Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The mood-altering drug marijuana derived from the hemp plant Cannabis sativa contains a group of biosynthetically related substances known collectively as cannabinoids. Tetrahydrocannabinol (THC), one of the major cannabinoids in marijuana, has potent analgesic and anti-inflammatory activities, but it also exhibits psychotropic effects, which limit its clinical application. Considerable effort has been expended toward the goal of creating nonpsychotropic cannabinoid derivatives that retain therapeutic actions but are free of psychotropic activity. A useful template for this search is the THC metabolite THC-11-oic acid (Fig. 1A), because it does not induce changes in mood in human or behavior in animal models. Unfortunately, THC-11-oic acid has only modest analgesic and anti-inflammatory activities (Burstein et al., 1992). By adding two carbons to the side chain of THC-11-oic acid, and introducing two methyl groups, the dimethylheptyl analog of THC-11-oic acid, ajulemic acid (AJA; Fig. 1B), has been developed (Burstein et al., 1992). AJA has potent and prolonged analgesic and anti-inflammatory activities that are comparable with those of morphine (Burstein et al., 1992, 1998; Dajani et al., 1999). Also, in contrast to the nonsteroidal anti-inflammatory drugs (NSAIDs), AJA is not ulcerogenic at therapeutic doses, and does not exhibit tolerance or cause mutagenesis (Dajani et al., 1999; Burstein, 2000). Furthermore, AJA does not have psychotropic actions in animal models (Burstein et al., 1992; Zurier et al., 1998). AJA also suppresses both acute inflammation induced by injection of interleukin (IL)-1 and tumor necrosis factor  in the murine subcutaneous air pouch model, and joint tissue damage in the adjuvant-induced polyarthritis model in rats (Zurier et al., 1998). Based on these findings, AJA has been approved recently for a phase II clinical trial for reduction of pain in humans. However, the mechanism by which AJA exerts its therapeutic activity remains largely unknown, because it does not bind efficiently to either of the cannabinoid-specific cell surface receptors CB1 and CB2 (Pertwee, 1997).

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Chemical structures of THC, THC-11-oic acid, and ajulemic acid

Peroxisome proliferator-activated receptor  (PPAR) is a pharmacologically important member of the nuclear receptor superfamily (Houseknecht et al., 2002). It plays important roles in a diverse array of biological processes, including lipid metabolism, glucose homeostasis, and adipocyte differentiation. The crystal structure of the PPAR ligand-binding domain reveals a large hydrophobic cavity for ligand binding (Uppenberg et al., 1998; Xu et al., 2001). Indeed, PPAR binds to a wide range of synthetic and naturally occurring substances, including the antidiabetic drugs thiazolidinediones (Lehmann et al., 1995; Willson et al., 1996), the synthetic tyrosine analog GW347845 (Cobb et al., 1998), polyunsaturated fatty acids (Kliewer et al., 1997), metabolites of arachidonic acid, including 15-deoxy-^12,14prostaglandin J₂ (Forman et al., 1995; Kliewer et al., 1995), NSAIDs (Lehmann et al., 1997), and compounds of oxidized low-density lipoprotein, such as 13-hydroxyoctadecadienoic acid and 15-hydroxyeicosatraenoic acid (Nagy et al., 1998). Several of these PPAR ligands exhibit anti-inflammatory activity in vivo (Kawahito et al., 2000; Naito et al., 2001), and activation of PPAR is directly linked to anti-inflammatory (Jiang et al., 1998) and antitumor (Patel et al., 2001) processes. Accordingly, activation of PPAR inhibits the expression of cytokines such as IL-1, tumor necrosis factor , and nitric oxide at both transcription and translation levels (Jiang et al., 1998; Ricote et al., 1998). PPAR is expressed in adipose tissue, skeletal muscle, adrenal gland, colonic epithelium, heart, pancreas, and liver (Mukherjee et al., 1997; Sarraf et al., 1998). It is also expressed in immune system-related cells such as splenocytes (Kliewer et al., 1994; Clark et al., 2000), synoviocytes (Kawahito et al., 2000; Ji et al., 2001; Simonin et al., 2002), helper T cells (Clark et al., 2000), and activated monocytes and macrophages (Jiang et al., 1998; Ricote et al., 1998; Kawahito et al., 2000), suggesting that PPAR has a direct role in modulating inflammation in addition to its role in lipid metabolism and glucose homeostasis.

The functional similarity between the anti-inflammatory activity of AJA and known PPAR ligands and the promiscuity of ligand binding by PPAR led us to consider AJA as a potential ligand for PPAR. In line with this speculation, some NSAIDs also bind to PPAR (Lehmann et al., 1997). In this study, we demonstrate that AJA binds specifically and selectively to PPAR but not PPAR or PPAR. Binding of AJA activates PPAR's transcriptional activity and enables recruitment of transcriptional coactivators to the receptor. We also show that AJA inhibits the induction of interleukin-8 promoter activity in a PPAR-dependent manner. Furthermore, we demonstrate that addition of AJA to 3T3 L1 cells in vitro induces differentiation of these cells into adipocytes, a process mediated by PPAR. Thus far, an intracellular receptor for AJA has not been reported; our data suggest that PPAR may serve this function.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Reagents. The PPAR-specific ligand GW347845, PPAR-specific ligand GW32, and PPAR-specific ligand GW36 were kindly provided by Dr. Timothy M. Willson (GlaxoSmithKline, Research Triangle Park, NC). Ajulemic acid was synthesized by Organix, Inc. (Woburn, MA). All other chemicals were purchased from commercial sources.

Plasmids. The pGEX-DRIP205 (527-970) was a gift from Dr. Leonard P. Freedman (Merck, Inc., West Point, PA). The Gal4-hPPAR ligand-binding domain (LBD) constructs were provided by Dr. Steven A. Kliewer (Southwestern Medical Center, Dallas, TX). The IL8-luciferase reporter was provided by Dr. Bernd Stein (Celgene, San Diego, CA). The plasmids mPPAR, mPPAR1, mPPAR, and hRXR were subcloned in the pCMX vector (Umesono et al., 1991). The pCMX-Gal4-mPPAR1 was constructed by fusing the mPPAR1 coding sequence to the yeast GAL4 DNA binding domain (amino acids 1-147) in the pCMX vector. The AF-2 helix-truncated mutant of mPPAR1 (PPAR AF2) was generated by polymerase chain reaction (PCR) amplification to introduce a stop codon and an NheI restriction site after amino acid 489, and then subcloned into the pCMX vector. The PPRE-TK-LUC, MH100-TK-LUC, and DR1-TK-LUC reporters have been described previously (Umesono et al., 1991; Forman et al., 1995).

Partial Protease Digestion Assay. Partial protease digestion was carried out as described previously (Leng et al., 1993). PPAR proteins were made by in vitro transcription/translation reactions in reticulocyte lysate according to manufacture's instructions (Promega, Madison, WI). AJA, GW compounds, or vehicle alone was incubated with the ³⁵S-labeled PPARs at room temperature for 1 hour before trypsin digestion. Reactions were stopped by boiling in SDS-containing sample buffer, and lysates were subjected to SDS-PAGE and autoradiography.

Expression and Purification of GST Fusion Protein. The glutathione S-transferase (GST)-DRIP205 (amino acids 527-970) and GST-RAC3 receptor interaction domain (amino acids 613-752) fusion proteins were expressed in the Escherichia coli BL21 cells by induction with 0.1 mM isopropyl--D-thiogalactopyranoside at room temperature for 4 h. The bacterial pellet was resuspended in buffer containing 150 mM NaCl, 10 mM Tris, pH 8.0, and 1 mM EDTA, with 5 mM dithiothreitol, 1 mM phenylmethlsulfonyl fluoride, and 1.5% Sarkosyl. The cell suspension was sonicated and centrifuged at 7400 rpm at 4°C for 30 min. The supernatant was isolated and 0.02% Triton-X 100 was added. The mixture was then incubated with 1 ml of 50% slurry of glutathione-Sepharose beads on a nutator at 4°C for 30 min. The beads were spun down at 3000 rpm for 10 min, the supernatant was removed, and the beads were suspended in 1 ml of cold phosphate-buffered saline (PBS).

GST Pull-Down Assay. Approximately 5 µg of GST-DRIP205 or GST-RAC3 bound on glutathione-Sepharose beads and 4 µl of [³⁵S]methionine-labeled mPPAR1 were incubated with the indicated concentrations of AJA, GW347845, or vehicle alone in H buffer (20 mM HEPES, pH 7.7, 75 mM KCl, 0.1 mM EDTA, 2.5 mM MgCl₂, 0.05% Nonidet P40, 0.1 mM methionine, and 1 mM dithiothreitol) containing 1 mg/ml bovine serum albumin on a nutator at 4°C overnight. After three washes with cold PBS, the bound proteins were eluted in SDS sample buffer and boiled for 10 min before SDS-PAGE and autoradiography. To ensure that equal amounts of GST fusion proteins were recovered in the pull-down assay, the gel was stained with Coomassie blue before autoradiography.

Cell Culture and Transient Transfection. For PPAR activation assay, the HEK293 cells were plated in 12-well cell culture plates and maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum at 37°C under 5% CO₂. Cells were changed to phenol red-free DMEM supplemented with 10% charcoal-stripped fetal bovine serum 3 h before transfection by a standard calcium-phosphate precipitation method. Twelve hours after transfection, the cells were washed with PBS and fed again with fresh medium containing the indicated concentrations of specified compounds. After 36 h, cells were harvested for -galactosidase and luciferase activities as described previously (Li et al., 1997). The average normalized luciferase activity was determined in triplicate experiments. For IL-8 promoter assay, HeLa cells were maintained and transfected as described above. After transfection, cells were recovered for 4 h before treatment with AJA, GW347845, or solvent. After 3 to 4 h, cells were treated with or without 25 nM of phorbol 12-myristate 13-acetate (PMA) for 24 h.

Adipocyte Differentiation Assay and RT-PCR. The adipocyte differentiation assay was performed as described by Mukherjee et al. (2000). The 3T3 L1 cells (American Type Culture Collection, Manassas, VA) were cultured in DMEM supplemented with 10% calf serum. Two days after reaching confluence, cells were treated with AJA, GW347845, or vehicle (0.1% DMSO) in the presence of 10 µg/ml insulin every other day. After 10 days of treatment with AJA or 7 days with GW347845 at confluence, cells were fixed and stained with Oil Red O (Sigma, St. Louis, MO). For RT-PCR, total cellular RNAs were isolated by TRIzol (Invitrogen, Carlsbad, CA). Reverse transcriptase (RT)-PCR was performed using the superscript first-strand synthesis kit (Invitrogen). After first-strand cDNA was synthesized by use of oligo(dT), cDNA was amplified by PCR. The forward and reverse primers used in the amplifications were 5'-GCT GTT ATG GGT GAA ACT CTG GGA G-3', and 5'-CTT CAT GAG GCC TGT TGT AGA GC-3', respectively, for PPAR2, 5'-GAG CAA ATG GAG TTC CCA GAT G-3' and 5'-GCA AAC AAT GGG AAT AGT TCA CAG TAG-3', respectively, for aP2, and 5'-GAC CAC AGT CCA TGC CAT CAC-3' and 5'-CAT ACC AGG AAA TGA GCT GAC-3', respectively, for GAPDH. PCR amplifications were performed in 50 µl volume with TaqDNA polymerase for 30 cycles.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Binding of Ajulemic Acid to PPAR. AJA is a synthetic dimethylheptyl analog of the THC metabolite THC-11-oic acid (Fig. 1). To test whether AJA binds to PPARs, we conducted partial proteinase digestion assays as described previously (Leng et al., 1993). Ligand binding is known to cause conformational changes of nuclear receptors, a crucial process for transcriptional activation by the receptors (Allan et al., 1992). Such conformational change can be detected by limited proteinase digestion based on alteration of the accessibility of proteolytic sites on the surface of the receptor. In vitro translated ³⁵S-labeled PPAR proteins were incubated with AJA for 60 min before being subjected to partial trypsin digestion. The proteolytic patterns of PPAR proteins were then analyzed by SDS-PAGE and autoradiography and compared with vehicle-treated control samples.

View larger version (43K): [in this window] [in a new window]   Fig. 2.   Binding of AJA to PPAR. The PPAR proteins were synthesized and labeled with [35S]methionine in reticulocyte lysate and incubated with AJA, GW compounds, or vehicle alone at room temperature for one hour before trypsin digestion. The reactions were terminated by boiling in SDS-containing protein sample buffer. The PPAR proteolytic patterns were analyzed by SDS-PAGE and autoradiography. A, AJA treatment protects mPPAR from trypsin digestion. The final concentrations of ligands were 20 µM AJA and 1 µM GW347845 in 0.2% DMSO (vehicle). Partial proteinase digestions were conducted with 30 µg/ml of trypsin for 30 or 60 min as indicated. Note that the two trypsin resistant fragments (30 and 24 kDa) were better protected from trypsin digestion in the presence of AJA or GW347845. Two additional smaller proteolytic fragments were observed only in the vehicle control. B, AJA concentration-dependent protection of mPPAR against trypsin digestion. The PPAR treated with the indicated concentrations of AJA was digested with 20 µg/ml of trypsin for 30 min. A clear protection of PPAR could be observed starting at 2 µM AJA. C and D, trypsin sensitivity of PPAR and PPAR in the absence or increasing concentrations of AJA. In contrast to PPAR, PPAR, and PPAR were not protected from trypsin digestion in the presence of 2 to 100 µM AJA, whereas the PPAR-specific ligand GW32 and the PPAR-specific ligand GW36 partially protected PPAR and PPAR from proteinase digestion, respectively.

Transcriptional Activation by PPAR upon Ajulemic Acid Treatment. The above observation indicates that AJA induces a conformational change in PPAR in a manner similar to a known PPAR agonist. This led us to speculate that AJA might be a PPAR agonist capable of activating PPAR. Therefore, transient transfection assay was performed using the PPRE-TK-LUC reporter (Kliewer et al., 1994) cotransfected with PPAR expression vector into the human kidney HEK293 cells followed by AJA treatment of the transfected cells. In this assay, 1 µM GW347845 stimulated the transcriptional activity of PPAR by 4-fold. Intriguingly, AJA also stimulated PPAR transcriptional activity 2- to 3-fold at 1 to 10 µM concentrations in a dose-dependent manner (Fig. 3A). These results demonstrate that AJA can indeed activate transcriptional activity of PPAR, consistent with its binding to PPAR and induction of an active conformation.

View larger version (29K): [in this window] [in a new window]   Fig. 3.   Transcriptional activation of PPAR by AJA. The reporter gene assay was performed in HEK293 cells by transient transfection with the indicated PPAR expression vectors and the PPRE-TK-LUC reporter. The average luciferase activities were normalized with the cotransfected -galactosidase in triplicate experiments. After transfections, cells were treated with indicated concentrations of ligands or vehicle alone for 36 h before harvesting for -galactosidase and luciferase assays. A, the mouse PPAR is activated by GW347845 and AJA in a concentration-dependent manner. B to C, the mouse PPAR and PPAR were not activated by AJA in the reporter gene assay, whereas they were activated by GW32 and GW36, respectively. D, AJA does not activate RXR. The hRXR and a DR1-TK-LUC reporter were cotransfected into HEK293 cells, followed by 9-cis-RA or AJA treatments. Unlike the 9-cis-RA, AJA does not activate hRXR homodimer on the DR1 element. E, activation of PPAR by AJA requires AF-2 helix of the receptor. The AF-2 helix truncated PPAR AF2 mutant could not be activated by AJA or GW347845, suggesting that AJA and GW347845 activate the ligand-dependent AF-2 function of the receptor.

View larger version (30K): [in this window] [in a new window]   Fig. 4.   Transcriptional activation by Gal4 DBD/PPAR fusion protein in response to AJA treatment. A, schematic of the Gal4 based reporter system. The PPARs were expressed as Gal4 DBD fusion proteins, which bind to the UAS promoter containing four copies of a synthetic Gal4 binding site upstream of the minimal thymidine kinase (TK) promoter. Transient transfection was conducted in HEK293 cells with the mPPAR expression vector and the MH100-tk-Luc reporter. The Gal4 DBD/PPAR fusion protein activates the reporter gene expression in response to agonist binding. B, PPAR was significantly activated by 20 µM AJA or 1 µM GW347845 in the Gal4 reporter system. C, AJA concentration-dependent activation of mPPAR. Transfected cells were treated with increasing concentrations of AJA from 10 to 60 µM, and the luciferase activities were determined in triplicated experiments. The estimated EC50 for activation of PPAR by AJA is 13 µM in this assay. D, activation of human PPAR by AJA. The Gal4 DBD/hPPAR fusion protein activates the reporter gene expression in a concentration-dependent manner in the Gal4 DBD system. In contrast, AJA failed to activate hPPAR and hPPAR. Normalized luciferase activity was plotted as fold activation relative to vehicle control.

AJA Stimulates Coactivator Binding to PPAR. Nuclear receptor coactivators are known to interact with ligand-activated receptors to enhance transcriptional activation by recruiting chromatin modifying enzymes and RNA polymerase. The receptor-associated coactivator-3 (RAC3) of the p160/SRC family (Leo and Chen, 2000) and the DRIP205 subunit of the DRIP coactivator complex (Yang et al., 2000), are known PPAR coactivators. To provide insight into the mechanism whereby AJA influences PPAR transcriptional activity, we examined the ability of PPAR to interact with DRIP205 and RAC3 in response to AJA treatment by GST-pull down assay. In this experiment, the [³⁵S]methionine-labeled PPAR showed negligible binding to GST alone, GST-DRIP205, or GST-RAC3 in the absence of ligand (Fig. 5, A and B). In contrast, AJA and GW347845 pretreatment caused significantly increased binding of PPAR to GST-DRIP205 and GST-RAC3. The interaction between PPAR and DRIP205 seemed to be stronger than the interaction between PPAR and RAC3, consistent with the finding that DRIP205 is a more potent coactivator than RAC3 for PPAR (Zhu et al., 1997; Yang et al., 2000). These data indicate that AJA treatment promotes the interaction of PPAR with nuclear receptor coactivators, further corroborating the hypothesis that AJA binds directly to PPAR and activates its transcriptional activity.

View larger version (28K): [in this window] [in a new window]   Fig. 5.   Binding of AJA enables coactivator recruitment to PPAR. GST pull-down assay shows interactions of PPAR with the coactivators DRIP205 and RAC3. The GST-DRIP205 (amino acids 527-970) and the GST-RAC3 (amino acids 613-752) fusion proteins were purified and used in binding reactions containing in vitro translated 35S-labeled PPAR in the presence of vehicle (0.1% DMSO), GW347845 (GW34, 1 µM), or AJA (20 µM). GST alone was used as a negative control in the pull-down reaction. A, the PPAR interacted with the GST-DRIP205 in the presence of GW347845 or AJA. B, the PPAR also interacted with GST-RAC3 in the presence of GW347845 or AJA but to a lesser degree than the interactions with GST-DRIP205.

Effect of PPAR Activation by AJA on IL-8 Promoter Activity. In an effort to link activation of PPAR by AJA to the anti-inflammatory activity of AJA, we determined the effect of AJA on IL-8 promoter activity and the role of PPAR in this process. Transient transfection assays were conducted using an IL-8 promoter-regulated luciferase reporter cotransfected with wild-type or PPAR AF2 mutant in mammalian cells. IL-8 is a biomarker for inflammation, and reduction of IL-8 levels correlates with a decrease in inflammation. The involvement of PPAR in regulating IL-8 promoter activity was determined by comparing the IL-8 promoter activity in the presence of wild-type PPAR or its AF2 mutant. In the experiment illustrated in Fig. 6A, PMA stimulated IL-8 promoter activity 3-fold, which was set as 100% promoter activity. AJA treatment reduced the IL-8 promoter activity by about 40% at 20 µM concentration in cells transfected with the wild-type PPAR. This effect seems concentration-dependent, and the reduction of IL-8 promoter activity is statistically significant at 10 and 20 µM AJA concentrations. AJA had no effect on IL-8 promoter activity in cells transfected with the PPAR AF2 mutant (Fig. 6B), suggesting that the reduction of IL-8 promoter activity is mediated through activation of PPAR by AJA. As a control, GW347845 dramatically reduced IL-8 promoter activity in cells cotransfected with the wild-type PPAR but not with the AF2 mutant (Figs. 6, C and D). These data indicate that activation of PPAR by AJA is involved in the inhibition of IL-8 promoter activity, suggesting a potential mechanism for the anti-inflammatory action of AJA.

View larger version (36K): [in this window] [in a new window]   Fig. 6.   Down-regulation of IL-8 promoter activity by AJA. The IL-8 luciferase reporter controlled by the human IL-8 promoter (97/69) was cotransfected with mPPAR or mPPAR AF2 mutant into HeLa cells. After transfection, cells were treated with AJA, GW347845, or solvent after PMA treatment for 24 h. PMA was able to induce IL-8 promoter activity 3-fold above untreated cells, which was set as 100% IL-8 promoter activity. A, in PPAR -transfected cells, AJA reduces the PMA-activated IL-8 promoter activity in a concentration (micromolar)-dependent manner. B, AJA had no effect on IL-8 promoter activity in the PPAR AF2-transfected cells, suggesting that the inhibition of IL-8 promoter activity by AJA depends on the transcriptional activity of PPAR. C, GW347845 reduces the IL-8 promoter activity in cells transfected with wild-type PPAR. D, in cells transfected with the PPAR AF2 mutant, GW347845 had no effect on IL-8 promotor activity.

Induction of Adipocyte Differentiation by Ajulemic Acid. Activation of PPAR by its ligands is an essential process for the onset of adipocyte differentiation, which is characterized by morphological changes, droplet formation, and induction of adipocyte-specific genes such as PPAR2 and aP2 (Tontonoz et al., 1994; Barak et al., 1999; Rosen et al., 2002). PPAR ligands can substitute for the adipogenic hormones during differentiation of preadipocytes into adipocytes (Chawla and Lazar, 1994), and ectopically expressed PPAR is able to transdifferentiate myoblasts into adipocytes (Hu et al., 1995). To provide biological evidence that AJA is a bona fide activator for PPAR, we assessed the ability of AJA to induce differentiation of 3T3 L1 fibroblasts into adipocytes. The cells were treated for 10 days with AJA, or 7 days with GW347845, and lipid accumulation in cells was assessed by Oil Red O staining. A dramatic increase in lipid droplet staining in the cytoplasm was observed after treatment with AJA or GW347845 (Fig. 7A), suggesting that both AJA and GW347845 can induce differentiation of 3T3 L1 fibroblasts into adipocytes.

View larger version (115K): [in this window] [in a new window]   Fig. 7.   Induction of adipocyte differentiation by AJA. 3T3 L1 fibroblasts were cultured in DMEM supplemented with 10% calf serum. Two days after confluence, cells were treated with 0.1% DMSO, 20 µM AJA, or 1 µM GW347845 in the presence of 10 µg/ml of insulin. A, Oil Red O staining for detection of adipocyte differentiation. After 7 days of treatment with GW347845 and 10 days of treatment with AJA, cells were fixed and stained with Oil Red O. Red staining indicates lipid droplets in the cytoplasm. B, RT-PCR analysis of adipocyte specific genes. PCR products were analyzed on 1% agarose gel and stained with ethidium bromide. PPAR2 and aP2 were induced significantly after treatment with 1 µM GW347845 or 20 µM AJA compared with vehicle-treated cells. The housekeeping gene GAPDH expressed equally in all samples.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In the present study, we investigated the molecular basis for the therapeutic action of AJA, a promising agent for relieving pain and inflammation. We find that AJA binds selectively to PPAR in vitro and AJA activates the transcriptional activity of PPAR in vivo. Activation of PPAR by AJA depends on the presence of the AF-2 helix in the receptor. AJA binding enables PPAR to recruit nuclear receptor coactivators. In addition, AJA inhibits IL-8 promoter activity in a PPAR-dependent manner and induces differentiation of 3T3 L1 fibroblasts into adipocytes. Our data suggest that AJA may exert its therapeutic actions at least in part through transcriptional activation of PPAR.

The structural and functional similarity between AJA and several known PPAR ligands led us to consider AJA as a potential PPAR ligand. Indeed, in a partial proteinase digestion assay, AJA effectively protects PPAR from proteinase digestion (Fig. 2), reflecting direct binding of AJA to PPAR. The two trypsin resistant fragments probably contain the LBD of PPAR, because ligand binding induces a compact conformation of the LBD (Xu et al., 2001), which is expected to be more resistant to proteinase digestion compared with the unliganded receptor. Because AJA causes proteinase resistance only to PPAR, but not PPAR or PPAR, it is clear that AJA binds selectively to PPAR. Indeed, the three PPARs have distinct ligand binding specificities (Xu et al., 2001) and physiological functions (Berger and Moller, 2002). The selective binding of AJA to PPAR suggests that PPAR may mediate the therapeutic activity of AJA.

Consistent with the binding of AJA to PPAR, the transient reporter gene assay further demonstrates that AJA activates the target promoter through PPAR but not PPAR or PPAR (Fig. 3). These data indicate that AJA is a PPAR-specific agonist, a conclusion that is further supported by the inability of AJA to activate the PPAR AF2 mutant, which is defective in ligand-dependent transcriptional activation (Fig. 3E). The observation that PPAR activation by AJA requires the AF-2 function is consistent with the hypothesis that AJA binds to PPAR LBD and activates its ligand-dependent transcriptional function. By using the Gal4-DBD fusion protein system, the EC₅₀ of AJA for PPAR activation was measured at 13 µM (Fig. 4), a concentration that is within the pharmacologically effective doses of AJA (Burstein, 2000). Furthermore, AJA also activates human PPAR equally well compared with mouse PPAR, implying that the PPAR-dependent therapeutic activity of AJA observed in mice might be applicable to humans.

In line with the hypothesis that AJA is an agonist for PPAR, binding of AJA enables PPAR to recruit nuclear receptor coactivators (Fig. 5). Nuclear receptor coactivators are known to interact with liganded receptors to facilitate transcriptional activation of target promoters by recruiting histone acetyltransferase activity to the receptor (Leo and Chen, 2000). The interaction of PPAR with DRIP205 in response to AJA treatment seems to be more prominent than the interaction with RAC3. This is consistent with a report showing that DRIP205 is a potent coactivator for PPAR (Yang et al., 2000). AJA may induce formation of a coactivator-binding surface to allow docking of the coactivator LXXLL motif. Furthermore, addition of AJA causes differentiation of 3T3 L1 fibroblasts into adipocytes (Fig. 7), a process that is mediated by PPAR. Together, these data provide strong evidence to suggest that AJA is an agonist of PPAR. These data also implicate a potential involvement of PPAR in the signaling of this cannabinoid class of analgesic and anti-inflammatory drugs.

It is likely that activation of PPAR is at least partly responsible for the anti-inflammatory action of AJA and perhaps other cannabinoids as well. In fact, several natural cannabinoids, such as THC and cannabidiol, also activate PPAR (J. Liu and J. D. Chen, unpublished data). The involvement of PPAR in AJA-mediated anti-inflammatory activity is further supported by the observation that AJA inhibits IL-8 promoter activity in a PPAR-dependent manner (Fig. 6). This inhibition occurs only in the presence of wild-type PPAR, not the AF-2 mutant, suggesting that transcriptional activation by PPAR is required for the repression of IL-8 promoter activity by AJA. It is not clear whether inhibition of cytokine promoter activity by PPAR is caused by direct binding of PPAR to the promoter. Because there is no evidence for direct binding of PPAR to the cytokine promoter, this inhibition may be indirect. Treatment of peritoneal macrophages with several PPAR ligands, such as 15-deoxy-^12,14-prostaglandin J₂ also suppresses expression of the inducible nitric-oxide synthase, gelatinase B, and scavenger receptor A in response to phorbol ester stimulation. The promoters of these genes were found to possess binding sites for activator protein-1, nuclear factor B, and the signal transducer and activator of transcription. Furthermore, the inhibition of inflammatory response by PPAR ligands in macrophages was produced in part by antagonizing the activities of these transcription factors (Ricote et al., 1998). Although 15-deoxy-^12,14-prostaglandin J₂ was shown in some studies to be able to mediate an anti-inflammatory action in a PPAR-independent manner (Straus et al., 2000; Tsubouchi et al., 2001), the inhibition of inflammatory responses by activation of PPAR was confirmed by several recent studies both in vitro (Ji et al., 2001) and in vivo (Dubuquoy et al., 2000; Kawahito et al., 2000; Naito et al., 2001). It is interesting to note that, like the activation of PPAR, the immunosuppressive functions of some cannabinoids were also reported to be exerted by the inhibition of activator protein-1, nuclear factor B, and the signal transducer and activator of transcription (Jeon et al., 1996; Zheng and Specter, 1996; Faubert and Kaminski, 2000).

The complete mechanism of the anti-inflammatory action of AJA is likely to involve a complicated signaling network. PPAR is highly expressed in several immune cells, such as macrophages and monocytes. Therefore, AJA may activate PPAR in these cells to modulate immune inflammatory responses. Further studies are needed to investigate whether other cannabinoids are capable of activating PPAR and whether other cytokines or chemokines are inhibited by AJA. Because PPAR is involved in several physiological processes including lipid metabolism, glucose homeostasis, and adipocyte differentiation, the discovery that AJA is an active ligand for PPAR suggests that AJA may have a broader range of pharmacological activities than previously expected. Activation of PPAR by several synthetic and natural compounds is known to correlate well with the ability of these compounds to control glucose concentrations. Therefore, it may be interesting to test in the future whether AJA is a useful agent for treatment of diabetes mellitus.