). Thioglycollate-induced macrophages incubated with isotypematched irrelevant antibody (solid line) as a control for nonspecific fluorescence. (B) PEC from mice administered polystyrene beads and thioglycollate were incubated with T cells and vehicle or 100 nM THC (see Fig. 1B legend). HSA staining on vehicle control macrophages (widely spaced dotted line). HSA staining on THCexposed macrophages (dashed line). Nonspecific fluorescence of vehicle control (solid line), and THC-exposed (closely spaced dotted line) macrophages. (C) HSA expression on bead- and thioglycollate-induced macrophages from wild-type mice (+/+ PEC) (dotted line) and from mice genetically deficient in CB2 receptor (CB2 / ) (dashed line). Nonspecific fluorescence of wild-type macrophages (solid line).

269

270

S. Chuchawankul et al. / International Immunopharmacology 4 (2004) 265–278

at 95 jC for 1.5 min, annealed at 60 jC for 1.5 min and extended at 72 jC for 2 min for 30 cycles. Fragments were extended by incubation at 72 jC for 8 min. PCR products were separated by agarose gel electrophoresis. Gels were stained with ethidium bromide, and bands were visualized by a UV transilluminator with a CCD camera. 2.8. Statistical analysis

3. Results 3.1. Induction of sensitivity to THC inhibition and functional HSA Immobilized monoclonal anti-CD3 antibody, which crosslinks the T cell receptor complex, was

Table 1 Induction of functional HSAa

Medium M1/69 20C9 a

Thioglycollateinduced PEC

Beads + thioglycollateinduced PEC

IL-2 (pg/ml)

IL-2 (pg/ml)

1627.3 F 108.6 1429.0 F 32.5 1366.0 F 123.8

Inhibition (%)

12.2 16.1

Treatment

IL-2 (pg/ml)

Treatment

IL-2 (pg/ml)

Vehicle THC Anti-HSA THC + Anti-HSA

1670.7 F 140.4 928.3 F 36.1* 536.4 F 54.4** 503.1 F 31.7**

Vehicle THC Anti-B7-1/2 THC + Anti-B7-1/2

1990.3 F 138.3 1314.7 F 78.7* 1024.2 F 136.3** 664.0 F 17.3**

a

Parametric analysis of variance was performed by a two-tailed Student’s t-test for unmatched pairs to compare treatment groups with vehicle control, and a Dunnett’s test to compare various treatments. Significance was designated at p < 0.05.

Treatment

Table 2 Inhibition by THC and antibodies specific for costimulatory moleculesa

1431.1 F 53.4 1285.2 F 70.1 209.7 F 12.5*

Inhibition (%)

10.2 85.4

Mice were administered either saline or polystyrene beads intraperitoneally. One day later, mice were administered thioglycollate intraperitoneally. Costimulatory activity of PEC was assessed after another 4 days in the presence and absence of monoclonal antibodies specific for HSA. M1/69 does not inhibit the function of HSA, whereas 20C9 does [36]. IL-2 values are the mean F S.D. from triplicate cultures. The experiment is representative of five. Background IL-2 levels without anti-CD3 antibody were 83.4 F 1.7 pg/ml with thioglycollate-induced PEC and 36.3 F 1.5 pg/ml with beads + thioglycollate-induced PEC. Cultures containing anti-CD3 antibody but lacking PEC had 92.6 F 11.7 and 34.0 F 3.5 pg/ml IL-2, respectively. * Significantly different from M1/69 control ( p < 0.001).

Costimulatory activity of PEC from mice administered polystyrene beads followed by thioglycollate was assessed (see Fig. 1 legend). Cells were incubated in the presence and absence of 100 nM THC and/or monoclonal antibodies specific for HSA or B7-1/2 costimulatory molecules. IL-2 values are the mean F S.D. from triplicate cultures. Experiments are representative of at least two repetitions. Background IL-2 levels without anti-CD3 antibody were 24.8 F 0.6 pg/ml for anti-HSA antibody experiment and 41.5 F 1.5 pg/ml for anti-B7-1/2 antibody experiment. * Significantly different from vehicle control ( p < 0.01). ** Significantly different from vehicle control ( p < 0.001).

used to provide the primary signal to helper T cell hybridoma 2B4.11. Thioglycollate-induced PEC were used as accessory cells to deliver secondary signals called costimulatory activity, and T cell activation was measured by IL-2 secretion. The impact of THC on costimulatory activity provided by PEC was examined. Macrophages were >75% of the peritoneal cell population. THC did not significantly influence the IL-2 quantity secreted by the T cells (Fig. 1A). Previously, we reported that THC interferes with costimulatory activity of a macrophage cell line that is associated with HSA, a costimulatory molecule [33,34]. HSA expression on thioglycollate-induced PEC was assessed by two-color immunofluorescence staining. Macrophages were distinguished from other cell types by expression of Mac-1 molecule. As illustrated in Fig. 2A, macrophages positively stained for cell surface HSA. To determine whether HSA was functional, an antibody blocking experiment was performed in the absence of THC. Although monoclonal anti-HSA antibody 20C9 interferes with HSA function [36], 20C9 did not appreciably inhibit the T cell response using thioglycollate-induced PEC as accessory cells (Table 1). M1/69 is another monoclonal antibody specific for HSA that does not block its function [36] and was used as a control. Although thioglycollate-induced PEC expressed HSA, the molecule was not functional.

S. Chuchawankul et al. / International Immunopharmacology 4 (2004) 265–278

We attempted to induce functional HSA on peritoneal macrophages. Other investigators reported that ingestion of polystyrene beads by a macrophage line causes expression of functional HSA [40]. Mice were administered polystyrene beads followed by thioglycollate, 24 h later. The percentage of macrophages within the peritoneal cell population was unaffected by bead administration compared to thioglycollate induction alone (data not shown). Again, functional HSA was examined by an antibody blocking experiment. The 20C9 antibody dramatically decreased costimulatory activity by PEC induced with polystyrene beads and thioglycollate (Table 1), indicating that HSA was now functional. However, overall cell surface expression of HSA on these macrophages did not change by immunofluorescence staining (Fig. 2A). In this case, THC significantly decreased

271

the quantity of secreted IL-2 (Fig. 1B), and cell death could not explain the inhibition (data not shown). Maximal inhibition occurred at 10 and 100 nM THC amounting to approximately 45%. When peritoneal macrophages expressed functional HSA, the cells were sensitive to THC inhibition of costimulatory activity. For all subsequent experiments, PEC from mice administered polystyrene beads and thioglycollate were used as accessory cells. We assessed whether THC altered HSA expression on peritoneal macrophages. PEC and T cells were incubated with vehicle or 100 nM THC as above. At the end of the culture period, cells were immunofluorescence-stained. THC did not change the level of Mac-1 expression on macrophages (data not shown). Fluorescence intensity of HSA staining on THCexposed PEC was similar to that on vehicle control

Fig. 3. Cannabinoid receptor involvement. Costimulatory activity of PEC from mice administered polystyrene beads followed by thioglycollate was assessed (see Fig. 1 legend). Values are the mean IL-2 in pg/ml F S.D. from triplicate cultures. (A) PEC were preincubated with vehicle or various concentrations of CP56,667, an inactive enantiomer, for 4 h before culture with T cells. Cultures lacking anti-CD3 antibody contained 13.3 F 6.6 pg/ml IL-2. The experiment shown is representative of three. (B) PEC were incubated with vehicle or 1.0 AM SR141716A, a CB1 selective antagonist, or SR144528, a CB2 selective antagonist, for 4 h, and then with vehicle or 100 nM THC for another 4 h. T cell costimulation assays were performed. Cultures lacking anti-CD3 antibody contained 31.2 F 1.0 pg/ml IL-2. *Significantly different from vehicle control ( p < 0.02). The experiment shown is representative of eight. (C) PEC were preincubated with vehicle or various concentrations of SR144528 for 4 h, and T cell costimulation assays were performed. (D) PEC were preincubated with vehicle or various concentrations of SR141716A for 4 h, and T cell costimulation assays were performed. Panels C and D are from the same experiment, which is representative of three. Cultures lacking anti-CD3 antibody contained 39.1 F 7.3 pg/ml IL-2. *Significantly different from vehicle control ( p < 0.05).

272

S. Chuchawankul et al. / International Immunopharmacology 4 (2004) 265–278

PEC (Fig. 2B). THC did not down regulate HSA expression at the concentration examined. Antibody blocking experiments were performed in conjunction with THC treatment. As before 100 nM THC significantly inhibited the T cell response (Table 2). Again, monoclonal anti-HSA antibody interfered with T cell activation. The combination of THC and anti-HSA antibody had the same inhibitory effect as that of the antibody alone. Monoclonal anti-B7-1/2 antibodies also significantly decreased the T cell response. In contrast, the degree of inhibition in cultures containing both THC and anti-B7-1/2 antibodies was significantly was greater than that caused by antibody or THC alone ( p < 0.02). THC-mediated suppression appeared to be additive with anti-B7-1/2 antibodies, but not with anti-HSA antibody.

3.2. Cannabinoid receptor participation We investigated whether the drug’s suppression of costimulatory activity was mediated through a cannabinoid receptor. Stereoselectivity of functional responses to cannabinoids is one criterion for a receptor-mediated mode of action. We utilized the inactive enantiomer CP56,667, as an indicator of stereospecificity. In contrast to THC, CP56,667 did not significantly decrease IL-2 production (Fig. 3A). In addition, we examined whether selective cannabinoid receptor antagonists would compete with THC and reverse the drug’s inhibition of the T cell response. SR141716A is a selective antagonist for the CB1 receptor, and SR144528 is a selective antagonist for the CB2 receptor. PEC were incubated with or

Fig. 4. Impact of cannabinoid antagonists and agonists using macrophages from CB2 / mice. Mice genetically deficient in CB2 receptor were administered polystyrene beads followed by thioglycollate. Costimulatory activity of PEC was assessed (see Fig. 1 legend). Values are the mean IL-2 in pg/ml F S.D. from triplicate cultures. (A) CB2 / PEC were preincubated with vehicle or various concentrations of SR144528, a CB2 selective antagonist, before culture with T cells. (B) CB2 / PEC were preincubated with vehicle or various concentrations of SR141716A, a CB1 selective antagonist, before culture with T cells. Panels A and B are from the same experiment, which is representative of four. Cultures lacking anti-CD3 antibody contained 122.6 F 20.3 pg/ml IL-2. *Significantly different from vehicle control ( p < 0.02). (C) CB2 / PEC were preincubated with vehicle or various THC concentrations before culture with T cells. The experiment shown is representative of four. Cultures lacking anti-CD3 antibody contained 41.1 F 6.6 pg/ml IL-2. (D) PEC were preincubated with vehicle or various concentrations of CP55,940, an active enantiomer, before culture with T cells. Cultures lacking anti-CD3 antibody contained 92.5 F 7.3 pg/ml IL-2. The experiment shown is representative of three. *Significantly different from vehicle control ( p < 0.05).

S. Chuchawankul et al. / International Immunopharmacology 4 (2004) 265–278

without either antagonist at 1 AM for 4 h followed by 100 nM THC or vehicle for another 4 h, and the T cell costimulation assay was performed as above. Neither antagonist alone affected IL-2 secretion in the experiment shown in Fig. 3B. As before, THC alone significantly decreased the T cell response compared to the vehicle control. SR141716A completely abrogated THC’s inhibition; the IL-2 quantity was slightly above the vehicle control level. On the other hand, SR144528 only partially reversed THC’s suppressive effect. Cannabinoid receptors may have a role in mediating the drug’s immunosuppression. In half of the preceding experiments with the antagonists, 1 AM SR144528 alone decreased the T cell response and did not affect THC’s inhibition, and, hence, we examined the effect of the antagonists alone on the T cell response. SR144528 at 1 and 10 nM significantly increased IL-2 production, while 1 AM was inhibitory as shown in Fig. 3C. This inhibition could not be attributed to cell death (data not shown). SR141716A at 10 and 100 nM also augmented IL-2 secretion (Fig. 3D), and no suppression was detected in any experiment. Both antagonists alone influenced the T cell response depending on their concentration. We further investigated the impact of cannabinoid antagonists and the possible involvement of cannabinoid receptors by using PEC from mice genetically deficient in the CB2 receptor. The percent Mac-1+ peritoneal cells from CB2 / mice did not differ from the frequency from wild-type mice (74.6 F 2.3% for CB2 / mice vs. 75.0 F 7.2% for CB2+/ + mice). Cell surface expression of Mac-1 (data not shown) and HSA (Fig. 2C) on macrophages from CB2 / mice was normal. Furthermore, the anti-HSA blocking antibody 20C9 inhibited the T cell response using CB2 / PEC by 81.0 F 13.7% as an average for three separate experiments. Unlike the preceding findings (see Fig. 3C), SR144528 alone did not affect the T cell response with CB2 / PEC as the accessory cells (Fig. 4A). In contrast, 10 and 100 nM SR141716A alone augmented the level of secreted IL-2 (Fig. 4B) similar to the above results (see Fig. 3D). The possible participation of another cannabinoid receptor besides CB2 was unexpected, because we previously reported CB2 / PEC are resistant to THC inhibition [35]. In agreement, THC did not influence IL-2 secretion in the presence of CB2 / PEC (Fig. 4C). The degree of THC suppression with

273

wild-type PEC from C57BL/6 mice was similar to that observed for PEC from C3D2F1 mice (data not shown). Next, the active enantiomer CP55,940, which has a higher affinity and greater potency than THC, was examined. In contrast to THC, 10 and 100 nM CP55,940 significantly decreased IL-2 production with CB2 / accessory cells (Fig. 4D). The degree of inhibition was approximately 27% compared to the vehicle control cultures. Finally, the ability of CP55,940 to reverse the enhanced response observed with SR141716A was examined (Fig. 5). CP55,940 at 0.1 nM did not affect the T cell response as in the preceding experiment (see Fig. 4D). As before (see Fig. 4A and B), 10 nM SR141716A, but not SR144528, significantly increased the IL-2 level (Fig. 5). In the presence of both CP55,940 and SR141716A, the IL-2 quantity was equivalent to that in vehicle control cultures. 3.3. Cannabinoid receptor expression Expression of cannabinoid receptors was assessed by RT-PCR. Transcripts encoding neither CB1 nor CB2 receptor were detected in thioglycollate-induced PEC (Fig. 6A). In contrast, CB2 mRNA, but not CB1

Fig. 5. Reversal of enhanced response with CB2 / PEC. Costimulatory activity of PEC from CB2 / mice administered polystyrene beads followed by thioglycollate was assessed (see Fig. 1 legend). PEC were incubated with vehicle or 0.1 nM CP55,940 for 4 h and then with vehicle or 10 nM SR141716A or SR144528 for another 4 h. T cell costimulatory assay was performed. Values are the mean IL-2 in pg/ml F S.D. from triplicate cultures. IL-2 was 111.7 F 15.0 in cultures with antiCD3 antibody but no PEC. Experiment is representative of four. *Significantly different from vehicle control ( p < 0.001).

274

S. Chuchawankul et al. / International Immunopharmacology 4 (2004) 265–278

Fig. 6. Cannabinoid receptor expression. RNA was isolated and prepared for RT-PCR. Reactions either lacked reverse transcriptase ( RT) or contained the enzyme (+ RT). (A) Thioglycollate-induced wild-type PEC. (B) Polystyrene bead- and thioglycollate-induced wild-type PEC. (C) Polystyrene bead- and thioglycollate-induced CB2 / PEC. (D) Murine macrophage cell line J774A.1 and murine T cell hybridoma 2B4.11.

mRNA, was readily detected in PEC induced with polystyrene beads and thioglycollate (Fig. 6B). As expected, CB2 / PEC lacked detectable CB2 mRNA, but CB1 mRNA was also not detected in Fig. 7. Direct influence of cannabinoids on T cell hybridoma. T cell hybridoma 2B4.11 was incubated in anti-CD3 antibodycoated wells with and without 2.5 nM phorbol 12,13-dibutyrate for 24 h. At culture initiation, vehicle or various concentrations of (A) CP55,940, (B) SR141716A or (C) SR144528 were added. Values are mean IL-2 in pg/ml F S.D. from triplicate cultures. The results are from the same experiment, which is representative of four. Background IL-2 level in cultures with anti-CD3 antibody alone was 149 F 22 pg/ml. IL-2 in cultures with phorbol 12,13dibutyrate alone was 755 F 36 pg/ml. *Significantly different from vehicle control ( p < 0.02).

S. Chuchawankul et al. / International Immunopharmacology 4 (2004) 265–278

these cells (Fig. 6C). The murine macrophage cell line J774A.1 expressed CB1 mRNA and served as a positive control (Fig. 6D). The T cell hybridoma 2B4.11 expressed a low level of CB2 mRNA, while CB1 mRNA was not detected. 3.4. Cannabinoid effect on T cells Because CB2 mRNA was detected in the T cell hybridoma, we developed a costimulatory assay that is independent of macrophages to investigate the direct effect of cannabinoids on the T cells. In this culture system, PEC were replaced with a phorbol ester. Although phorbol ester alone stimulated IL-2 secretion, the combination of phorbol ester and antiCD3 antibody induced a substantially higher level of IL-2. Neither CP55,940 (Fig. 7A) nor SR141716A (Fig. 7B) at the concentrations examined influenced the T cell response in this system. However, 10 nM to 1 AM SR144528 significantly inhibited T cell activation (Fig. 7C).

4. Discussion We investigated the consequence of THC on peritoneal macrophages to deliver costimulatory signals to a helper T cell hybridoma. The primary signal was provided by immobilized anti-CD3 antibody that bypasses T cell receptor specificity. IL-2 secretion by the T cells requires a second signal that was provided by macrophages. THC diminished the level of T cell activation when the accessory cells were peritoneal macrophages elicited by polystyrene beads and thioglycollate. However, THC had no effect on the T cell response with macrophages elicited by thioglycollate alone. These results cannot be explained by differences in the proportion of macrophages within the peritoneal population, nor differential cell death due to THC exposure. Macrophages induced with beads and thioglycollate expressed CB2 mRNA, whereas CB2 transcripts were not detected in thioglycollate-induced macrophages. Other investigators have reported that the level of CB2 expression depends on the activation state of macrophages [41], which is consistent with our findings. Thus, sensitivity of peritoneal macrophages to THC inhibition was accompanied by CB2 mRNA expression.

275

The functional role of cannabinoid receptors was assessed by multiple criteria. Diminished IL-2 secretion caused by THC exposure was stereoselective, whereby the stereoisomer CP56,667 did not impair the T cell response. CP56,667 binds both cannabinoid receptors, but does not induce a biological response via the receptors [18,19]. SR141716A, a CB1 receptor selective antagonist, completely reversed the suppressive effect of THC, while SR144528, a CB2 receptor selective antagonist, partially eliminated the drug’s inhibition. The partial reversal by SR144528 may be related to a suppressive effect observed with 1 AM SR144528 alone in half of the experiments. Reversal by SR141716A was a surprise, because we previously found that macrophages from CB2 / mice are refractory to THC suppression of costimulation [35]. We compared the influence of high affinity agonist CP55,940 with that of THC which is categorized as a partial agonist [18,19]. In agreement with our previous findings, THC did not affect the T cell response in the presence of peritoneal cells from CB2 / mice. In contrast, CP55,940 decreased T cell activation with macrophages from CB2 / mice as the accessory cells, suggesting a differential influence depending on ligand affinity. Yet the action of the antagonists is complex, because both antagonists can behave as inverse agonists in some systems [42 – 44]. Cannabinoid receptors transduce signals in other cell types in the absence of exogenous agonists, and the antagonists inhibit signaling of constitutively active receptors [42 –44]. In our present study, both antagonists alone at low concentrations slightly, yet significantly, increased the IL-2 secretion with wild-type accessory cells. The augmentation appeared to be receptorselective, because SR141716A, but not SR144528, enhanced the T cell response with CB2 / accessory cells. Furthermore, CP55,940 at a concentration that was not inhibitory completely reversed the increased response caused by SR141716A with CB2 / macrophages. Taken altogether, our results are consistent with the possible participation of cannabinoid receptors, especially the CB2 receptor subtype, in this immunosuppression. Typically, the CB1 receptor is invoked when SR141716A has an influence [23 – 26]. However, CB1 transcripts were not detected in either wild-type or CB2 / macrophages. The CB1 receptor when detected is expressed at a lower level relative to the

276

S. Chuchawankul et al. / International Immunopharmacology 4 (2004) 265–278

CB2 receptor in leukocytes [13 – 15]. Perhaps, the level of CB1 mRNA expression was too low to be detected by the RT-PCR. Alternatively, evidence for the existence of another, yet unidentified, cannabinoid receptor subtype has been recently published [45 – 47]. For example, some CB2-selective deoxy-THC analogs with low affinity for the CB1 receptor cause unexpected suppression of locomotor activity, hypothermia and antinociception, and their effects are blocked by SR141716A, but not SR144528 [45]. Additionally, SR141716A reverses cannabinoid-induced hypotension and mesenteric vasodilation in mice genetically deficient in both CB1 and CB2 receptors [46]. Our results with SR141716A suggest the possible involvement of another cannabinoid receptor in addition to the CB2 receptor. The T cell hybridoma used in the costimulation system expressed a low level of CB2 mRNA. Other investigators have reported that THC decreases IL-2 secretion by T cells activated by mitogenic stimuli [2,3]. In the case of another T cell line, cannabinoid modulation of IL-2 gene expression is independent of cannabinoid receptors [17]. If THC exerted its inhibition through the T cells, decreased IL-2 production should have occurred regardless of the source of macrophages, which was not observed. We previously reported that THC does not diminish surface expression of the CD3 complex on the T cells used in our current study [37] and does not decrease their response in a costimulation system lacking macrophages [33]. Neither CP55,940 nor SR141716A affected the T cell response in a system independent of macrophages, whereby the costimulatory signal was provided by a phorbol ester. On the other hand, SR144528 was suppressive in the phorbol ester system, which may have contributed to the partial reversal by this antagonist in the macrophage system. Our findings suggest that the inhibitory effect of cannabinoids on the T cell hybridoma activation in the macrophage system is most likely exerted mainly at the level of macrophages, not T cells. Thus, cannabinoids may influence on an immune response at a step before helper T cell cytokine production. Although several costimulatory molecules have been identified, we focused on HSA expressed by peritoneal macrophages. THC depresses costimulatory activity of a macrophage cell line [33,34], which constitutively expresses functional HSA [34]. THC

exposure of this macrophage cell line down regulates expression of HSA, but not B7-1/2 molecules [33,34]. Although resident and thioglycollate-induced peritoneal macrophages express HSA, the molecule lacks costimulatory activity [48]. Our antibody blocking experiments verified that HSA on thioglycollateelicited macrophages was not functional. In addition, the cells were resistant to THC inhibition. While cytokine stimulation of peritoneal macrophages does not increase HSA expression or induce its function [48], ingestion of polystyrene beads by a macrophage cell line in culture induces functional HSA [40]. Based on this report, we developed a protocol to induce functional HSA on peritoneal macrophages by administering polystyrene beads followed by thioglycollate. Acquisition of HSA function was confirmed by antibody blocking experiments, while HSA cell surface expression was not changed. Once macrophages expressed HSA with costimulatory activity, the cells became sensitive to THC suppression. However, THC exposure did not decrease HSA expression on peritoneal macrophages, unlike our findings with the macrophage cell line [33,34]. Inhibition of costimulatory activity by THC and anti-HSA antibody was equivalent to that with the antibody alone. In contrast, the combination of THC and anti-B7-1/2 antibodies was significantly more suppressive than either agent alone; the degree of suppression was nearly additive. The lack of an additive effect with THC and anti-HSA antibody suggests a possible overlap in their mode of action. Other investigators have postulated that functional activity of HSA is determined by its glycosylation, not necessarily by the level of protein expression [36,40,48]. Although THC may affect other costimulatory molecules, the drug may, perhaps, alter HSA glycosylation causing the molecule to revert back to an inactive form. Additional experiments are needed to either verify or refute this possibility. In summary, cannabinoids impaired costimulatory activity provided by macrophages leading to a diminished helper T cell response. Collectively, the findings suggest that cannabinoid receptors may mediate the suppression. Disruption of costimulatory signals during T cell receptor occupancy causes long-lasting T cell unresponsiveness [31,32]. Cannabinoid interference with costimulation mediated through cannabinoid receptors could potentially compromise immune

S. Chuchawankul et al. / International Immunopharmacology 4 (2004) 265–278

responsiveness. Constitutively active cannabinoid receptors expressed by leukocytes may affect immune regulation during the natural course of an immune response. Therefore, evidence is growing that the role of cannabinoids and their receptors in immune modulation is complex.

Acknowledgements The authors would like to thank Marina Matveyeva for technical assistance. This work was supported by P50 DA05274 and R01 ES07199 grants from the National Institutes of Health. S.C. was supported, in part, by a Royal Thai Scholarship.

References [1] Berdyshev EV. Cannabinoid receptors and the regulation of immune response. Chem Phys Lipids 2000;108:169 – 90. [2] Schatz AR, Koh WS, Kaminski NE. D9-Tetrahydrocannabinol selectively inhibits T-cell dependent humoral immune responses through direct inhibition of accessory T-cell function. Immunopharmacology 1993;26:129 – 37. [3] Nakano Y, Pross SH, Friedman H. Modulation of interleukin 2 activity by D9-tetrahydrocannabinol after stimulation with concanavalin A, phytohemagglutinin, or anti-CD3 antibody. Proc Soc Exp Biol Med 1992;201:165 – 8. [4] Coffey RG, Yamamoto Y, Snella E, Pross S. Tetrahydrocannabinol inhibition of macrophage nitric oxide production. Biochem Pharmacol 1996;52:743 – 51. [5] Jeon YJ, Yang KH, Pulaski JT, Kaminski NE. Attenuation of inducible nitric oxide synthase gene expression by D9-tetrahydrocannabinol is mediated through the inhibition of nuclear factor-n B/Rel activation. Mol Pharmacol 1996;50:334 – 41. [6] Fischer-Stenger K, Dove Pettit DA, Cabral GA. D9-Tetrahydrocannabinol inhibition of tumor necrosis factor-a: suppression of post-translational events. J Pharmacol Exp Ther 1993; 267:1558 – 65. [7] Zheng Z.-M., Specter S, Friedman H. Inhibition by delta-9tetrahydrocannabinol of tumor necrosis factor alpha production by mouse and human macrophages. Int J Immunopharmacol 1992;14:1445 – 52. [8] Zhu W, Newton C, Daaka Y, Friedman H, Klein TW. D9-Tetrahydrocannabinol enhances the secretion of interleukin 1 from endotoxin-stimulated macrophages. J Pharmacol Exp Ther 1994;270:1334 – 9. [9] Derocq JM, Segui M, Marchand J, Le Fur G, Casellas P. Cannabinoids enhance human B-cell growth at low nanomolar concentrations. FEBS Lett 1995;369:177 – 82. [10] Klein TW, Newton C, Friedman H. Cannabinoid receptors and immunity. Immunol Today 1998;19:373 – 81.

277

[11] Yuan M, Kiertcher SM, Cheng Q, Zoumalan R, Tashkin DP, Roth MD. Delta 9-tetrahydrocannabinol regulates Th1/Th2 cytokine balance in activated human T cells. J Neuroimmunol 2002;133:124 – 31. [12] Matusuda LA, Lolait SJ, Brownstein MJ, Young AC, Bonner TI. TI. Structure of a cannabinoid receptor and functional expression of the cloned cDNA. Nature 1990;346:561 – 4. [13] Munro S, Thomas KL, Abu-Shaar M. Molecular characterization of a peripheral receptor for cannabinoids. Nature 1993; 365:61 – 5. [14] Valk P, Verbakel S, Vankan Y, Hol S, Macham S, et al. Anandamide, a natural ligand for the peripheral cannabinoid receptor, is a novel synergistic growth factor for hematopoietic cells. Blood 1997;90:1448 – 57. [15] Noe SN, Newton C, Widen R, Friedman H, Klein TW. AntiCD40, anti-CD3, and IL-2 stimulation induce contrasting changes in CB1 mRNA expression in mouse splenocytes. J Neuroimmunol 2000;110:161 – 7. [16] Derocq JM, Bouaboula M, Marchand J, Rinaldi-Carmona M, Segui M, Casellas P. The endogenous cannabinoid anandamide is a lipid messenger activating cell growth via a cannabinoid receptor-independent pathway in hematopoietic cells. FEBS Lett 1998;425:419 – 25. [17] Jan TR, Rao GK, Kaminski NE. Cannabinol enhancement of interleukin-2 (IL-2) expression by T cells is associated with an increase in IL-2 distal nuclear factor of activated T cell activity. Mol Pharmacol 2002;61:446 – 54. [18] Martin BR, Compton DR, Thomas BF, Prescott WR, Little PJ, et al. Behavioral, biochemical, and molecular modeling evaluations of cannabinoid analogs. Pharmacol Biochem Behav 1991;40:471 – 8. [19] Thomas BF, Compton DR, Martin BR. Characterization of the lipophilicity of natural and synthetic analogs of D9-tetrahydrocannabinol and its relationship to pharmacological potency. J Pharmacol Exp Ther 1990;255:624 – 30. [20] Carayon P, Marchand J, Dussossoy D, Derocq JM, Jbilo O, et al. Modulation and functional involvement of CB2 peripheral cannabinoid receptors during B-cell differentiation. Blood 1998;92:3605 – 15. [21] Jbilo O, Derocq JM, Segui M, Le Fur G, Casellas P. Stimulation of peripheral cannabinoid receptor CB2 induces MCP-1 and IL-8 gene expression in human promyelocytic cell line HL60. FEBS Lett 1999;448:273 – 7. [22] Germain N, Boichot E, Advenier C, Berdyshev EV, Lagente V. Effect of the cannabinoid receptor ligand, WIN 55,212-2, on superoxide anion and TNF-alpha production by human mononuclear cells. Int Immunopharmacol 2002;2:537 – 43. [23] Smith SR, Terminelli C, Denhardt G. Effects of cannabinoid receptor agonist and antagonist ligands on production of inflammatory cytokines and anti-inflammatory interleukin10 in endotoxemic mice. J Pharmacol Exp Ther 2000;293: 136 – 50. [24] Gross A, Terraza A, Marchant J, Bouaboula M, OuahraniBettache S, et al. A beneficial aspect of a CB1 cannabinoid receptor antagonist: SR141716A is a potent inhibitor of macrophage infection by the intracellular pathogen Brucella suis. J Leukoc Biol 2000;67:335 – 44.

278

S. Chuchawankul et al. / International Immunopharmacology 4 (2004) 265–278

[25] Smith SR, Denhardt G, Terminelli C. The anti-inflammatory activities of cannabinoid receptor ligands in mouse peritonitis models. Eur J Pharmacol 2001;43:107 – 19. [26] Klein TW, Newton CA, Nakachi N, Friedman H. Delta 9-tetrahydrocannabinol treatment suppresses immunity and early IFN-gamma, IL-12, and IL-12 receptor beta 2 responses to Legionella pneumophila infection. J Immunol 2000;164: 6461 – 6. [27] Kane LP, Lin J, Weiss A. Signal transduction by the TCR for antigen. Curr Opin Immunol 2000;12:242 – 9. [28] Villadangos JA, Ploegh HL. Proteolysis in MHC class II antigen presentation: who’s in charge? Immunity 2000;12:233 – 9. [29] Lenschow DJ, Walunas TL, Bluestone JA. CD28/B7 system of T cell costimulation. Annu Rev Immunol 1996;14:233 – 58. [30] Watts TH, DeBenedette MA. T cell co-stimulatory molecules other than CD28. Curr Opin Immunol 1999;11:286 – 93. [31] Janeway Jr CA, Bottomly K. Signals and signs for lymphocyte responses. Cell 1994;76:275 – 85. [32] Linsley PS, Ledbetter JA. The role of CD28 receptor during T cell responses to antigen. Annu Rev Immunol 1993;11: 191 – 212. [33] Clements DJ, Cabral GA, McCoy KL. D9-Tetrahydrocannabinol selectively inhibits macrophage costimulatory activity and down-regulates heat-stable antigen expression. J Pharmacol Exp Ther 1996;277:1315 – 21. [34] Clements DJ, Matveyeva M, McCoy KL. Delta9-tetrahydrocannabinol suppresses macrophage costimulation by decreasing heat-stable antigen expression. Int J Immunopharmacol 1998;20:415 – 28. [35] Buckley NE, McCoy KL, Mezey E, Bonner T, Zimmeretal A, et al. Immunomodulation by cannabinoids is absent in mice deficient for the CB2 cannabinoid receptor. Eur J Pharmacol 2000;396:141 – 9. [36] Liu Y, Jones B, Aruffo A, Sullivan KM, Linsley PS, Janeway Jr CA. Heat-stable antigen is a costimulatory molecule for CD4 T cell growth. J Exp Med 1992;175:437 – 45. [37] McCoy KL, Gainey D, Cabral GA. D9-Tetrahydrocannabinol modulates antigen processing by macrophages. J Pharmacol Exp Ther 1995;273:1216 – 23.

[38] McCoy KL, Matveyeva M, Carlisle SJ, Cabral GA. Cannabinoid inhibition of the processing of intact lysozyme by macrophages: evidence for CB2 receptor participation. J Pharmacol Exp Ther 1999;289:1620 – 5. [39] Buckley NE, Hansson S, Harta G, Mezey E. Expression of the CB1 and CB2 receptor messenger RNAs during embryonic development in the rat. Neuroscience 1998;82:1131 – 49. [40] De Bruijn MLH, Peterson PA, Jackson MR. Induction of heatstable antigen expression by phagocytosis is involved in in vitro activation of unprimed CTL by macrophages. J Immunol 1996;156:2686 – 92. [41] Carlisle SJ, Marciano-Cabral F, Staab A, Ludwick C, Cabral GA. Differential expression of the CB2 cannabinoid receptor by rodent macrophages and macrophage-like cells in relation to cell activation. Int Immunopharmacol 2002;2:69 – 82. [42] Landsman RS, Burkey TH, Consroe P, Roeske WR, Yamamura HI. SR 141716A is an inverse agonist of the human cannabinoid CB1 receptor. Eur J Pharmacol 1997;334:R1 – 2. [43] Portier M, Rinaldi-Carmona M, Pecceu F, Combes T, PoinotChazel C, et al. SR144528, an antagonist for the peripheral cannabinoid receptor, behaves as an inverse agonist. J Pharmacol Exp Ther 1999;288:582 – 9. [44] Rhee MH, Kim SK. SR144528 as inverse agonist of CB2 cannabinoid receptor. J Vet Sci 2002;3:179 – 84. [45] Wiley JL, Martin BR. Cannabinoid pharmacology: implications for additional cannabinoid receptor subtypes. Chem Phys Lipids 2002;121:57 – 63. [46] Jarai Z, Wagner JA, Varga K, Lake KD, Compton DR, et al. Cannabinoid-induced mesenteric vasodilation through an endothelial site distinct from CB1 or CB2 receptors. Proc Natl Acad Sci U S A 1999;96:14136 – 41. [47] Facchinetti F, Del Giudice E, Furegato S, Passarotto M, Leon A. Cannabinoids ablate release of TNFa in rat microglial cells stimulated with lypopolysaccharide. Glia 2003;41:161 – 8. [48] Ding L, Linsley PS, Huang L.-Y., Germain RN, Shevach EM. IL-10 inhibits macrophage costimulatory activity by selectively inhibiting the up-regulation of B7 expression. J Immunol 1993;151:1224 – 34.