_ 33“.“... 15.2.3115. 1......r:§ . . u h . win... .543... .u 5.. i. . 5...... .- .5. . 1.. :H u. .5... air? .5... m r... 13.... r r cum H... n .1: c 3.. . E. “39?... mursl. H. :zflfiflfiwflm .3. (i. a... . z m. Kuwfixfi Ash”... .4. 3.. . n. _ $.23. .93.“... . I n...“ n? W. d! a..- at... vi 4. v v . l...... :1... and..- 4.2!. 1.11.9le1.». . . .. . _ .323... u... ,maéfinfiw.wwhfiiwfiflmmflwfiamfimfi_, ..m.....w..._»..w. . _ . fi?§¥®§?2§uw.%.$a ..._....m§.a% .. \ I {‘5\ r‘ r) A , a /7 “*3"; Q) J {‘3 / LIBRARIES MICHIGAN STATE UNIVERSITY EAST LANSING, MICH 48824-1048 This is to certify that the dissertation entitled INHIBITION OF lNTERLEUKIN-Z SECRETION BY 2-ARACHIDONYL GLYCEROL AND ANANDAMIDE OCCURS THROUGH PEROXISOME PROLIFERATOR ACTIVATED RECEPTOR y INDEPENDENTLY OF THE CANNABINOID RECEPTORS presented by CHERYL ELIZABETH ROCKWELL has been accepted towards fulfillment of the requirements for the PHARMACOLOGY 8. TOXICOLOGY PHD. degree in r Major Professor’s Signature 3/29/0 S" Date MSU is an Affinnative Action/Equal Opportunity Institution PLACE IN RETURN Box to remove this checkout from your record. To AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE 2/05 c':/c'incxomoua.m-p.ts fir— INHIBITION OF INTERLEUKIN-Z SECRETION BY 2-ARACHIDONYL GLYCEROL AND ANANDAMIDE OCCURS THROUGH PEROXISOME PROLIFERATOR ACTIVATED RECEPTOR ‘y INDEPENDENTLY OF THE CANNABINOID RECEPTORS By Cheryl Elizabeth Rockwell A DISSERTATION Submitted to Michigan State University in partial fiilfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Pharmacology and Toxicology 2005 ABSTRACT INHIBITION OF INTERLEUKIN-z SECRETION BY 2-ARACHIDONYL GLYCEROL AND ANANDAMIDE OCCURS THROUGH PEROXISOME PROLIFERATOR ACTIVATED RECEPTOR 'y INDEPENDENTLY OF THE CANNABINOID RECEPTORS By Cheryl Elizabeth Rockwell 2-Arachidonyl glycerol (2-AG) and anandamide (ABA) are endogenous arachidonic acid derivatives which activate the cannabinoid receptors, CB1 and CB2, hence termed endocannabinoids. 2-AG and AEA modulate a variety of immunological responses, including induction of transient calcium influx in HL-60 cells, modulation of lymphocyte proliferation, and suppression of cytokine production: As is observed with the plant-derived cannabinoids, 2-AG and AEA suppress interleukin (IL)-2 production. The overall goal of these studies was to determine the mechanism of the inhibitory effects of ABA and 2-AG upon IL-2 secretion. The suppression of IL-2 by 2-AG and AEA in splenocytes derived from CB1/CB2 null mice coupled with the failure of the CB1/CB2 antagonists to block the decrease in IL-2 by ABA and 2-AG, demonstrates that the cannabinoid receptors are not involved. Interestingly, arachidonic acid causes a concentration-dependent suppression of IL-2 secretion, which was similar to that of structurally-related ABA and 2-AG. The decrease in IL-2 by ABA and 2-AG was partially reversed by pretreatment with the nonspecific cyclooxygenase (COX) inhibitor, flurbiprofen, as well as the COX-2 specific inhibitor, NS398, suggesting that COX-2 metabolites of 2-AG and AEA are responsible for the inhibitory effects upon IL-2, rather than the parent molecules. Because peroxisome proliferator activated receptor 7 (PPAR'y) activation has been correlated with IL-2 suppression in T cells and a number of COX metabolites are known PPARy agonists, the ability of 2-AG to activate PPAR'y was investigated. Both 2-AG and 2-AG ether, a non-hydrolyzable analogue of 2-AG, activate PPAR'y, as evidenced by differentiation of 3T3 -L1 cells into adipocytes, induction of aP2 mRN A levels, and activation of a PPARy -specific luciferase reporter in transiently transfected 3T3-L1 cells. Consequently, the putative role of PPARy in IL-2 suppression by 2-AG and 2-AG ether was examined. IL-2 suppression by 2-AG and 2-AG ether in activated T cells was blocked by T0070907, a potent PPAR'y -specific antagonist. Similarly, T0070907, also blocked ABA-mediated IL-2 suppression. Additionally, 2-AG was also found to inhibit the transcriptional activity of nuclear factor of activated T cells (NF AT) and nuclear factor KB (NF KB), transcription factors that are critical for IL-2 transcription. In addition to its effects upon IL-2, 2-AG suppresses the transcription of IL-4 and IFN'y, cytokines that are also regulated by NFAT. Moreover, the inhibition of NFAT and NFKB transcriptional activity by 2-AG was abrogated in the presence of TOO70907. Collectively, the aforementioned studies provide evidence that suppression of IL-2 by COX-2 metabolites of 2-AG and AEA is mediated through activation of PPAR'Y independently of CBl/CBZ. In addition, evidence is provided that PPAR'y activation by 2-AG or its metabolite inhibits the transcriptional activity of NFAT and NFKB, which ultimately results in suppression of IL-2 by activated T cells. DEDICATION This dissertation is dedicated to my husband, Bryan, for his love and support throughout all the good times as well as the less-than-great times that have marked these crazy years at graduate school, and to my parents, for a lifetime of support and encouragement. iv ACKNOWLEDGEMENTS I am deeply grateful to my thesis advisor, Dr. Norbert Kaminski, for giving me an opportunity to learn the critical aspects of becoming an independent researcher. I have found our research project to be more stimulating than I ever could have imagined. I have truly enjoyed pouring my energy into it. I also thank you for all of your advice and support, which has been invaluable. I would also like to thank the members of my thesis committee, Dr. James Pestka, Dr. Robert Roth, and Dr. Stephanie Watts, for all of their insight and for some very stimulating discussions during committee meetings. I have found the comments to be both useful and interesting. I also thank you for lending me chemical reagents, antibodies, and laboratory equipment, to further my research project. I would be remiss if I did not acknowledge the assistance of my husband, Dr. Bryan Copple. Thank you for allowing me to requisition your computer for statistical analysis and for showing me how to stain cells and to photograph them. I have shamelessly taken advantage of you. Additionally, I would like to acknowledge all the members of my laboratory. To Natasha Snider, thanks for being a great friend, a tremendous source of support, as well as for all the technical assistance and tag-teaming on the project. To Gautham Rao, thanks for tolerating more than four years’ worth of bad jokes and for sharing some great times in Italy. To Kim Townsend, thanks for the good times both inside and outside the lab. To Dr. Barb Kaplan, Dr. Courtney Sulentic, John Buchweitz, Dina Schnaider, Bob Crawford, Dr. Alison Springs, and Dr. Susan McKarns, thanks for lots of laughs and many entertaining conversationsll You have made the lab a great place to work!! TABLE OF CONTENTS LIST OF FIGURES ............................................................................... ix LIST OF ABBREVIATIONS .................................................................. xiii LITERATURE REVIEW II. Cannabinoid receptors ............................................................... 1 11. Cellular signaling of cannabinoid receptors ............................... 1 III. Expression of CB1 ............................................................ 4 IV. Physiological effects of CB1 activation .................................... S V. Expression of CB2 ............................................................ 6 VI. Physiological effects of CB2 activation .................................... 7 VII. Cannabinoid receptor specific agonists, antagonists, and knockout mice ................................................................. 8 III. Endocannabinoids ................................................................... 8 A. Identification of the endogenous cannabinoids ........................... 8 B. AEA biosynthesis and physiological concentrations ..................... 9 C. Metabolism of ABA ......................................................... 13 D. AEA binding affinity for CB1/CB2 and cellular signaling ............ 16 B. Other targets of ABA ........................................................ 19 F. 2-AG biosynthesis and physiological concentrations .................. 20 G. Metabolism of 2-AG ........................................................ 24 H. 2-AG binding affinity for CB1/CB2 and cellular signaling. . . . . . . . 25 IV. Therapeutic potential of endogenous cannabinoids ........................... 27 V. Other putative endogenous cannabinoids ....................................... 28 VI. Immunomodulatory activity of cannabinoids ................................... 29 VII. Regulation of IL-2 .................................................................. 31 A. IL-2 function, receptors, and clinical importance ....................... 31 B. NFAT ......................................................................... 32 C. NFKB, AP-l, Oct, and CD28 .............................................. 34 D. Modulation of IL-2 production by cannabinoids ........................ 35 VIII. COX .................................................................................. 36 IX. PPARs ............................................................................... 37 A. PPAR subtypes .............................................................. 37 B. PPAR'Y isoforms ............................................................. 37 C. PPAR'y immune effects ..................................................... 38 X. Rationale ............................................................................. 40 MATERIALS AND METHODS 1. Reagents ............................................................................. 43 II. Animals .............................................................................. 43 III. Cell Lines ........................................................................... 44 IV. IL-2 protein quantification ........................................................ 45 vi V. 3T3-Ll differentiation assay and oil red staining .............................. 46 VI. Real-time PCR ...................................................................... 46 A. Reverse transcription and amplification .................................. 46 B. Quantification by AACt protocol .......................................... 47 VII. Transient transfection assay ...................................................... 47 A. 3T3-L1 cells (PPARy-LBD/Gal4-DBD, Gal4-luc) ..................... 47 B. Jurkat T cells ................................................................. 48 C. Plasmids ....................................................................... 48 D. Luciferase Assay (chemiluminescence method) ........................ 48 VIII. Statistical analysis .................................................................. 49 EXPERIMENTAL RESULTS 1. Effect of ABA, 2-AG, and 2-AG ether upon IL-2 secretion in primary splenocytes activated with PMA/ionomycin .................................... 50 11. Role of the cannabinoid receptors, CB1 and CB2, in the suppression of IL-2 secretion by AEA, 2-AG, and 2-AG ether ................................ 54 III. The role of the vanilloid receptor, VRl , in ABA-mediated IL-2 suppression .................................................................... 64 IV. Effect of 2-AG and AEA upon calcium influx in resting splenocytes and activated splenocytes and thymocytes ...................................... 65 V. The effect of arachidonic acid upon IL-2 secretion and the roles of FAAH, MAG lipase, and the AMT in the suppression of IL-2 by ABA and 2-AG ........................................................ 73 VI. The role of COX in the suppression of IL-2 by ABA and 2-AG ............ 84 VII. Effect of prostanoids and PPAR'y agonists upon IL-2 production ........... 99 VIII. The activation of PPAR'y by 2-AG and 2-AG ether .......................... 107 IX. The role of PPAR‘y in the suppression of IL-2 by 2-AG, 2-AG ether, AEA ......................................................... 1 19 X. The role of NFAT, NFKB, and AP-l in the suppression of IL-2 by 2-AG .................................................................. 127 DISCUSSION I. Effects of AEA, 2-AG, and 2-AG ether upon IL-2 secretion ............... 142 11. Role of the cannabinoid receptors in the suppression of IL-2 by 2-AG, 2-AG ether and AEA .................................................... 143 III. Role of the vanilloid receptor, VRl, in the suppression of IL-2 by AEA ............................................................................ 146 IV. Effects of ABA and 2-AG upon calcium influx .............................. 148 V. Role of hydrolysis in the suppression of IL-2 by ABA and 2-AG ......... 150 VI. Role of COX in the suppression of IL-2 by ABA and 2-AG ............... 151 VII. Role of PPAR'y in the inhibition of IL-2 secretion by 2-AG, 2-AG ether, and AEA ............................................................ 153 vii VIII. Role of NFAT, NFKB, and AP-l in the suppression of IL-2 secretion by 2-AG ............................................................................ 158 IX. Summary .......................................................................... 159 LITERATURE CITED ......................................................................... 163 viii 10. 11. 12. 13. 14. 15. 16. 17. LIST OF FIGURES Structures of various cannabinoids .................................................... 10 Schematic representation of ABA biosynthesis and catabolism ................... 12 Schematic representation of 2-AG biosynthesis and catabolism .................. 21 Schematic representation of alternative modes of 2—AG biosynthesis ............ 22 Schematic representation of the IL-2 minimal essential promoter region ........ 33 Schematic representation of the hypothesis .......................................... 42 Effect of ABA upon PMA/ionomycin-stimulated IL-2 production in murine primary splenocytes ...................................................................... 51 Effect of 2-AG and 2-AG ether upon PMA/ionomycin-stimulated IL-2 production in murine primary splenocytes ............................................ 53 Effect of cannabinoid receptor antagonists on ABA-mediated suppression of PMA/ionomycin-stimulated IL-2 production ..................................... 56 Effect of cannabinoid receptor antagonists on suppression of IL-2 production by 2-AG and 2-AG ether .................................................. 58 Effect of the CB2-specific agonist, JWH-133, upon PMA/ionomycin— stimulated IL-2 production in murine primary splenocytes ........................ 59 Effect of ABA upon PMA/ionomycin—stimulated IL-2 production in human Jurkat T cells .................................................................. 61 Effect of 2-AG and 2-AG ether upon PMA/ionomycin-stimulated IL-2 production in human Jurkat T cells .................................................... 63 Effect of ABA upon PMA/ionomycin-stimulated IL-2 production in splenocytes derived from CB1/CB2 null and wild-type mice ..................... 66 Effect of 2-AG upon PMA/ionomycin-stimulated IL-2 production in splenocytes derived from CB1/CB2 null and wild-type mice ..................... 68 Effect of 2-AG ether upon PMA/ionomycin-stimulated IL-2 production in splenocytes derived from CB1/CB2 null and wild-type mice ..................... 70 Effect of capsaicin upon PMA/ionomycin-stimulated IL-2 production in ix 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. murine primary splenocytes ............................................................ 71 Effect of the vanilloid receptor antagonist, capsazepine, on AEA-mediated suppression of PMA/ionomycin-stimulated IL-2 production ...................... 72 Effect of ABA and 2-AG upon intracellular calcium ............................... 75 Effect of 2-AG upon calcium influx by ionomycin and concanavalin A ......... 77 Effect of arachidonic acid upon PMA/ionomycin-stimulated IL-2 secretion in primary murine splenocytes ............................................. 78 Effect of AMT inhibitors on suppression of IL-2 secretion by AEA ............. 81 Effect of the FAAH inhibitor, MAF P, on suppression of IL-2 secretion by ABA and arachidonic acid .......................................................... 83 Effect of the AMT inhibitor, AM404, and the F AAH inhibitor, MAFP, on suppression of IL-2 secretion by 2-AG ............................................... 86 Effect of the nonselective COX inhibitor, flurbiprofen, upon suppression of IL-2 secretion by ABA and arachidonic acid ..................................... 89 Effect of the COX-1 selective inhibitor, piroxicam, upon suppression of IL-2 secretion by AEA and arachidonic acid ..................................... 91 Effect of the COX-2 specific inhibitor, NS398, upon suppression of IL- 2 secretion by AEA and arachidonic acid .............................................. 93 Piroxicam and NS398 attenuate the suppression of IL-2 secretion by AEA in a concentration-dependent manner .......................................... 95 Effect of the COX-1 specific inhibitors, SC560 and FR122047, on the suppression of IL-2 secretion by AEA ........................................... 97 Effect of the COX-1 specific inhibitor, SC560, on arachidonic acid— mediated suppression of PMA/ionomycin-stimulated IL-2 production .......... 98 Effect of the nonselective COX inhibitor, flurbiprofen, on 2-AG-mediated suppression of PMA/ionomycin-stimulated IL-2 production ..................... 100 Effect of the COX-1 selective inhibitor, piroxicam, on 2-AG-mediated suppression of PMA/ionomycin-stimulated IL-2 production ..................... 101 Effect of the COX-1 specific inhibitors, SC560 and FR122047, on 2-AG- mediated suppression of PMA/ionomycin-stimulated IL-2 production ......... 103 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 45. 46. 47. 48. Effect of the COX-2 specific inhibitor, NS398, on 2-AG-mediated suppression of PMA/ionomycin—stimulated IL-2 production .................... 104 The effect the prostaglandin inhibitor, AH6809, on the suppression of IL-2 secretion by ABA and the effect of PGEz-ethanolamine on IL-2 production ..................................................................... 106 Effect of PGJz and 15-deoxy-PGJ2 upon IL-2 secretion ........................... 109 Effect of ciglitazone and ETYA upon IL-2 secretion .............................. 111 Differentiation of 3T3-L1 cells by 2-AG ............................................ 114 Differentiation of 3T3-L1 cells by 2-AG ether, ciglitazone, and differentiation media ................................................................... 116 Quantification of 3T3-L1 differentiation ............................................ 117 Induction of aP2 by the PPAR'Y agonist, ciglitazone, is attenuated by the PPARy antagonist, GW9662 ...................................................... 118 Effects of 2-AG and 2-AG ether upon aP2 production ............................ 121 Effects of 2-AG and 2-AG ether upon luciferase activity in 3T3-L1 cells transfected with PPAR'y-LBD/Gal 4-DBD Gal 4 luciferase reporter ............ 123 Effect of the PPARy antagonist, T0070907, upon suppression of IL-2 by 2-AG in primary murine splenocytes and human Jurkat T cells ............. 125 Effect of the PPARy antagonist, T0070907, upon suppression of IL-2 by 2-AG ether in human Jurkat T cells .............................................. 126 Effect of the PPARy antagonist, T0070907, upon suppression of IL-2 by AEA in primary murine splenocytes ............................................. 128 Effects of 2-AG upon IFN'y and IL-4 production in primary splenocytes ...... 130 Effect of PMA/ionomycin and PHA/PMA upon NFAT transcriptional activity ................................................................................... 132 xi 49. 50. 51. 52. 53. 54. 55. Effect of 2-AG upon NFAT transcriptional activity ............................... 133 Effect of the PPAR'y antagonist, T0070907, upon 2-AG-mediated suppression of NFAT transcriptional activity ....................................... 136 Effect of 2-AG upon AP-l transcriptional activity ................................. 137 Effect of T0070907 and 2-AG upon AP-l transcriptional activity ............... 138 Effect of 2-AG upon NFKB transcriptional activity ............................... 139 Effect of the PPAR'y antagonist, T0070907, upon 2-AG-mediated inhibition of NFKB transcriptional activity .......................................... 141 Schematic representation of suppression of IL-2 by 2-AG and AEA in activated T cells ......................................................................... 162 xii AEA 2-AG AMT ANOVA AP-l CB1 CB2 COX CP55940 CREAE DAG DMEM EAE ELISA ERK ETYA FAAH FR122047 GW9662 HETE LIST OF ABBREVIATIONS Anandamide (arachidonyl ethanolamine) 2-Arachidonyl glycerol Anandamide membrane transporter Analysis of variance Activator protein-1 Cannabinoid receptor 1 Cannabinoid receptor 2 Cyclooxygenase (-)-cis-3-[2-hydroxy-4-(1,l-dimethylheptyl) phenyl]-trans- 4-(3-hydroxypr0pyl) cyclohexanol Chronic relapsing experimental allergic encephalomyelitis Diacylglycerol Dulbecco’s modified Eagle’s medium Experimental allergic encephalomyelitis Enzyme-linked immunosorbent assay Extracellular signal regulated kinase Eicosatetraynoic acid Fatty acid amidohydrolase 1 -[[4,5-bis(4-methoxyphenyl)-2-thiazolyl]carbony1]4- methylpiperazine 2-Chloro-5-nitro-N-pheny1benzamide Hydroxyeicosatetraenoic acid xiii IKB IFN IL 1P3 JWHOIS JWH133 NAPE LPS MAFP MS NFkB NFAT NMDA NO NS398 Oct PBS PEA PG 1 Sd-PGJz Inhibitor of NFKB Interferon Interleukin Inositol 1,4,5-triphosphate (2-Methy1— 1 -propy1—1H-indol-3-yl)-1 - naphthalenylmethanone (6aR, 1 OaR)-3 -(1 ,1 -dimethylbutyl)-6a,7,1 0,1 Oa-tetrahydro- 6,6,9-trimethyl-6H-dibenzo [b,d]pyran N-arachidonyl phosphatidylethanolamine Lipolysaccharide Methyl arachidonyl fluorophosphonate mitogen activated protein Multiple sclerosis Nuclear factor of K B Nuclear factor of activated T cells N—methyl-D-aspartic acid Nitric oxide N-[-2-(cyclohexyloxy)-4-nitrophenyl]-methanesulfonamide Octamer phosphate buffered saline Palmitoylethanolamine Prostaglandin lS-deoxy-A'z’” prostaglandin J2 xiv PHA PKC PL PMA PPAR PPRE SC560 SC58125 SR141716A SR144528 A9-THC TNFa T0070907 UCM707 VRl Phytohemagglutinin Protein kinase C Phospholipase Phorbol 12-myristate 13-acetate Peroxisome proliferator activated receptor PPAR response element Retinoid X receptor 5-(4-chlorophenyl)-1 —(4-methoxyphenyl)-3- (trifluoromethyl)- 1 H-pyrazole 5-(4-fluorophenyl)-1-[4-(methylsulfonyl)phenyl]-3- (trifluoromethyD-lH-pyrazole N-(piperidin-l-yl)-5-(4-chlorophenyl)-1 -(2,4- dichlorphenyl)-4-methyl-H-pyrazole—3 carboxyamidehydrochloride N-[(1 S)-endo-1 ,3,3-trimethyl-bicyclo[2,2,1]heptan-2-y1]-5- (4-chloro-3-methylpheny1)-1-(4-methylbenzyl)-pyrazole-3- carboxamide Delta 9-tetrahydrocannabinol Tumor necrosis factor at 2-chloro-5-nitro-N-4-pyridinyl-benzamide N-(3-fi1ranylmethyl)-SZ,8Z,1 12,14Z-eicosatetraenamide Vanilloid receptor 1 XV WIN55212-2 VVIN55212-3 R (+)-[2,3-dihydro-5-methyl-3-(4- morpholinylmethyl]pyrrolo[l ,2,3-de]-1 ,4-benzoxazin—6- y1]-1 -naphthaleny1methanone mesylate S (-)-[2,3-dihydro-5-methyl-3-(4- morpholinyhnethyl]pyrrolo[1,2,3-de]-1 ,4-benzoxazin-6- yl]-1-naphthalenylmethanone mesylate xvi LITERATURE REVIEW 1. Cannabinoid Receptors Twoicannabinoid receptors have been isolated and cloned to date, CB1 and CB2 (1-3). Both CB1 and CB2 have seven transmembrane domains and are G-protein coupled. A splice variant, CBlA, has also been discovered, although its physiological significance has yet to be determined (4). CB1 and C82 are 44% homologous, which increases to 68% when considering only the transmembrane domains that compose the ligand binding site. Although there is considerable overlap in the ligands for CB1 and CB2, the receptors differ substantially in distribution and in activity. A. Cellular signaling of cannabinoid receptors Agonist-induced activation of both CB1 and CB2 inhibits adenylate cyclase in a variety of different model systems (1, 5-8). Inhibition of adenylate cyclase by both CB1 and CB2 is pertussis toxin-sensitive, indicating that both receptors associate with Gm, proteins (6, 7, 9, 10). Conversely, CB1 activation has also been shown to stimulate adenylate cyclase in pertussis toxin-treated cells, suggesting that in the absence of Gm, proteins, CB1 can associate with Gs (11). It has been hypothesized that CB1 and CB2 can couple to multiple isoforms of adenylate cyclase and that the result of cannabinoid receptor activation is dependent upon which adenylate cyclase isoforms are present (12). Activation of CB] and CB2 has also been correlated with ion channel regulation. Modulation of cAMP levels by CB1 has been shown to activate A-type potassium channels in rat hippocampal cells (13). Similarly, exogenously expressed CB1 in AtT-20 pituitary cells induces inwardly rectifying potassium channels in a pertussis toxin- sensitive manner (14, 15). In addition to potassium channels, CB1 activation has also been associated with modulation of calcium channel activity. Cannabinoid-mediated inhibition of L-type calcium channels is blocked with CB1 antagonist pretreatment and has been correlated with vasorelaxation of feline cerebral arterial rings (16). It has also been demonstrated that CB1 activation inhibits N-type calcium channels through Gm, (17- 21). Likewise, CB1 activation has also been associated with a pertussis toxin-sensitive inhibition of Q-type calcium channels in AtT-20 pituitary cells expressing recombinant CB1 as well as P/Q-type calcium channels in rat cortical or cerebellar brain slices (15, 22). In addition to its ability to inhibit voltage-gated calcium channels, CBl activation also induces a rapid, transient increase in intracellular calcium in certain cell types. Cannabinoid-evoked transient calcium influx in NG108-15 neuroblastoma/glioma cells, is sensitive to pertussis toxin and inhibition of CB1 and phospholipase C (PLC), suggesting that activation of PLC by 01/0 (B/Y subunit) induces the release of inositol- 1,4,5-triphosphate (1P3) that subsequently results in calcium influx (23-25). Likewise, cannabinoids augment depolarization-induced calcium influx in cultured cerebellular granule cells, which is also sensitive to pertussis toxin and inhibition of CB1 and PLC (26). Similar to CB1 activity in NG108-15 cells, CB2 activation has been correlated with the induction of transient calcium influx in HL-60 cells, which is blocked with a CB2 antagonist (27). Furthermore, pretreatment with both CB] and CB2 antagonists attenuates calcium influx by cannabinol in primary splenocytes, suggesting that CB1 and CB2 also modulate calcium channels in this cell type (28). Interestingly, A9-THC induces a robust increase in intracellular calcium in primary splenocytes, which is also inhibited with both C31 and CB2 antagonists, but is also observed in CB1/CB2 null mice (29) (unpublished observations). The aforementioned results suggest that another receptor may exist which is also activated by Ag-THC and inhibited by CB1 and CB2 antagonists. Consequently, studies characterizing the efl‘ects of CB1 and CB2 solely through the use of cannabinoid receptor antagonists may need to be reevaluated. CB1 and CB2 have also been implicated in the activation of mitogen activated protein (MAP) kinase in a number of different cell types (30-32). Cannabinoid-mediated MAP kinase activation is inhibited by pertussis toxin and a CB1 antagonist in C6 glioma cells and primary astrocytes (33, 34). In U373 MG astrocytoma cells, phosphatidylinositol 3-kinase (PI3K) appears to be involved with CB1-mediated MAP kinase activation as protein kinase B is phosphorylated upon CB1 activation and PI3K inhibitors block MAP kinase signaling by CB1 (35). The activation of MAP kinase by CB1 has been correlated with the expression of a number of immediate early genes, such as krox-24, c-fos, and c-jun (36-40). Similarly, CB2 activation has also been associated with MAP kinase activation and an upregulation of krox-24 (41). Conversely, cannabinoids have also been shown to inhibit MAP‘kinase activation and immediate early genes under certain conditions, such as cellular activation (42). In addition to their effects upon calcium channel regulation and MAP kinase signaling, one or both cannabinoid receptors have also been linked to the activation of a number of other second messengers, such as cyclic guanosine monophosphate (cGMP), focal adhesion kinase, ceramide, and NO (43-46). B. Expression of CB1 CB1 receptors are expressed in a number of different areas within the central nervous system (CNS) but have also been detected in a variety of peripheral tissues, including immune cells at low levels, sympathetic ganglia, reproductive tissues, gastrointestinal tissues, urinary bladder, heart, adrenal gland, lung, spleen, and the pituitary gland (47-53). Within the brain, CB1 is highly expressed in the cerebral cortex, hippocampus, arnygdala, lateral caudate-putamen, substantia nigra pars reticulata, globus pallidus, entopeduncular nucleus, basal ganglia, and the cerebellum (54-58). The distribution pattern of CB1 in neuronal tissues is thought to be consistent with the ability of cannabinoids to diminish locomotor activity, produce catalepsy, as well as to impair cognition and memory (59). The expression of CB1 in the basal ganglia may have clinical applications for the treatment of diseases which are caused by the degeneration of basal ganglia neurons, such as Parkinson’s disease and Huntington’s disease (60). Additionally, CB1 has also been detected in the neuronal tissues of pain pathways in the spinal cord and brain, which may be responsible for cannabinoid-mediated analgesia (61). Conversely, there are relatively few CB1 receptors in the brainstem, which may explain the lack of toxicity associated with high doses of Ag-THC and other cannabinoids (62). Several studies have demonstrated that CB1 is presynaptically localized on both inhibitory and excitatory neuronal fibers, suggesting a modulatory role in neurotransmission (58, 63-68). Consistent with the aforementioned findings, CB1 activation has been correlated with inhibition of GABA release from hippocampal interneurons and suppression of glutamate secretion from cerebellar basket cells (69, 70). In addition to its role in short-term neurotransmission, CB1 activation has also been shown to inhibit long-term potentiation in rat hippocampal slices (71). C. Physiological effects of CB1 activation A variety of different physiological effects have been attributed to CB1 in a number of different systems, including the cardiovascular, gastrointestinal, reproductive, and the CNS. Within the CNS, CB1 activation results in decreased locomotion, analgesia, hypothermia, immobility, prolonged sleep duration, and hyperphagia (72-80). Additionally, activation of CB1 has been demonstrated to reduce spasticity in chronic relapsing experimental allergic encephalomyelitis (CREAE), an animal model of multiple sclerosis (MS), which has contributed to the interest in the clinical use of cannabinoids for the treatment of MS. In the cardiovascular system, CB1 activation causes bradycardia and vasorelaxation, which is not observed in CB1 null mice (81-85). Interestingly, CB1 may also play a role in reproduction as CB1 is highly expressed in the uterus and cannabinoids have been correlated with inhibition of embryo implantation and increased numbers of stillbirths (86-90). Within the gastrointestinal system, there has been considerable clinical interest in the ability of CB1 activation to prevent emesis (91 , 92). Indeed, A9-THC is prescribed in the US. and other countries for this purpose. Additionally, CB1 activation has been correlated with certain immunological effects, although CB1 is typically expressed at much lower levels than CB2 in immune cells. CB1 activation inhibits nitric oxide (NO) in primary murine astrocytes as well as in rat microglial cells and feline macrophages stimulated with interferon (IFN) y/lipopolysaccharide (LPS) (45, 93, 94). In a mouse peritonitis model, CB1 has been shown to suppress neutrophil migration to the peritoneal cavity, which has been attributed to a delay in neutrophil chemokine production (95). Additionally, agonist- activated CB1 is associated with inhibition of interleukin (IL) -12 and tumor necrosis factor (TNF) or and induction of the immunosuppressive cytokine, IL-1 0, in LPS-treated mice (96, 97). Conversely, CB1 activation has also been correlated with immunostirnulatory effects, such as induction of IL-6 in mouse astrocytes (96). D. Expression of CB2 While CB1 is highly expressed in the CNS, CB2 is predominantly expressed in immune cells. In addition to immune cells, CB2 hasalso been detected in astrocytes, C6 glioma cells, rat oligodendrocytes, rat retina, embryonic rat liver, rat placenta, and rat uterus (98-101). Interestingly, CB2 is upregulated in neuritic plaque-associated glia in Alzheimer’s disease as well as in the lumbar spinal cord in peripheral nerve injury (102, 103). Within the immune system, CB2 has been detected in B cells, natural killer cells, monocytes, macrophages, neutrophils, cytotoxic T cells, helper T cells, mast cells, dendritic cells, and microglial cells (53, 104-110). In addition, CB2 is differentially expressed in macrophages and microglia relative to activation state. CB2 is upregulated in IFN'y-primed macrophages and microglia and thioglycollate-treated macrophages, but down-regulated in LPS-treated macrophages and microglia (107). Additionally, CB2 is downregulated in LPS-treated splenocytes and upregulated in CD40-stimulated splenocytes (106, 111). Interestingly, CB2 is also upregulated in the peripheral blood mononuclear cells of marijuana smokers (112). E. Physiological effects of CB2 activation - CB2 activation has been correlated with a number of immune effects in a variety of different models. Agonist-induced CB2 has been shown to enhance the proliferation of CD40-stimulated B cells, while conversely inducing apoptosis in thymocytes and splenocytes (113, 114). In primary murine astrocytes and feline macrophages, CB2 activation has been associated with inhibition of LPS-induced NO release (45, 94). Activated CB2 has also been shown to inhibit macrophage-induced activation of helper T cells, which is likely due to diminished antigen processing by the macrophages (115, 116). Furthermore, CB2 activation has been correlated with induction of cell migration in a number of different immune cell types, including microglia, myeloid precursor cells, differentiated HL-60 cells, and human peripheral blood monocytes (117-120). Migration of HL-60 cells may be related to the induction of the chemokines, monocyte chemotactic protein-1 (MCP-l) and IL—8, by CB2 (121). In addition to its effects upon the migration of myeloid precursor cells, CB2 also blocks neutrophilic differentiation in this cell type (119, 122). Modulation of cytokine release has also been reported to occur through CB2 activation, including induction of the immunosuppressive cytokine, transforming growth factor B (TGF-B), in hmnan peripheral blood lymphocytes, and suppression of TNFor in human mononuclear cells (123, 124). Diminished overall antitumor immunity has also been attributed to CB2, resulting in larger tumors in cannabinoid-treated animals which is blocked with a CB2 antagonist (125). In addition to effects upon the immune system, CB2 activation has also been associated with antinociceptive effects to thermal stimuli as well as to formalin (126, 127). F. Cannabinoid receptor specific agonists, antagonists, and knockout mice Synthetic antagonists have been developed for both CB1 and CB2. SR141716A, AM251, and AM281 are CB1 antagonists, which are widely used (128-130). Likewise, SR144528 and AM630 are the most widely used CB2 antagonists (131, 132). SR141716A and SR144528 were developed first, while AM251, AM281, and AM630 were developed more recently. There are also a number of CB1 antagonists which are not commercially available (133). In addition to the antagonists, CBl- and CB2-specific agonists have also been developed. Arachidonyl-2’-chloroethylamide (ACEA) and arachidonyl cyclopropylamide (ACPA) are potent CB1-specific agonists, whereas JWH133 is a potent CB2-specific agonist (134, 135). JWH015 is less potent than JWH133, but is also CB2-selective agonist that is widely used (136). Perhaps the most powerful tools that have been developed for the identification of CB1/CB2-mediated activity are transgenic animals, including CB1 heterozygous knockout mice, CB1 homozygous knockout mice, CB2 homozygous knockout mice, and CB1/CB2 homozygous double knockout mice (84, 116, 137, 138). II. Endocannabinoids A. Identification of the endogenous cannabinoids With the discovery of the cannabinoid receptors, a search for the endogenous ligands ensued. Isolated from porcine brain, anandamide (AEA) was the first endocannabinoid discovered (139). Shortly thereafter, a second endogenous ligand was isolated by two separate laboratories simultaneously (140, 141). 2-Arachidonyl glycerol (2-AG) was isolated from canine gut by Mechoulam’s group, who also demonstrated that 2-AG possessed cannabimimetic activity, and from rat brain by Sugiura’s laboratory, who determined that 2-AG is extremely sensitive to hydrolysis. Both AEA and 2-AG are arachidonic acid derivatives, and are structurally distinct from plant-derived and synthetic cannabinoids (Figure 1). Prior to its identification as a cannabinoid receptor agonist, 2- AG was regarded as a degradation product of inositol phospholipids and as a potential source of arachidonic acid under certain conditions, but was not considered to have biological activity of its own (142). B. AEA biosynthesis and physiological concentrations Two different pathways have been proposed for the synthesis of AEA (Figure 2). The first pathway, initially described by Deutsch and Chin in 1993, is the direct condensation of arachidonic acid and ethanolamine (143). Although other laboratories have confirmed that AEA can be formed from free arachidonic acid and ethanolamine, the physiological relevance of this has been called into question due to the exceptionally high concentrations of ethanolamine required (144-150). The direct condensation of free arachidonic acid and ethanolamine is now believed to occur through the enzyme, fatty acid amidohydrolase (FAAH), which is also considered to be the principal enzyme responsible for the catabolism of ABA (146, 148). While FAAH generally hydrolyzes AEA into arachidonic acid and ethanolamine, it is also thought to catalyze the reverse reaction, resulting in AEA synthesis. The evidence for the role of F AAH in AEA synthesis comes from the ability of FAAH inhibitors to block the condensation of arachidonic acid and ethanolamine (150). SR141716A (c131 ANTAGONIST) SR144528 (c132 ANTAGONIST) K3 W III-7° 159-THC CP55,940 OH OH NW 0—<: CEO/é m .. ANANDAMIDE 2-ARACHIDONYL GLYCEROL .. . _ _ OH .. o—C __ OH ARACHIDONIC ACID 2-ARACHIDONYL ervcshor. amen Figure 1. The structures of various cannabinoids. The putative endogenous cannabinoids, AEA, 2-AG, and 2-AG ether, are structurally related to arachidonic acid. Ag-THC is a plant-derived cannabinoid. SR141716A and SR144528 are antagonists of CB1 and CB2, respectively. CP55,940 is a synthetic agonist of both CB1 and CB2. 10 Due to the extraordinarily high concentrations of substrate needed for the condensation pathway, it is considered to be a minor mode of ABA biosynthesis. The bulk of ABA synthesis is believed to occur through a transacylase/phosphodiesterase pathway (Figure 2). According to the proposed pathway, the addition of an arachidonyl group from a donor phospholipid (thought to be phosphatidylcholine) onto phosphatidylethanolamine occurs through a transacylase to produce N-arachidonyl phosphatidylethanolamine (NAPE). NAPE is then hydrolyzed by phospholipase D (PLD), producing ABA and phosphatidic acid (151, 152). One troubling aspect of this pathway is that it is dependent upon the arachidonyl group being esterified in the sn-1 position of the donor phospholipid, which rarely occurs. As a result, it is tempting to postulate that there may be other pathways that also contribute to AEA synthesis. Nonetheless, there is evidence to suggest that this pathway is at least partially responsible for AEA production in N18TG2 neuroblastoma cells, J774 macrophages as well as in the rat brain and testis (153-156). Additionally, it has also been demonstrated that both the transacylase and PLD are calcium dependent, which corresponds to ionomycin- stirnulated AEA release from N18TGZ and J774 cells (155-157). Calcium influx is not the only stimulus for AEA synthesis as simultaneous stimulation of NMDA and nicotinic receptors has also been shown to result in increased AEA levels in cortical neurons (158). Within the immune system, AEA has been detected in U937 lymphoma cells, macrophages, and dendritic cells (108, 154, 159). Furthermore, it has been demonstrated that AEA is induced in LPS-treated macrophages (160). While the physiological concentrations of AEA vary somewhat between tissue types, AEA has been detected in mouse brain (10-15 pmol/g tissue), human brain (25-150 11 92:323.: :2 BVAW /\/= a 8.2.2.8...— .— o . L Eva—2.33 can anon—«533: a. $305 conga 28 a w_w>._OW_D>I :o /o 33: .93226852 228083 .285 _e8>_u_>8_o / .285 _>coo_._oa<-~ O: :Olc :OIO .. .. . +1008 3.... 9.3:... 4 ._... ..>>\\_o 382633.. .1 LJXXUU o 88:988.... 8388& 66:30:35 638533230533 Ba 22228283 38259: .< 08923“. 3888»... §853B~fi8i 22 into DAG, which is then subsequently cleaved by DAG lipase (Figure 3) (231, 232). While the biological significance of the multiple proposed routes of 2-AG biosynthesis remains largely undetermined, there is evidence to suggest that the PLC/DAG lipase pathway is active in cortical neurons and platelets (71, 142). Likewise, the phosphatidic acid pathway appears to operate in mouse neuroblastoma cells and rat microglial cells (231, 23 2). While the upstream events that result in the activation of PLC, phosphatidic acid phosphohydrolase, and DAG lipase have not been well characterized, increased intracellular calcium and glutamate receptor activation result in elevated 2-AG synthesis. Likewise, the activation of P2X; purinergic receptors has been implicated in 2-AG biosynthesis in murine astrocytes and microglial cells (233, 234) . Activation of immune cells, such as macrophages and dendritic cells, with LPS or platelet activating factor also results in increased 2-AG levels (108, 235, 236). Physiological concentrations of 2-AG are generally higher than those of ABA in most cell types and can exceed the levels of AEA by as much as nearly 3 orders of magnitude. 2-AG levels have been detected in rat brain (3-65 nmol/g tissue), human brain (65 nmng tissue), rat liver (1.15 nmng tissue), rat spleen (1.17 nmng tissue), rat lung (0.78 nmol/g tissue), and rat kidney (0.98 nmng tissue) (71, 159, 237). Additionally, 2-AG has been detected in rat plasma (12 11M), human sera (10 nM), and human sera from patients with endotoxic shock (30 nM) (164). The accuracy of the reported 2-AG levels is not entirely clear due to a number of confounding factors. 2-AG is highly unstable, both chemically and metabolically, which results in the rapid isomerization and hydrolysis of 2-AG under a variety of different experimental 23 conditions. Furthermore, it has been reported that the half-life of 2-AG under culture condi tions ranges from 2 to 10 minutes, depending upon the presence and concentration of serum or albumin (238). Moreover, it has also been reported that a number of the detection techniques commonly used are inefficient in the detection of both 2-AG and AEA ( 164). G. Metabolism of Z-AG 2-AG hydrolysis occurs through at least two different pathways (Figure 3). In addition to AEA, FAAH can also hydrolyze 2-AG into arachidonic acid and glycerol (239) . FAAH-mediated hydrolysis of 2-AG has been demonstrated in several different cell types but is not likely to be the primary mode of 2-AG hydrolysis. While AEA has Sigllificantly greater binding affinity for CB1 in FAAH null mice, there is no difference in title binding affinity of 2-AG for car between FAAH null and wild-type control arliIfilals (240). Furthermore, the silencing of monoacylglycerol (MAG) lipase through RNA interference results in marked increases in 2-AG accumulation, suggesting that MAG lipase is the primary mechanism of 2-AG hydrolysis (241). Like FAAH, MAG lipase has been cloned and it has been determined that the two enzymes are not hotrlologous (242, 243). Additionally, there is evidence to suggest that 2-AG may also be transIaorted into some cell types through facilitated diffusion by either the putative AMT or a similar mechanism (244). In addition to hydrolysis, 2-AG is also undergoes oxidative metabolism. As 0bServed with AEA, 2-AG is metabolized directly by COX-2, but is not a substrate for COX-1 (245). The products of 2-AG metabolism by COX-2 include, 12- 24 hydroxyheptadecatrienoic acid glyceryl ester (G), PGEz-G, PGDz-G, PGan-G, thromboxane (Tx) Bz-G, a hydrolysis product of TxAz-G, as well as 6-keto-PGFla-G, a hydrolysis product of prostacyclin-G (177). The efficiency of COX-2 and the corresponding PG synthases to metabolize 2-AG is comparable to their ability to metabolize arachidonic acid. In contrast, metabolism of PGHz-G by thromboxane synthase is markedly less efficient than the free acid, which suggests that the formation of TxAz-G may not be physiologically relevant. Furthermore, platelets, which express only COX-l , are the predominant source for TxAz in vivo. Additionally, 2-AG binds to and is oxidized by leukocyte l2-LOX and 15-LOX but is not efficiently metabolized by human 5-LOX or platelet 12-LOX (177, 246). The metabolites generated fi'orn 2-AG metabolism by LOX have been identified as 12(S)- HETE-G and 15-HETE-G. Although the physiological activity of the LOX metabolites is largely unknown, 15-HETE-G, which is the result of 2-AG metabolism by 15-LOX, has been found to be a ligand for peroxisome proliferator activated receptor a (247). It is unknown whether 2-AG, like ABA and arachidonic acid, can be metabolized by cytochrome P450 enzymes. H. 2-AG binding affinity for CB1/C82 and cellular signaling Initial radioligand binding experiments with 2-AG demonstrated that 2-AG has lower affinity for both CB1 (K, = 472 nM) and CB2 (K, = 1400 nM) than AEA (140). Subsequent experiments demonstrated that the affinity of 2-AG for the cannabinoid receptors is likely higher than was initially reported. The more recent studies to examine 2-AG binding to CB1 and CB2 report K values of 58 and 145 nM for CB1 and CB2, 25 respectively (248). Despite the comparable affinity for CB1 and CB2, 2-AG exhibits greater activity in a variety of different cell systems than AEA, which is due to greater efficacy through CB1 and CB2. 2-AG is a full agonist of CB1 as determined by activation of G and G0 as well as by inhibition of cAMP accumulation (71, 198, 200,). Likewise, 2-AG also produces maximal inhibition of cAMP accumulation through CB2 activation (200). While modulation of ion channels by AEA has been the focus of a number of published studies, fewer studies have explored the effects of 2-AG upon ion channel activation. A recently published study found, however, that 2-AG inhibits L-type calcium channels and activated inwardly rectifying potassium channels (Girk 1 and 4) through CB1 in rat sympathetic neurons (249). Additionally, 2-AG also mediates a transient calcium release in neuroblastoma cells, which is inhibited with a CB1 antagonist (24, 25, 250, 251). Likewise, 2-AG induces a transient calcium current in HL-60 cells, which is blocked with a CB2 antagonist and pertussis toxin (27). Like AEA, 2-AG modulates MAP kinase pathways in a variety of different models. Activation of ERK] and ERK2 by 2-AG in I-IL-60 cells appears to be mediated by c132 as it can be blocked with a 0132 antagonist as well as by pertussis toxin (252). Similarly, 2-AG also activates ERK 1 in microglial cells in a CB2-dependent manner, which results in increased proliferation (232). In murine hippocampal slices, 2-AG activates p38 MAP kinase in wild-type mice, but not in CB1 null mice (208). Interestingly, activation of ERKs by 2-AG, like AEA, results in downregulation of prolactin receptors and subsequently reduced proliferation of breast cancer cells (211, 253,254) 26 III. Therapeutic potential of endogenous cannabinoids There has been considerable interest in the therapeutic potential of ABA and 2- AG. In particular, there has been considerable energy devoted to the inhibitors of FAAH and the putative AMT, which could be used clinically to raise endogenous AEA levels. There is diverse array of clinical conditions which might benefit fiom increased AEA concentrations, such that FAAH/AMT inhibitors are currently being investigated for their use as analgesics, anxiolytics, antispasmodics, cancer chemotherapeutics, anti-emetics, and for the treatment of neuropsychiatric disorders, such as schizophrenia (173, 254-268). Arguably, the analgesic effects of the F AAH/AMT inhibitors have. received the majority of interest thus far. FAAH null mice have been found to possess a higher baseline for pain perception, which may be because AEA levels have been found to be 15 times higher in FAAH null mice than wild-type controls (255). Additionally, treatment of wild-type mice with a CB1 antagonist or antisense oligonucleotides against CB1 mRNA results in hyperalgesia, which is thought to be due to constitutive endocannabinoid tone (269). Moreover, several AMT inhibitors have been shown to produce analgesia in mice treated with a subeffective dose of ABA (173, 256, 257, 270). While the majority of focus has been upon the increase of endogenous AEA through FAAH/AMT inhibitors, a similar strategy could be employed for 2-AG. With an increasing number of studies reporting physiological effects of 2-AG that are of therapeutic interest, the clinical benefits of increased levels of 2-AG for a number of different conditions are currently being investigated. 27 IV. Other putative endogenous cannabinoids In addition to ABA and 2-AG, there are several other putative endogenous cannabinoids which have been identified. Early binding studies revealed the existence of two other lipids in brain tissue, homo-gamma-linolenylethanolamide and docosatetranylethanolamide, which also bind to CB1 (271). Additionally, palmitoylethanolrnine (PEA) was shown to have analgesic effects, similar to AEA but of longer duration, which were blocked with a CB2 antagonist (170, 272). Notably, it was later determined that PEA does not bind to CB2 and therefore cannot activate the receptor directly. It is still unclear whether the ability of the CB2 antagonist, SR144528, to block PEA-mediated analgesia is due to non-specific effects, blockage of a CB2-like receptor, or some other mechanism. There have also been a number of metabolically stable analogs of both ABA and 2-AG that have been developed to avert rapid hydrolysis in tissue culture as well as in whole animal studies. Methanandamide and fluoromethanandarnide are more resistant to hydrolysis by FAAH than AEA (273-275). In contrast, 2-fluoroanandamide is equally susceptible to hydrolysis by FAAH as AEA, but has increased binding affinity for CB1 (192). Likewise, 2-AG ether was developed as a metabolically stable analog of 2-AG (25). In contrast to the synthetic AEA analogs, 2-AG ether has been detected in murine brain tissue and consequently has been named as another endogenous cannabinoid (276). This may be premature, however, as other investigators report that they are unable to detect 2-AG ether in brain tissue from a variety of different mammals, including rat, mouse, and pig (277). 28 V. Immunomodulatory activity of cannabinoids The earliest observations of the immunomodulatory effects of cannabinoids were in human patients. Juel-Jensen noted more frequent recurrences of herpes simplex virus infections in his patients who smoked marijuana (278). This small study observing effects in only a few patients was followed by a number of larger controlled studies. While the immune effects of marijuana in humans remains a contentious issue, published studies indicate that high doses of marijuana cause a significant drop in immunoglobulin production by B cells (279). The effect of low to moderate doses of cannabis upon immune cells in humans remains a debatable issue. In vivo studies performed in animals indicate that A9-THC exposure is correlated with a decreased resistance to bacterial, viral, and protozoan infections. A9-THC has been shown to increase susceptibility of animals to herpes simplex virus as well as to enhance the progression of the infection (280, 281). A9-THC has also been shown to cause an increased susceptibility to bacterial infections, such as Listeria monocytogenes and Treponema pallidum (280, 282). More recently it has been demonstrated that Ag-THC exacerbates brain infection by opporttmistic amebae of the genus Acanthamoeba, which is associated with decreased production of the cytokines, IL-1 and TNFa (283). Additionally, A9-THC causes immunosuppressive effects in immune cells from severely immunocompromised animals. Immune cells from mice infected with Friend leukemia virus, which causes profound immunodeficiency, were suppressed firrther upon Ag-THC exposure, such that T cells and NK cells were 100% inhibited (284). The increased susceptibility of animals treated with A9-THC to bacterial and viral infections may be due to some of the effects of Ag-THC upon macrophages, B cells, and 29 T cells. In macrophages, Ag-THC causes inhibition of phagocytosis as well as inhibition of NO production (285-287). In addition, A9-THC inhibits antigen processing, but not antigen presentation by macrophages (288). Inhibition of antigen processing in macrophages by A9-THC appears to be mediated through CBZ as the effect was completely blocked by the CB2 antagonist, SR144528 (115). B and T lymphocytes have also been shown to be sensitive to cannabinoids. Proliferation of B and T cells are both suppressed by cannabinoids in the uM range, however low concentrations of synthetic cannabinoids enhance proliferation of B cells (289-291). While A9-THC does not appear to affect T cell-independent production of irnmunoglobulins, it markedly inhibits antibody formation that is dependent upon T cells (290). Ag-THC as well as other plant- derived cannabinoids also cause a decrease in the production of cytokines, including IL-2 (292, 293). Because plant-derived cannabinoids modulate a variety of different immune - effects and AEA and 2-AG have been shown to mimic the effects of plant-derived cannabinoids, a number of studies have investigated the effects of AEA and 2-AG upon the immune system. AEA has been shown to modulate a number of immunological responses, including inhibition of NO and IL-6 production in macrophages as well as inhibition of TNFa and neutrophil recruitment in LPS-induced pulmonary inflammation in mice (294, 295). Additionally, AEA suppresses the release of IFN'y, TNFa, as well as the soluble TNFor receptor in human peripheral blood mononuclear cells and inhibits the migration of activated CD8+ T cells (296, 297). Whether these effects are in fact mediated through the cannabinoid receptors has yet to be rigorously examined, however, there are reports of CB2-mediated immune 3O effects by AEA and methanandarnide. Induction of the immunosuppressive cytokine, TGFB by methanandamide is blocked with the CB2 antagonist, SR144528 (123). Recently published studies suggest that AEA induces apoptosis of immature dendritic cells, which can be blocked with CB1 and CB2 antagonists (298). Likewise, a number of studies have also investigated the effects of 2-AG upon immune cells. 2-AG has been reported to induce calcium influx in HL-60 cells, increase NO production in human monocytes, enhance antibody formation in murine splenocytes, and to induce the migration of human peripheral blood monocytes and HL-60 cells (27, 120, 299, 300). 2- AG has also been shown to inhibit cytokine production, including TNFOL release from both LPS-treated rat microglial cells as well as murine macrophages, IL-6 production in J 774 macrophages, and IL-2 secretion in activated murine splenocytes (295, 301-303). While CB2 has been implicated in many 2-AG-mediated immune effects, the role of CB2 and/or CB1 in the immunosuppressive effects of 2-AG upon cytokine release has yet to be conclusively determined. VI. Regulation of IL-2 A. IL-2 function, receptors, and clinical importance IL-2 is an autocrine/paracrine factor secreted by activated T cells and is important for T cell survival, proliferation, and in some cases, differentiation (304). As such, IL-2 is a central cytokine for the development of an adaptive immune response. IL-2 is highly regulated both transcriptionally through a number of different transcription factors and post-transcriptionally through the stabilization of IL-2 mRNA. There are three IL-2 receptors: the high-affinity receptor, containing the IL-2R0t, IL-2RB, and IL-2R'y chains, 31 the intermediate receptor, containing the IL-ZRB and IL-2Ry chains, and the low-affinity receptor, which consists of IL-2Ra alone. The IL-2R'y chain is a shared component of several other cytokine receptors, including IL-4, IL-7, IL-9, and IL-15. Consequently, 7 chain deficiency in humans results in severe combined immunodeficiency. Conversely, IL-2R0t-deficient mice exhibit automimmunity, inflammatory bowel disease, and premature death. Similarly, IL-2RB-null mice exhibit elevated IgG, IgE, and autoantibodies as a result of overdifferentiation of B cells into plasma cells. B. NFAT The IL-2 minimal essential promoter comprises a number of different response elements, including the NFAT, NFKB, AP-l, Oct, and CD28RE sites (Figure 5). NFAT binds to the IL-2 promoter in two well-characterized sites, referred to as the proximal and distal NFAT binding sites, but at least three other putative NFAT binding sites have also been identified within the IL-2 promoter (3 05, 306). The binding of NFAT to the IL-2 promoter is essential for IL-2 transcription (307). The distal NFAT site requires cooperative binding of both nuclear and cytosolic NFAT components (308). The nuclear NFAT component is a heterodimer of jun and fos family members. The cytosolic NFAT component is related to the rel family of proteins and requires dephosphorylation by the Ca+2/calmodulin-dependent phosphatase, calcineurin, for activation (305, 309). Upon dephosphorylation of cytosolic NF AT by calcineurin, the nuclear localization sequence is exposed, causing translocation of cytosolic NFAT to the nucleus, where it can then bind with the nuclear component to the IL-2 promoter (310). Additionally, there is evidence 32 n . . _ v Hafiz define 38:83 33:88 3.538 aid 2: he coca—.323.— ecanAom .m 95me The, The. 30 . 1.1.1.211... {all} .0 ..r. I. ....r.,..,..... . .. ... “haunt... . ., a... ,.w..~a.......-ti.~.W "W...-.. ,_ , ... a. t“. W.“ . t 2. . . . w in. .i..........t..pz...nu.\.~.....i..,.n4 . \ awNGU Mural 54ml 33 to suggest that NFAT may also play a cooperative role with other transcription factors in binding to the Oct-1 and CD28RE binding sites (306, 311, 312). C. NFKB, AP-l, Oct, and CD28 Nuclear factor for K chain in B cells (N FKB) has one binding site in the IL-2 promoter (307). In resting T cells, NF KB exists as a dimer with inhibitor of NFtcB (IKB) in the cytoplasm. Upon T cell activation, IKB is phosphorylated by a number of different kinase 5. The primary kinase responsible for phosphorylation of IKB is IKB kinase (IKK), but I ICE can also be phosphorylated by PKA and PKC (313-317). Phosphorylation of [ICE cames it to dissociate from NFKB, such that NFKB can then translocate to the nucleus and bind to the IL-2 promoter. AP-l binds to the IL-2 promoter at two different sites, the proximal and distal AP- 1 bind ing sites (318). While AP-l binding at the distal site enhances IL-2 transcription, AP '1 binding at the proximal site seems to be essential for IL-2 transcription (318, 319). The AP-l protein is best known as a heterodimer of jun and fos family members (320, 321 ) - Additionally, the AP-l complex may also consist of other members of the b-ZIP superfarnily, such as CREB and CREB family members (321). Oct has two binding sites in the IL-2 promoter, which are bound by the conStitutively expressed Oct-1 and the lymphoid-specific Oct-2 (307, 322-324). The Oct- ] bitlding site is actually a composite NFAT:AP-1:Oct site, as cooperative binding occurs With both AP-l and NFAT (306). Maximal IL-2 production requires a costimulatory signal in addition to the acti\*ation of the T cell receptor (325). CD28 is a membrane protein that provides a 34 costimulatory signal during T cell activation, which results in the binding of transcription factors to the CD28RE. The stimulation of the T cell receptor and CD28 need to be temporme coincident and spatially proximal to produce full T cell activation (326). While the CD28 costimulatory pathway has not yet been fully elucidated, there is evidence to support that CD28 ligation results in activation of NF AT and NFKB (326- 330)- Furthermore, it has been determined that both NFAT and NFKB can bind to the CD28 BE, in addition to other members of the rel family of proteins (311, 312). D. Modulation of IL-2 production by cannabinoids A number of different cannabinoids have been shown to inhibit IL-2 production, includi ng A9-THC, cannabinol, cannabidiol, WIN55,ZI2-2, WINSS,212-3 and 2-AG (28, 292. 293, 331). While much insight has been gained into the mechanism of action of some of these compounds, much has yet to be learned about how cannabinoids inhibit IL- 2 Production. Cannabinol inhibits IL-2 secretion through inhibition of ERK MAP kinase activ ation, which causes a decrease in nuclear c-fos and ultimately leads to a decrease of AP“ 1 binding in the IL-2 promoter (42). In contrast, 2-AG inhibits NFAT binding and to a 1eSser extent NFKB binding to the IL-2 promoter in activated splenocytes (303). Co’L'?l\1ersely, cannabinol, cannabidiol and CP55,94O have also been shown to increase IL- 2 prOduction in suboptimally activated cells (332). Cannabinol-mediated enhancement of IL‘2 transcription is likely to occur due to activation of NFAT by calcium calmodulin kinase 11 and PKC (332, 333). 35 VII. Cyclooxygenase Cyclooxygenases (COX) are enzymes that catalyze the two-step formation of PGHz from arachidonic acid and oxygen, which is the committed step in prostanoid biosynthesis. The cyclooxygenase activity of COX initially converts arachidonic acid to PGGz, which subsequently is reduced to PGHz by a peroxidase reaction. PGH; is the common substrate for an array of different synthase enzymes that produce prostaglandins and thromboxanes, although some prostanoids can be produced from PGHz nonenzymatically (334). Two isoforms of COX have been identified. COX-1 is consti tutively produced in most cell types, although the levels can vary during different developmental stages and can be modulated in certain cell types (335, 336). COX-1 pI’OduCes an array of different PGs and TXs, which function to maintain homeostasis in a variety of different tissues, including the gastrointestinal tract, the kidneys, and in platelets, COX-2 is undetectable in most tissues, but is rapidly upregulated in certain cell types in response to a variety of agents, including growth factors, cytokines, LPS, tumor Promoters, and hormones (33 5, 336). Because COX-2 is believed to be the predominant iSOfOI‘m involved in prostanoid production during inflammation, a number of drugs have been developed to specifically inhibit COX-2 without affecting COX-l activity. Both COX enzymes have been found in most cell types within the immune system, including T cells (334). In purified human T cells and Jurkat T cells, cox-1 is constitutively expressed at relatively low levels, whereas COX-2 is induced with PMA/ionomycin tlrealtt*rlent (337). There is evidence to support that a number of different prostanoids can cause suppression of lL-2 secretion, including PGEz, 15-deoxy-A'2’” PG12 (lSd-PGJz), and PGIz 36 (338-340). Of the aforementioned eicosanoids, 15d-PGJ2, is the most recent to be recognized to suppress IL-2 secretion. There has been growing interest in lSd—PGJz in a number of different research areas due to its identification as one of the most potent endoge nous ligands of peroxisome proliferator activated receptor 7 (341). VIII. Peroxisome Proliferator Activated Receptors (PPARs) A. PPAR subtypes PPARs are members of the nuclear receptor superfarnily. There are three subtypes, PPARor, PPAR5 (also called PPARB and NUCl), and PPAR'y. PPARs form hetero dimers with retinoid X receptors (RXR), which are also nuclear receptors, and bind to perO xisome proliferator response elements (PPREs) found in the promoter regions of PPAR 1arget genes (342). PPARor is highly expressed in metabolically active tissues, 811011 as liver, heart, kidney, and muscle, where it plays a key role in regulating lipid metabolism (343). PPARa has also been recently discovered in B and T lymphocytes (342) - Clofibrate, and other members of the fibrate class of drugs, as well as a number of fatty acids are agonists of PPARor. Interestingly, a 15-LOX metabolite of 2-AG, 15- HETE-G, is also an agonist of PPARor (247). PPAR8 is ubiquitously expressed and although its physiological role is not well characterized, PPAR6 may also play a role in lipid metabolism (343). PGDz and other fatty acids are ligands of PPARS. B. PPAR‘Y isoforms PPARy has three different mRNA isoforms, 7,, 72, and 73. PPARyt is expressed in mmnne cells, such as macrophages, B cells, and T cells, and seems to play a role in the 37 regulation of a number of immune effects (342). Interestingly, activated PPAR'yt sequesters NF AT such that NFAT cannot bind to the IL-2 promoter causing an inhibition of IL-2 transcription, a process called transrepression (340). PPARYI has also been reported to transrepress NFKB and AP-l in T cells and macrophages (340, 344, 345). PPARyz is predominantly expressed in adipose tissue and plays a role in lipid homeostasis. The protein expressed by PPAR‘Y3 is identical to that expressed by PPAR'yl. The ligand binding domains of the PPARY subtypes are identical (342). Troglitazone and its analogs, as well as 15d-PGJ2, and other eieosanoids are agonists of PPAR'y. In addition to specific agonists, there are two commercially-available PPAR'Y antagonists, GW9662 and T0070907. While PPARY knockout mice are embryonic lethal, PPAR’y null macrophages have been developed from the creation of PPAR‘y null embryonic stem cells via homologous recombination and the subsequent in vitro differentiation of the stem cells into macrophages (346). C. PPAR? immune effects A number of immune effects are attributed to PPAR'Y , which have generally been suppressive in nature. In macrophages, PPAR’y ligands suppress inflammatory cytokines, such as IL-6, TNFor, lL-lB, IL-10, and IL-12 (344, 347, 348). Conversely, PPARy activation has also been demonstrated to result in the upregulation of pro-inflammatory surface receptors, including CD14, CD11b/CD18, and CD36 (349, 350). Decreased expression of inducible NO synthase and induction of apoptosis have also been attributed to PPAR‘Y in macrophages (351). The interpretation of some of these studies may need to be reviewed, however, as a recent report demonstrated that troglitazone, ciglitazone, 38 and 15d-PGJ; suppress TNFor and IL-6 production in both wild-type and PPARy null macrophages (346). While relatively few studies have investigated the role of PPAR'y in B cells, a number of effects have been attributed to PPAR'y in T cells. PPAR'y activation has been associated with decreased cytokine production, inhibition of proliferation, as well as both the induction or prevention of apoptosis (340, 344, 352-354). The effects of PPAR‘Y upon apoptosis in T cells may be dependent upon a variety of factors, including activation state and the particular model (transformed cells appear to be more sensitive to PPAR? —mediated apoptosis) (353-3 59). While apoptosis may be responsible for some of the immunosuppressive effects of PPAR‘y in T cells, a number of studies have reported that PPARy activation suppresses cytokines, such as IL-2, IL-4, and IFN'y (340, 345, 352, 360-362). Many investigators have suggested that the suppression of cytokine production by PPARY has pathophysiological and clinical implications. Consequently, there has been considerable interest in the therapeutic potential of PPAR'y ligands for autoimmune diseases. Indeed, a number of different animal models of autoimmune disease have been ameliorated with PPARy agonist treatment, including allergic asthma and experimental crescentic glomerulonephritis (363-365). The therapeutic effects of PPAR'y agonist treatment appear to involve suppression of T cell function in certain autoimmune models, such as autoimmune myocarditis, autoimmune diabetes, and experimental allergic encephalomyelitis (EAE), which is an animal model of multiple sclerosis (MS) (366- 369). EAE, in particular, has been the focus of a number of studies, which have demonstrated that PPARy agonists suppress T cell proliferation and cytokines, including 39 IFN'y, IL-4, and IL-10 in this model (367, 368, 370). Furthermore, PPARy deficient heterozygous mice develop an exacerberated form of EAE, which suggests that constitutively expressed PPAR'y may be protective against EAE (367). Moreover, PPARy agonists have been shown to inhibit T cell proliferation and suppress cytokine production in T cells derived from human MS patients as well (371). Interestingly, 2-AG and AEA also ameliorate the symptoms of EAE (258). Suppression of T cell function by PPAR‘Y may also involve secondary indirect mechanisms, as PPAR'y activation has been shown to decrease major histocompatibility complex (MHC) II receptor expression on atheroma-associated cells (3 72). MHC 11 receptors are responsible for antigen presentation to T cell receptors and as such are critical for T cell activation. IX. Rationale It is widely established that 2-AG and AEA modulate a variety of immune responses, including suppression of cytokine production, inhibition of immunoglobulin secretion, and suppression of CD8+ T cell migration. It has also been demonstrated that 2-AG and AEA are produced by a number of different immune cell types upon activation, including macrophages, dendritic cells, basophils, and microglial cells (108, 160, 172, 232, 234-236). Consequently, 2-AG and AEA may play an important role in immune regulation and the maintenance of immune homeostasis. Additionally, the growing interest in the therapeutic uses of AEA and 2-AG, particularly as anti-emetics, anti-spasmodics, and appetite stimulants, in immunocompromised patients, necessitates the characterization of their effects in the immune system. While previous studies from this laboratory have determined that suppression of IL-2 production by 2-AG occurs 40 through inhibition of NFAT, the upstream events leading to the impaired response have not yet been elucidated. The mechanism of the suppression of IL-2 by AEA is less well characterized than that of 2-AG. Because the role of the cannabinoid receptors had not yet been determined in the suppression of IL-2 by ABA and 2-AG, the initial studies focused upon CB1 and CB2. Subsequent studies examined the involvement of other targets, including COX-2 and PPAR'y. The overall goal of the project was to determine the mechanism of IL-2 suppression by 2-AG and AEA through the investigation of the following hypothesis: 2-AG and AEA suppress IL-2 secretion in activated T cells through PPAR‘Y independent of CB1 and CB2 (Figure 6). 41 $853.3 2: «e 55588.52 umuanaom .e 8.53% .8889:— «a _ sauces; . 2.355.. _ .522 ._.<.a.z 8de fig / k a see axe”. a. . ewe «we am a w a we age O O O O O O O O O O O O O O O O O O O O O O O O O O O O l< BKG PMA/IVH 0.01 0.1 1 10 20 I 10 . l Anandarmde (11M) Figure 7. Effect of AEA upon PMA " J -2“ " ‘ ‘ ’ IL-2 production in mil tine primary splenocytes. Splenocytes (1 x 106 cell/ml) were treated with 0.01-20 “M of ABA or VH (0.1% ethanol) for 30 min followed by activation of the cells with PMA (40 nM) and ionomycin (0.5 11M). Cells were harvested 24 h later and the suIDematants were analyzed for IL-2 protein by ELISA. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean :1: StaI‘Ldard error of triplicate cultures. *, p<0.05 compared to VH group. These data are representative of at least four separate experiments. 51 Figure 8. Effect of 2-AG and 2-AG ether upon PMA/ionomycin-stimulated IL-2 production in murine primary splenocytes Splenocytes (1 x 106 cell/ml) were treated with either A.) 2-AG (0.01-50 nM) or B.) 2-AG ether (0.1-50 nM) for 30 min followed by activation of the cells with PMA/ionomycin (40nM/0.5 11M). Cells were harvested 24 h later and the supematants were analyzed for IL-2 protein by ELISA. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean :t standard error of triplicate cultures. * p<0.05 compared to VH group. These data are representative of at least three separate experiments. 52 fated N60 IL-2 (U/ml) IL-2 (U/ml) Figure 8. 12507 1000- 750' 500- 250‘ N.D. — BKG P/IVHl0.01 0.1 l 2.5 5 10 15 20 50l 2-Arachidonyl glycerol (11M) 8001 400- 200' 0 N.D. * a: e BKG P/I VHIOJ l 2.5 5 10 15 20 50l 2-Arachidonyl glycerol ether (11M) e 53 II. Role of the cannabinoid receptors, CB1 and CB2, in the suppression of IL-2 secretion by AEA, 2-AG, and 2-AG ether With the discovery that the putative endogenous cannabinoids suppress IL-2, the subsequent experiments were designed to determine the mechanism for this effect with the initial studies focusing upon the role of the cannabinoid receptors. Although CB2 is the predominant cannabinoid receptor expressed in immune cells, low levels of CB1 transcripts have also been detected in many immune cells, including T cells (104). Consequently, the role of both CB1 and CB2 in the suppression of IL-2 secretion by AEA, 2-AG, and 2-AG ether was evaluated. Pretreatment of primary splenocytes with the CB1 and CB2 antagonists, SR141716A and SR144528, used in combination (0.05/0.05, 0.5/0.5, and 5/5 11M), did not attenuate AEA-mediated suppression of IL-2 secretion (Fig. 9a). Additionally, pretreatment with SR144528 alone did not attenuate AEA-mediated suppression of IL-2 secretion (Figure 9b). Similar to AEA, pretreatment with SR141716A and SR144528, in combination, did not block the suppression of IL-2 secretion by 2-AG and 2-AG ether, suggesting that CB1 and CB2 are not involved (Figure 10). At the highest concentration used (5/5 nM), treatment of SR141716A and SR144528 alone decreased IL-2 production, which diminishes the utility of the antagonists at this level. The role of the cannabinoid receptors in the modulation of IL-2 production was further examined through the use of the CBZ-specific agonist, JWH133. Treatment of primary splenocytes with JWH133 did not suppress IL-2 except at the highest concentration used (50 nM), which filrther suggests that the decrease in IL-2 secretion is independent of CB2 (Figure 11). Additionally, AEA, 2-AG, and 2-AG ether all suppress 54 Figure 9. Effect of cannabinoid receptor antagonists on AEA-mediated suppression of PMA/ionomycin-stimulated IL-2 production. Splenocytes (1 x 10‘5 cells/ml) were pretreated with either A.) both SR141716A and SR144528 or B.) SR144528 alone for 30 min followed by ABA (10 nM) treatment for 30 min. Cells were then stimulated with PMA (40 nM) and ionomycin (0.5 11M) for 24 h. The supematants were harvested and interleukin-2 production was measured by ELISA analysis. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean 1 standard error of triplicate cultures. A.) * p<0.05 compared to VH + AEA group. B.) None of the groups were significantly different from the 0 + AEA group. These data are representative of at least three separate experiments. 55 IL-2 (U/ml) IL-2 (U/ml) Figure 9. 8001 I Vehicle AEA (10 nM) 600-1 7 i 7" * 400- g % g r / 2 3 / 2 / / é / 4 % 4 a 2 a a / / a a t a 0 N.D. /; A 4 4. BKG PMA/ I VH 00510.05 0.5105 515] 1" SR141716A/SR144528 (11M) + 1500- | Vehicle AEA (10 1M) 10004 500- .7 i 2 / / 0 Ni». . . /. BKG PII 0 0.1 0.5 1 l l I SR144528 (11M) 56 Figure 10. Effect of cannabinoid receptor antagonists on suppression of IL-2 production by 2-AG and 2-AG ether. Splenocytes (1 x 106 cells/n11) were pretreated with both SR141716A and SR144528 or VH (0.1% DMSO) for 30 min followed by treatment with either A.) 2-AG (20 nM) or B.) 2-AG ether (20 111M) for 30 min. Cells were then stimulated with 40nM PMA/0.5 uM ionomycin for 24 h. The supematants were harvested and interleukin-2 production was measured by ELISA analysis. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean i standard error of triplicate cultures. * p<0.05 compared to VH + 2-AG ether group. None of the groups were significantly different from the VH + 2-AG group. These data are representative of at least three separate experiments. 57 A. 800- I Vehicle 2-AG (20 nM) 600 - IL—2 U/ml 400' 200~ 0_ . . '::'- . BKG P/I L0 VH 0.1/0.1 0.5/0.5 l/l 5/5I SR141716A/SR144528 (nM) 3- 600- . Vehicle 500‘ Z 2-AG ether (20 11M) 400— IL-2 (U/ml) 300- 200- 100- BKG P/I I 0 VII 0.1/0.1 0.5/0.5 1/1 5/5. L SR141716A/SR144528 (nM) Figure 10. 58 1000] 750- IL-2 (U/ml) 500- * 250' 0“N.D. BKG P/I VHL0.01 0.1 l 2.5 5 10 20 50J JWH133 (11M) Figure 11. Effect of the CB2-specific agonist, JWH133, upon PMA/ionomycin- stimulated IL-2 production in murine primary splenocytes. Splenocytes (1 x 106 cell/ml) were treated with JWH133 (0.01-50 11M) or VH (0.1% ethanol) for 30 min followed by activation of the cells with PMA (40 nM) and ionomycin (0.5 uM). Cells were harvested 24 h later and the supematants were analyzed for IL-2 protein by ELISA. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the meani standard error of triplicate cultures. * p<0.05 compared to VH group. These data are representative of at least two separate experiments. 59 PMA/Io-induced IL-2 production in the Jurkat T cell line, which lacks CB] and expresses a nonfunctional form of CB2 (Figures 12 and 13) (53). Interestingly, AEA is markedly more potent and efficacious in the human Jurkat T cells than in primary murine splenocytes. The increased potency of ABA in Jurkat cells may be due to species differences in sensitivity to AEA. Alternatively, the presence of B cells and macrophages in the mixed splenocyte preparation may diminish the activity of ABA in this model. Although there is a pronounced difference in potency of ABA in the Jurkat and splenocyte models, both 2-AG and 2-AG ether are only modestly more potent in Jurkat cells compared to splenocytes. While the studies with the CB1/CB2 antagonists and JWH133 coupled with the ability of the putative endogenous cannabinoids to suppress IL-2 in the Jurkat model suggested that the effects of AEA, 2-AG, and 2-AG ether upon IL-2 were independent of CB1 and CB2, the evidence was not entirely conclusive. Due to the relatively few number of laboratories investigating the immune effects of cannabinoids, there was a dearth of published studies with SR144528 to confirm its utility as a CB2 antagonist. Furthermore, the suppression of IL-2 by SR141716A and SR144528 treatment alone, confounded their ability to act as antagonists. Additionally, the presence of the aberrant CB2 transcripts in the Jurkat model was problematic because it was initially unclear whether the aberrant CB2 receptor was fimctional. Moreover, the lack of specific CB2 antibodies made it difficult to determine whether CB2 protein is expressed in Jurkat cells. Although previous studies from this laboratory demonstrated that, unlike normal CB2, the aberrant CB2 receptor is not capable of inhibiting cAMP production, it was unknown whether the aberrant receptor could couple to other second messengers and 60 100007 7500‘ IL-Z (pg/m1) 5000- 2500- 3|: * 0_ N.D. N.D. N.D. BKG P/I VHl0.1 1 2.5 5 10 15 20 SOJ Anandamide (11M) > Figure 12. Effect of AEA upon PMA/ionomycin-stimulated IL-2 production in human Jurkat T cells. Jurkat cells (5 x 105 cell/ml) were treated with 0.1-50 uM of AEA or VH (0.1% ethanol) for 30 min followed by activation of the cells with PMA (40 nM) and ionomycin (0.5 uM). Cells were harvested 24 h later andpthe supematants were analyzed for IL-2 protein by ELISA. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean :1: standard error of triplicate cultures. *, p<0.05 compared to VH group. These data are representative of at least two separate experiments. 61 Figure 13. Effect of 2-AG and 2-AG ether upon PMA/ionomycin-stimulated IL-2 production in Jurkat T cells. Jurkat T cells (5 x 105 cell/ml) were treated with 0.1-50 11M of either A.) 2-AG or B.) 2-AG ether for 30 min followed by activation of the cells with PMA/ionomycin (40nM/0.5uM). Cells were harvested 24 h later and the supematants were analyzed for IL-2 protein by ELISA. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean :t standard error of triplicate cultures. *, p<0.05 compared to VH group. These data are representative of at least two separate experiments. 62 5000‘ 4000- IL-2 (pg/ml) 3000‘ 2000* 1000- t 2 e . N.D. 50 BKG P/I lVH 0.1 1 2.5 5 10 15 20 50I .115 2-Arachidonyl glycerol (M O B 30001 /ofor r. re 2000‘ IL-2 (Pg/ml) 1000- a: t * 0_ N.D. * BKG PII VH 0.1 l 2.5 5 10 15 20 50 I l 2-Arachidonyl glycerol ether (11M) ’ Figure 13. 63 thereby fimction in other capacities. Consequently, the involvement of CB1 and CB2 in IL-2 suppression by the putative endocannabinoids was examined further through the use of CB l/CB2 double-knockout mice. AEA causes a suppression of IL-2 in splenocytes derived from CB1/CB2 null mice which is similar to that observed in splenocytes derived from wild-type controls, confirming that CB1 and CB2 are not involved (Figure 14). Likewise, both 2-AG and 2-AG ether also produce a comparable decrease in IL-2 in CB1/CB2 null splenocytes to that observed in wild-type controls (Figures 15 and 16). III. The role of the vanilloid receptor, VRl, in AEA-mediated IL-2 suppression Because AEA has also been shown to activate the vanilloid receptor, VRl , the role of VRl in AEA-mediated IL-2 suppression was investigated (214). The effect of VRl activation upon PMA/Io-induced IL-2 secretion was initially assessed with the VRl agonist, capsaicin. In contrast to the positive control, CP55940, increasing concentrations of capsaicin had little effect upon lL-2 secretion (Figure 17). Furthermore, pretreatment with the VRl antagonist, capsazepine, did not block AEA- mediated suppression of IL-2 secretion. Collectively, the aforementioned results suggest that VRl is not involved (Figure 18). IV. Effect of’ 2-AG and AEA upon calcium influx in resting splenocytes and activated splenocytes and thymocytes The activation of T cells requires both the induction of the MAP kinase pathway as well as a rise in intracellular calcium, which in turn activates the Ca+2-dependent phosphatase, calcineurin, and ultimately results in the dephosphorylation and 64 Figure 14. Effect of AEA upon PMA/ionomycin-stimulated IL-2 production in splenocytes derived from CB1/CB2 null and wild-type mice. Splenocytes (1 x 106 cells/ml) were treated with 0.1-50 1.1M of ABA for 30 min followed by activation of the cells with PMA/ionomycin (40nM/0.5uM). Cells were harvested 24 h later and the supematants were analyzed for IL-2 protein by ELISA. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean :1: standard error of triplicate cultures. * p<0.05 compared to VH group. These data are representative of at least two separate experiments. 65 Give—.8855, 5m 8 a S m 3 _ 2}; E 8:— AHZ-e remN * teem rem» reeefi 25-2; .3: 00h 1x18 anofme r ime 28590111 :25 8885.2 :5: «m0: m0 mmi ata are .2 2:5 .3 8 a S m 3 H 3:5 5 8E re remN .23 2:5: NLHH rams race“ 66 Figure 15. Effect of 2-AG upon PMA/ionomycin-stimulated IL-2 production in splenocytes derived from CB1/CB2 null and wild-type mice. Splenocytes (l x 106 cells/ml) were treated with 2-AG (0.1-50 11M) for 30 min followed by activation of the cells with PMA/ionomycin (40nM/0.511M). Cells were harvested 24 h later and the supematants were analyzed for IL-2 protein by ELISA. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean i standard error of triplicate cultures. * p<0.05 compared to VH group. These data are representative of at least two separate experiments. 67 1 G25 .9892» SeeEeoafira Jm em mm 3 m m.N ~ Sig—E UMm .Q.Z 893-23? lA—ZB Enoch—w 352553-" re * * tch or * reev Gaga urn: * -eoe ._. h as ==z 82:6 f the 1.10 {01 1t er .Q.Z .3 25$ aem en 3 3 m m.~ m Eng—E UV:— -e ream -2... a - o8 . cam .53 Nut: 68 Figure 16. Effect of 2-AG ether upon PMA/ionomycin-stimulated IL-2 production in splenocytes derived from CB1/CB2 null and wild-type mice. Splenocytes (1 x 106 cells/ml) were treated with 2-AG ether (0.1-50 11M) for 30 min followed by activation of the cells with PMA/ionomycin (40nM/0.511M). Cells were harvested 24 h later and the supematants were analde for IL-2 protein by ELISA. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean :1: standard error of triplicate cultures. *, p<0.05 compared to VH group. These data are representative of at least two separate experiments. 69 93 .850 .9892» Reactions—«TN .3 233 [A25 .550 3.89:» REE-Luanda" Em an mu 3 m m.~ fl :1? Ch Ova 5m an m“ 3 m md — fiiES— Ova .Qilc I: * * * 1:3 rem" * 18m :58 team case * mid NH: tomb lamb mgr—.53 1:2: :52 NM—U\~m—U r53 m .0... m h w... .M U m am... no. 311% mar? 70 300- 250— T 200- T IL-2 (U/ml) 150- 100- * 50- _ N.D. BKG P/I VHl0.01 0.1 1 10 20 50 I CP Capsaicin (11M) V Figure 17. Effect of capsaicin upon PM.A ’° _, -2“ " ‘ ‘ IL-2 production in murine primary splenocytes. Splenocytes (1 x 10‘5 cell/ml) were treated with capsaicin (0.01-50 nM), CP55940 (10 11M), which served as a positive control, or VH (0.1% ethanol) for 30 min followed by activation of the cells with PMA (40 nM) and ionomycin (0.5 11M). Cells were harvested 24 h later and the supematants were analyzed for IL-2 protein by ELISA. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean 3: standard error of triplicate cultures. *, p<0.05 compared to VH group. These data are representative of two separate experiments. 71 800- ' Vehicle AEA (10 11M) 600- IL-2 (Ulml) 400- 200- 0 N.D. f BKG WI 0.01 0.1 Capsazepine (11M) Figure 18. Effect of the vanilloid receptor antagonist, capsazepine, on AEA- mediated suppression of PMA/ionomycin-stimulated IL-2 production. Splenocytes (1 x 106 cells/ml) were pretreated with capsazepine (0.01-l 11M) for 15 min followed by AEA treatment for 15 min. Cells were then stimulated with PMA (40 nM) and ionomycin (0.5 nM) for 24 h. The supematants were harvested and IL-2 production was measured by ELISA analysis. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean :h standard error of triplicate cultures. *, p<0.05 compared to 0 + AEA group. These data are representative of two separate experiments. 72 translocation of NFAT to the nucleus. As a result, disruption of the normal Ca+2 influx associated with T cell activation would inhibit the binding of NFAT to the IL-2 promoter and suppress transcription. Conversely, the initiation of a calcium signal in the absence of simultaneous MAP kinase activation generally results in T cell anergy (375). Furthermore, published studies suggest that both ABA and 2-AG modulate calcium channels, including inhibition of voltage-gated calcium channels in a variety of cell types and induction of transient calcium influx in HL-60 cells (17, 20, 27). Collectively, these results suggest that 2-AG and AEA could mediate their effects upon IL-2 through modulation of the calcium signal, which thereby could either induce T cell anergy or inhibit activation of NFAT. Unlike the plant-derived cannabinoids, however, neither 2-AG nor AEA induce calcium influx in primary murine splenocytes (Figure 19). In addition, calcium influx by ionomycin or concanavalin A is not inhibited by 2-AG pretreatment, suggesting that modulation of intracellular calcium is not the major mechanism of 2-AG-mediated IL-2 suppression (Figure 20). V. The effect of arachidonic acid upon IL-2 secretion and the roles of FAAH, MAG lipase, and the AMT in the suppression of IL-2 by AEA and 2-AG Because it was unclear whether the inhibitory effect of ABA upon IL-2 secretion was due to the parent molecule or a hydrolysis product of AEA, the effect of arachidonic acid upon IL-2 secretion was also evaluated. Treatment of primary splenocytes with various concentrations of arachidonic acid (0.01 — 20 11M) caused a concentration- dependent decrease in IL-2 secretion (Figure 21). The magnitude of the decrease in IL-2 secretion produced by arachidonic acid was very similar to that observed with AEA as 73 Figure 19. Effect of AEA and 2-AG upon intracellular calcium. Splenocytes (5 x 106 c/ml) were loaded with fura-2 AM dye for 30 minutes at room temperature in the dark. Cells were harvested and washed four times in Ca+2 KREB buffer to remove excess fura- 2 AM dye from the buffer. Three ml of cells were added to a quartz cuvette and calcium concentration determined. At 300 see, either A.) ABA (20 11M) or B.) 2-AG (20 11M) was added. In a second run, cannabidiol (10 nM) was added at 300 sec, which served as a positive control. Maximum and minimum values were obtained with 0.1% Triton-X and 0.5 mM EGTA, respectively. Calcium concentration was calculated from the change in ratio of bound to free calcium. These data are representative of at least four separate experiments. 74 L A. 400- 300“ CBD (10 nM) retelli / (nM) we 0 500 1000 1500 Time (sec) B. 300- 200_ CBD (10 nM) [cw], , 7 i “7“": “* “m :— (nM) 100 1 n4 _ ---_n -_ .______.A,__v.___..._- . urn.“ 0' 2-AG (20 nM) -100- . 0 200 400 600 800 1000 1200 1400 1600 Time (sec) Figure 19. 75 Figure 20. Effect of 2-AG upon calcium influx by ionomycin and concanavalin A. Splenocytes (5 x 106 c/ml) were loaded with fura-2 AM dye for 30 minutes at room temperature in the dark. Cells were harvested and washed four times in Ca+2 KREB buffer to remove excess fura-Z AM dye from the buffer. Three ml of cells were added to a quartz cuvette and were either pretreated with 2-AG (20 11M) or left untreated. At 300 see, either A.) ionomycin (0.5 nM) or B.) concanavalin A (10 rig/ml) was added. Maximum and minimum values were obtained with 0.1% Triton-X and 0.5 mM EGTA, respectively. Calcium concentration was calculated from the change in ratio of bound to free calcium. These data are representative of at least two separate experiments. 76 valill A. room REB 3 added to 1. A1300 4 E0111. ibound ‘0 /10 200 / 100- " 0- 10+2-AG [Ca+2]l _100.. (nM) -2003 -300- 0 200 400 600 800 1000 1200 1400 1600 Time (sec) B. 300- ConA 200' loan]. M “m“ " i” ‘ (nM) 100: ConA+2-AG 0.. -1m- *1 ‘1 If T *Ii I I I 02004006008001000120014001600 Time (sec) Figure 20. 77 1000] 750 ‘ IL-2 CU/ml) 500 - 250 - * 0- BKG PMA/ VH 0.01 0.1 1 10 20 lo I J l Arachidonic acid (11M) V Figure 21. Effect of arachidonic acid upon PMA " J -2. " ' ‘ ‘ IL-2 production in murine primary splenocytes. Splenocytes (1 x 106 cell/m1) were treated with 0.01-20 uM of arachidonic acid or VH (0.1% ethanol) for 30 min followed by activation of the cells with PMA (40 nM) and ionomycin (0.5 11M). Cells were harvested 24 h later and the supematants were analyzed for IL-2 protein by ELISA. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean i standard error of triplicate cultures. * p<0.05 compared to VH group. These data are representative of at least four separate experiments. 78 evidenced by the calculated ICso values: 11.4 uM and 10.3 uM for ABA and arachidonic acid, respectively. Because arachidonic acid produced a concentration-dependent suppression of IL-2 secretion similar to the one observed with AEA, it seemed reasonable that suppression of IL-2 by AEA was mediated by arachidonic acid, which is likely formed from the hydrolysis of ABA, rather than by the parent molecule of ABA itself. It has been postulated that the AMT may be coupled to F AAH, the chief enzyme responsible for AEA hydrolysis and that these two proteins represent a major mechanism for the uptake and catabolism of ABA (3 76). As such, the potential roles of the AMT and FAAH in AEA-mediated IL-2 suppression were studied. Pretreatment with the AMT inhibitors, AM404 and UCM707 (ICso values = luM and 0.8 11M, respectively), at various concentrations, did not attenuate AEA-mediated suppression of IL-2 secretion, suggesting that the AMT is not involved (Figure 22). At the highest concentration used (10 11M), AM404 alone caused substantial suppression of IL-2. In the presence of ABA, 10 11M AM404 modestly potentiated the decrease in IL-2 secretion by AEA, but this was not statistically significant. Conversely, UCM707 had little effect upon IL-2 secretion in the absence of ABA. With the elimination of the AMT as a potential target of AEA, the role of F AAH was subsequently examined using the potent, irreversible FAAH inhibitor, MAFP (1C50 = 2.5 nM). Pretreatment with a broad range of concentrations of MAFP (0.001 —— 1 11M) did not attenuate the suppression of IL-2 secretion by AEA (Figure 23a). It has been demonstrated that in addition to the hydrolysis of ABA, FAAH also catalyzes the reverse reaction, which results in the synthesis of AEA from arachidonic acid and ethanolamine. Consequently, it is possible 79 Figure 22. Effect of AMT inhibitors on suppression of IL-2 secretion by AEA. Splenocytes (1 x 10‘5 cells/ml) were pretreated with either A.) AM404 (0.1-10 11M) or B.) UCM707 (1-20 nM) for 30 min followed by treatment with AEA for 30 min. Cells were then stimulated with 40nM PMA/0.5 11M ionomycin for 24 h. The supematants were harvested and interleukin-2 production was measured by ELISA analysis. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean :1: standard error of triplicate cultures. Neither the groups that were treated with AM404 and AEA nor those that were treated with UCM707 and AEA were significantly different from the control groups, VH + AEA. The data are representative of at least three separate experiments. 80 18. 1000! IIr2 (U/ml) IIp2 (U/ml) Figure 22. 750- 250- II‘Vehhfle AEA (10 11M) "I ‘ 1 | k\\\\\\\‘ SSSBBQQRSSSEE 800- 600- 400- 200- :.__/ ”A“... . g n. .u-"‘.' ;_.._-,_..._-‘,_..a.'. .. -I.'.iiv‘_.-.- __.. ' IR" '4 "" “.339: 'I 1.72.10.” " hI“"‘ 1:17.. ‘I'. L... "4 v u. 5‘. -".' ‘ 'o .' '. " ". ‘l- “ ‘ -l' - - '. ' ..- '-".-"... '.. -". .'. ' .‘ ". te'\_.;|'_' ’. '1" . -. {2.9. If. .'.:v .3 . 3“. - ..OJ. ' J. _ . . J. , ,-.-‘. . r “(.'.-9‘ ‘D i‘ I; v_i.§.._-_......-,-. -'_'.'..ev"'l. .__..,.o,_ _' ... -_-,,. alumina-e": u.) ._ - I I'*,_ v'- "_'-.-‘ 9..-";‘.'-.e- .'. .- '.. \ ..- l'.‘. ...-U "' t\\\\\\\‘- E 3’ E I Vehicle AEA (20 11M) 2 0. IBEX} [VI ‘" SEDB\‘§ l0 VH 1 10 20I UCM707 (11M) O 81 Figure 23. Effect of the FAAH inhibitor, MAFP, on suppression of IL-2 secretion by AEA and arachidonic acid. Splenocytes (l x 106 cells/ml) were pretreated with MAF P at various concentrations for 30 min followed by treatment with either A.) AEA or B.) arachidonic acid for 30 min. Cells were then stimulated with 40nM PMA/0.5 11M ionomycin for 24 h. The supematants were harvested and interleukin-2 production was measured by ELISA analysis. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean i standard error of triplicate cultures. *, p<0.05 compared to VH + AEA group. None of the groups treated with arachidonic acid were significantly different from the control group, O + arachidonic acid. These data are representative of at least three separate experiments. 82 lps as riplicalf With njc 36111. A. 12501 1000- 750- IL-2 (mm!) 250 - _ N.D. BKG P/I B. 1250- 1000‘ 750‘ IL-2 (U/ml) 500- 250- PMA/Io alone I Vehicle AEA (11M) :. ... f1 3k * r I I * 4 6 fl 7. 7.. 0 '1 VH 0.001 0.01 0.1 l l MAFP(1|M) PMA/Io alone I Vehicle Arachidonic Acid (15 11M) Figure 23. 83 that the observed effects of arachidonic acid upon IL-2 secretion are actually mediated by newly-synthesized AEA, produced by FAAH. As such, the effect of MAFP upon the suppression of IL-2 by arachidonic acid was studied. Pretreatment with a broad range of concentrations of MAP P (0.01 -— 10 nM) did not attenuate the decrease in IL-2 secretion by arachidonic acid (Figure 23b). At the highest concentration used (10 nM), MAFP alone caused significant inhibition of IL-2 secretion. Because MAFP is also a PLAz inhibitor and previous studies from this laboratory have shown that PLAz inhibitors cause a decrease in IL-2 production, the inhibitory effect of MAFP upon IL-2 secretion was not unexpected (331). In addition to studies suggesting that the AMT catalyzes the transport of ABA across cell membranes, there is also evidence to suggest that the AMT can also participate in 2-AG transport. Likewise, FAAH has also been found to hydrolyze 2-AG in addition to AEA. In addition to FAAH, 2-AG is also hydrolyzed by MAG lipase, which is also inhibited by MAP P. As such, the roles of the putative AMT, FAAH, and MAG lipase in the inhibitory effect of 2-AG upon IL-2 were investigated. Neither pretreatment with AM404 nor MAF P affected suppression cf IL-2 by 2-AG, thus ruling out the involvement of either the AMT, FAAH, or MAG lipase in IL-2 suppression by 2-AG (Figure 24). VI. The role of COX in the suppression of IL-2 by AEA and 2-AG Because arachidonic acid suppresses IL-2 with a potency comparable to that of ABA and 2-AG, the role of the COX enzymes was investigated. Initial evaluation of concentration responses of flurbiprofen determined 50 uM to be an optimum 84 Figure 24. Effect of the AMT inhibitor, AM404, and the FAAH inhibitor, MAFP, on suppression of IL-2 secretion by 2-AG. Splenocytes (l x 106 cells/ml) were pretreated with either A.) AM404 (0.1-10 nM) or B.) MAFP (0.1-10 nM) for 30 min followed by treatment with 2-AG (10 nM) for 30 min. Cells were then stimulated with 40nM PMA/0.5 uM ionomycin for 24 h. The supematants were harvested and IL-2 production was measured by ELISA analysis. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean :t standard error of triplicate cultures. *, p<0.05 compared to VH + 2-AG group. These data are representative of two separate experiments. 85 A- 15001 I Vehicle 2-AG (10 nM) 1000'“ IL-2 (U/ml) 500‘ I ‘1 5 :r‘ I * 3‘ ,_ 2 2 2 2 2 2 BKGP L0 0.1 1 2.5 5 10l AM404 (nM) A B. 1250- I Vehicle 1000‘ 2-AG (10 nM) 750‘ IL-2 (U/ml) 500‘ I pl. 250- a * ... ... _ 7 3 Z 3 Z Z Z 0_ e 2 2 2 2 2 2 BKG P/I lVH 0.1 l 2.5 5 10 I MAFP(uM) > Figure 24. 86 concentration for inhibition of COX-1 and COX-2 in splenocytes (data not shown). The higher concentration of 100 uM flurbiprofen caused marked suppression of IL-2 by itself, which is not surprising since flurbiprofen (in the range of 100 — 1000 uM) is known to inhibit NFKB, an important transcription factor for IL-2 transcription (377). Although 50 uM flurbiprofen also caused decreased IL-2 secretion in the absence of ABA or arachidonic acid, the decrease was far less robust than that which was produced by 100 uM flurbiprofen and nonetheless still caused an almost complete reversal of ABA- mediated IL-2 inhibition (Figure 25a). Likewise, flurbiprofen pretreatment also significantly attenuated the suppression of IL-2 mediated by arachidonic acid (Figure 25b). Similar to flurbiprofen, pretreatment of primary splenocytes with piroxicam partially attenuated the inhibitory activity of both AEA as well as arachidonic acid upon IL-2 secretion (Figure 26). Moreover, the effect of piroxicam upon AEA-mediated suppression of IL-2 is concentration-dependent (Figure 28a). In order to determine which COX subtype is involved, the ability of NS398, a COX-2 specific inhibitor, to attenuate suppression of IL-2 by ABA and arachidonic acid was evaluated. Similar to flurbiprofen and piroxicam, N83 98 also attenuated the decrease in of IL-2 secretion by ABA and arachidonic acid (Figure 27). In addition, the attenuation of AEA-mediated decrease in IL-2 secretion by N83 98 is also concentration- dependent (Figure 28b). Conversely, the COX-1 specific inhibitors, SC560 and FR122047 (ICso values = 9nM and 28 nM, respectively), had no effect upon the suppression of IL-2 by AEA (Figure 29). Likewise, pretreatment with SC560 did not attenuate the decrease in IL-2 by arachidonic acid (Figure 30). Collectively, this suggests that COX-2 plays a role in the inhibitory effects of ABA and arachidonic acid upon IL-2. 87 Figure 25. Effect of the nonselective COX inhibitor, flurbiprofen, on suppression of IL-2 secretion by AEA and arachidonic acid. Splenocytes (1 x 10‘5 cells/ml) were pretreated with 50 uM flurbiprofen (F BN) or VH (0.05% ethanol) for 30 min followed by treatment with either A.) AEA (1-20 uM) or B.) arachidonic acid (1-20 nM) for 30 min. Cells were then stimulated with 40nM PMA/0.5 uM ionomycin for 24 h. The supematants were harvested and IL-2 production was measured by ELI SA analysis. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean i standard error of triplicate cultures. a denotes p<0.05 compared to VH + VH group. b denotes p<0.05 compared to FBN + VH group. c denotes p<0.05 compared to the matched VH group. These data are representative of at least three separate experiments. 88 » Arachidonic acid (nM) 89 k m k m m m m m e m e m V V In I r A I I c quI/I/l/ld 0 + I I a 2 c 07/////////. 0 M a 1 I..\ a fir//////////1.m d nr/l/ll/lllll. m n A L H.7///////////l 0 radwwflfi .2 a . a... 2km. .. ....e... P I o D. G . q . . . . N. m _ . . . a m m m m m m o m m m m o 7 5 8 4 2 l l D ) 2 m 2 .m u l - I m w m w A. B. Figure 25. Figure 26. Effect of the COX-1 selective inhibitor, piroxicam, upon suppression of IL-2 secretion by AEA and arachidonic acid. Splenocytes (l x 10‘5 cells/ml) were pretreated with piroxicam (20 or 50 nM) or VH (O. 1% ethanol) for 30 min followed by treatment with either A.) anandamide (AEA) or B.) arachidonic acid (AA) for 30 min. Cells were then stimulated with PMA (40 nM) and ionomycin (0.5 nM) for 24 h. The supematants were harvested and IL-2 production was measured by ELISA analysis. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean 3: standard error of triplicate cultures. a denotes p<0.05 compared to VH + VH group. b denotes p<0.05 compared to piroxicam + VH group. c denotes p<0.05 compared to the matched VH group. These data are representative of at least three separate experiments. 90 I Vehicle Piroxicam (5011M) C .0 1.7//////////6 m . c b r/////////1 M J a V///////////////A 5 rl////////////Jr/////A 1 u7///////////////Z m 8007 600- IL-2 (U/ml) 400- 200 - mu ?////////////////1 0 Anandamide (nM) (20 M) I Vehicle Piroxicam c b nr/lf/KI m. c b 7.. V////////1 m wrur/V/IV/lf/f/l/V/ld 5 ur//////uf////Jr///////lr///a 1 Arachidonic acid (M) w?////////////////.. m V/////////////////. 0 800 7 600 ‘ (U/ml) 400- Figure 26. 91 Figure 27. Effect of the COX-2 specific inhibitor, NS398, on the suppression of IL-2 secretion by AEA and arachidonic acid. Splenocytes (1 x 106 cells/ml) were pretreated with NS398 (10 nM) or VH (0.02% ethanol) for 30 min followed by treatment with either A.) anandamide or B.) arachidonic acid for 30 min. Cells were then stimulated with PMA (40 nM) and ionomycin (0.5 uM) for 24 h. The supematants were harvested and IL-2 production was measured by ELISA analysis. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean at standard error of triplicate cultures. a denotes p<0.05 compared to VH + VH group. b denotes p<0.05 compared to NS398 + VH group. c denotes p<0.05 compared to the matched VH group. These data are representative of at least three separate experiments. 92 I Vehicle Nss9s (10 M) c V. ?/////1 ml ..an/l/l/d m ..W/I/l/l/l/l. 5 n?/////////////. 1 7/////////A m _r///////////A 0 l BKG P/I 1500 - 1000- 500- 0_ k 0 IL-2 (U/ml) Anandamide (M) + I Vehicle D C b 7//////’1 m a NS398 (10 nM) c b V//////////.. 0 .. l a whOI///////////1 5 ?////////////////A l nr/////////////. W .flV//Jr//////Jr////l 0 Arachidonic acid (nM) I‘LD. 1000- 750- 250- IL-2 (U/ml) 500- Figure 27. 93 Figure 28. Piroxicam and N8398 attenuate the suppression of IL-2 secretion by AEA in a concentration-dependent manner. Splenocytes (l x 106 cells/ml) were pretreated with either A.) piroxicam or B.) NS398 for 30 min followed by treatment with either anandamide (AEA) or VH (0.04% ethanol) for 30 min. Cells were then stimulated with PMA (40 nM) and ionomycin (0.5 nM) for 24 h. The supematants were harvested and IL-2 production was measured by ELISA analysis. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean :9: standard error of triplicate cultures. * denotes p<0.05 compared to VH + anandamide group. These data are representative of least three separate experiments. 94 A. 800-} I Vehicle . Anandamide 600- (20 nM) IL-2 (U/ml) 400- 3|: 3|: 200' I" 3' a i 2 fl 5 5 .D. 2 2 2 2 2 BKGP/I lo VH 5 10 20 sol [ Piroxicam (pM) > B' 1250] I Vehicle . Anandamide 1000'- (20 FM) 7501 IL-2 as (U/ml) 500- , 2.! , 7 250- , . 7' 7 5 i 2“ / / / 2 2 2 2 2 0_N.I) 4 4 4. 5. 5. BKG P/I l0 VH 1 5 10 2L NS398 (uM) a Figure 28. 95 Figure 29. Effect of the COX-1 specific inhibitors, SC560 and FR122047, on the suppression of IL-2 secretion by AEA. Splenocytes (1 x 106 cells/ml) were pretreated with either A.) SC560 (1 nM) or B.) FR122047 (0.01-5 nM) for 30 min followed by treatment with AEA or VH (0.04% ethanol) for 30 min. Cells were then stimulated with PMA (40 nM) and ionomycin (0.5 nM) for 24 h. The supematants were harvested and IL-2 production was measured by ELISA analysis. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean :h standard error of triplicate cultures. a, p<0.05 compared to VH + VH group. b, p<0.05 compared to VH + SC560 group. *, p<0.05 compared to VH + AEA group. These data are representative of at least two separate experiments. 96 .fi. I Vehicle SC560 (1 uM) IL-2 (U/ml) B- 1000' I Vehicle AEA (20 nM) 750‘ IL-2 (U/ml) 500' 250' _ 2 . BKG P/I l0 VH 0.01 0.1 1 5I FR122047 (pM) Figure 29. 97 500- I Vehicle SC560 (1 nM) 400— w a, ‘/ 3.. 2 2 IL-2 é % (U/ml) 2 2 _ / ¢ 200 6 ¢ 8 / b 100- g g g / 2 / 2 . 2 2 2 . BKG P/I VB 1 5 10 20I Arachidonic acid (nM) Figure 30. Effect of the COX-1 specific inhibitor, SC560, on arachidonic acid- mediated inhibition of PMA/ionomycin-stimulated IL-2 production. Splenocytes (1 x 10‘5 cells/ml) were pretreated with SC560 (1 nM) for 30 min followed by arachidonic acid or VH (0.04% ethanol) treatment for 30 min. Cells were then stimulated with PMA (40 nM) and ionomycin (0.5 nM) for 24 h. The supematants were harvested and IL-2 production was measured by ELISA analysis. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean i standard error of triplicate cultures. a, p<0.05 compared to VH + VH group. b, compared to VH + SC560 group. These data are representative of at least two separate experiments. 98 Similar to the attenuation of AEA-mediated IL-2 suppression in primary splenocytes, flurbiprofen pretreatment also blocked inhibition of IL-2 secretion by 2-AG in Jurkat T cells (Figure 31). In contrast to AEA and arachidonic acid, however, piroxicam only modestly attenuated 2-AG-mediated suppression of IL-2 In Jurkat cells (Figure 32). This is likely because piroxicam has been shown to be highly selective for human COX-1 over human COX-2 (378-380). Furthermore, pretreatment with the COX-1 specific inhibitors, SC560 and FR122047, did not affect inhibition of IL-2 secretion by 2-AG in Jurkat cells (Figure 33). Conversely, the COX-2 specific inhibitor, NS398, caused a concentration-dependent reversal of 2-AG-mediated suppression of IL-2, suggesting the involvement of COX-2 in these effects (Figure 34). VII. Effect of prostanoids and PPAR? agonists upon IL-2 production Although the experiments with NS398 strongly suggest that both 2-AG and AEA are metabolized by COX-2 and that the metabolites are responsible for suppression of lL-2, the identity of the metabolites is unknown. The subsequent studies were designed to determine which metabolites inhibit IL-2 secretion. Initially the effect of AH6809, an antagonist of the EPl, EP2, and DP prostanoid receptors, was investigated. Although AH6809 did not antagonize the effects of AEA, it caused a significant degree of IL-2 suppression by itself (Figure 35a). As such, it was not possible to rule out the EPl, EP2 or DP receptors as targets of the metabolites of ABA. The EP receptors were of particular interest because a number of published studies demonstrated that PGEz causes inhibition of IL-2 secretion (339). Additionally, it has been demonstrated that PGEz- 99 250001 I Vehicle FBN(50uM) 20000- \‘ a 3 0 # # \ S § ‘8 ‘5 cs (pg/nu) \ \ \ \ *\ \ \ \ \ \ § *\ \ \ \ § \ ,_ N. s s s r s BKG 0 VB 1 s 10 20 _ — PMA/Io + 2-Arachidonyl glycerol (nM) Figure 31. Effect of the nonselective COX inhibitor, flurbiprofen, on 2-AG- mediated suppression of PMA/ionomycin-stimulated IL-2 production. Jurkat T cells (5 x 105 cells/ml) were pretreated with 50 uM flurbiprofen (FBN) for 30 min followed by 2-AG (1-20 nM) or VH (0.04% ethanol) treatment for 30 min. Cells were then stimulated with PMA (40 nM) and ionomycin (0.5 nM) for 24 h. The supematants were harvested and IL-2 production was measured by ELISA analysis. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean :1: standard error of triplicate cultures. *, p<0.05 compared to VH + VH group. #, p<0.05 compared to the matched vehicle group. These data are representative of three separate experiments,. 100 30001 b § c b I Vehicle ii S :5 ° PXM(50lIM) s \ ‘s 2000- \ \ \ ‘\ $ $ . S: lL-2 § § 8 S (pg/ml) § 3 s a\ S r s s . 1000- § S S S c b \ \ \ \ .‘. c S s s S . e. \ \ \ \ a\ \ \ \ \ \ \ \ § § § \ s 0 NI.D. § .\ § § § & BKG PII Io VII 1 ~ 5 10 20 2—Arachidonyl glycerol (nM) Figure 32. Effect of the COX-1 selective inhibitor, piroxicam, on 2-AG-mediated suppression of PMA/ionomycin-stimulated IL-2 production. Jurkat T cells (5 x 105 cells/ml) were pretreated with 50 uM piroxicam (PXM) for 30 min followed by 2-AG (1- 20 nM) or VH (0.04% ethanol) treatment for 30 min. Cells were then stimulated with PMA (40 nM) and ionomycin (0.5 nM) for 24 h. The supematants were harvested and IL-2 production was measured by ELISA analysis. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean a: standard error of triplicate cultures. a, p<0.05 compared to VH + VH group. b, p<0.05 compared to VH + piroxicam group. c, p<0.05 compared to the matched vehicle group. These data are representative of four separate experiments 10] Figure 33. Effect of the COX-1 specific inhibitors, SC560 and FR122047, on 2-AG- mediated suppression of PMA/ionomycin-stimulated IL-2 production. Jurkat T cells (5 x 105 cells/ml) were preheated with either A.) scsso (0.01-0.5 nM) or B.) FR122047 (0.01-5 nM) for 30 min followed by 2-AG (20 nM) or VH (0.04% ethanol) treatment for 30 min. Cells were then stimulated with PMA (40 nM) and ionomycin (0.5 nM) for 24 h. The supematants were harvested and IL-2 production was measured by ELISA analysis. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean i: standard error of triplicate cultures. A.) None of the groups treated with both 2-AG and SC560 were significantly different from the control group, VH + 2-AG. B.) *, compared to VH + 2-AG group. These data are representative of at least two separate experiments. 102 A- 5000] I Vehicle 4000“ 2-AG (20 nM) 3000- IL-2 (pg/ml) 2000- 1000- °'K'e' a B . . . . P Q) VH 001 005 or 05I scsao (11M) B 3000- . Vehicle 2500- 2-AG (20 M) 2000- IL-2 _ (pg/ml) 1500. 1000- 500- s 0_ N.D. 5 § BKG P/I I0 VH 0.01 0.1 1 5I FR122047(uM) Figure 33. 103 I Vehicle 2-AG (20 11M) 15000‘ IL-2 * 7" (pg/ml) 10000- ' ' z 22 2 '2 2 2 5000- . % f % 5' 2 2 ¢ 7 2 / / 2 2 2 2 f / 041.1). 4 4 4 Q Q BKG Pl! 1 0 VII 2.5 12.5 25 I L NS398 (11M) Figure 34. Effect of the COX-2 specific inhibitor, NS398, on 2-AG—mediated inhibition of PMA/ionomycin-stimulated IL-2 production. Jurkat T cells (5 x 105 cells/ml) were pretreated with NS398 (2.5-25 11M) for 30 min followed by 2-AG (20 nM) or VH (0.04% ethanol) treatment for 30 min. Cells were then stimulated with PMA (40 nM) and ionomycin (0.5 nM) for 24 h. The supematants were harvested and IL-2 production was measured by ELISA analysis. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean :t standard error of triplicate cultures. * , p<0.05 compared with VH + 2-AG group. These data are representative of at least two separate experiments. 104 Figure 35. Effect of the prostaglandin inhibitor, AH6809, on the suppression of IL- 2 secretion by AEA and the effect of PGEz-ethanolamine on IL-2 production. Splenocytes (1 x 106 cells/ml) were either A.) pretreated with AH6809 (1-20 nM) for 30 min followed by treatment with ABA (20 nM) or VH (0.04% ethanol) for 30 min or B.) treated with prostaglandin Ez-ethanolamine (0.1-50 nM) for 30 min. Cells were then stimulated with PMA (40 nM) and ionomycin (0.5 uM) for 24 h. The supematants were harvested and IL-2 production was measured by ELISA analysis. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean d: standard error of triplicate cultures. A.) None of the AEA-treated groups were significantly different fiom the VH + AEA group. #, p<0.05 compared to the VH + VH group. B.) *, p<0.05 as compared to the VH group. These data are representative of two separate experiments. 105 A. 800 I Vehicle AEA (2011M) 600- IL-2 (U/ml) 400- 200- 2 . . 2 4 BKG P/I l0 VB 1 5 10 20I AH6809 (11M) _ B 400— __ 300— * * * >1: * IL-2 (U/ml) 200- ... 100- _N.D. BKG P/I VH 10.1 1 2.5 5 10 15 20 501 Prostaglandin Ez-ethanolamine Figure 35. 106 ethanolamine is a COX-2 metabolite of AEA and that PGEz-ethanolamine binds to and activates all four EP receptors (176, 178). Moreover, all four EP receptors are expressed in T cells. Consequently, the effect of PGEz-ethanolamine upon IL-2 was assessed. IL-2 production was only modestly affected by increasing concentrations of PGE2- ethanolamine, however, suggesting that the EP receptors are not involved (Figure 35b). In addition to PGEz, other prostaglandins have been reported to suppress IL-2, including lSd-PGJz and PGIz (339, 340). Treatment of primary splenocytes with PG]; or lSd—PGJz produced a robust concentration-dependent decrease in IL-2 secretion (Figure 36). Because both PGJ; and 15d-PGJ2 have been found to activate PPAR'y, the effect of other known PPARy agonists upon IL-2 production was investigated. The PPAR‘y agonists, ETYA and ciglitazone, both suppressed IL-2 in a concentration-dependent manner, which supports the published findings of other laboratories that activated PPAR'y inhibits IL-2 production (Figure 37). The inhibition of IL-2 secretion by ciglitazone was more modest than that of the other PPAR’y agonists, which is consistent with observations of other laboratories (352). Although it is unclear why ciglitazone is less potent and efficacious than the other PPAR'Y agonists in the suppression of IL-2, it may be related to differences in chemical structure that result in subtle variations in the conformation of PPARy, which are not conducive to transrepression. VIII. The activation of PPAR’y by 2-AG and 2-AG ether The ability of PPAR'Y agonists to inhibit IL-2 secretion coupled with the similarity in structure of 2-AG and AEA to the putative endogenous ligands of PPARy, suggested that PPAR‘Y may play a role in the suppression of IL-2 by 2-AG and/or AEA. Evidence 107 Figure 36. Effect of PGJz and lS-deoxy PGJ; on IL-2 secretion. Splenocytes (1 x 106 cells/ml) were treated with either A.) prostaglandin 12 or B.) lS-deoxy prostaglandin 12 for 30 min. Cells were then stimulated with PMA (40 nM) and ionomycin (0.5 nM) for 24 h. The supematants were harvested and IL-2 production was measured by ELISA analysis. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean :1: standard error of triplicate cultures. *, p<0.05 as compared to the VH group. These data are representative of two separate experiments. 108 tes(lx :tion was aups as ‘m'plicatc e of tWO A. 1000' 750‘ IL-2 (U/ml) 500- 250- 0-N‘D' * * T T BKG P/I VHL0.1 l 2.5 5 10 15 20l Prostaglandin J2 4 B- 1000- 750- IL-2 (U/ml) 500‘ * 250‘ 2|: 0 NiD' * * BKG P/I VH Ifl-l 1 2.5 5 10 15. lS-deoxy PGJ2 (nM) > Figure 36. 109 Figure 37. Effect of ciglitazone and ETYA on IL-2 secretion. Splenocytes (1 x 106 cells/ml) were treated with either A.) ciglitazone (0.1-50 11M) or B.) ETYA (0.1-50 nM) for 30 min. Cells were then stimulated with PMA (40 nM) and ionomycin (0.5 11M) for 24 h. The supematants were harvested and IL-2 production was measured by ELISA analysis. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean i: standard error of triplicate cultures. ’, p<0.05 as compared to the VH group. These data are representative of at least three separate experiments. 110 6 es(1x10 (0.1-501M] ).5 W) for ,' ELISA 3‘ trypan bite _ p<0.0515 paraIC A. 800— 600‘ IL-2 * ml!!!” 400‘ >1: :1: 200- _ N.D. BKG P/I VH £1 1 2.5 5 10 15 50] Ciglitazone (M) B. 1500- 1000— IL-2 (U/ml) 500- * * _ N.D. BKG P/IVH&.1 1 2.5 5 10 15 20 50I ETYA (11M) Figure 37. 111 that 2-AG or AEA treatment caused PPARy activation was lacking, however. In an effort to simplify the preliminary studies, this line of investigation initially focused solely upon the ability of 2-AG and its non-hydrolyzable analogue, 2-AG ether, to activate PPAR'y. To characterize whether 2-AG and/or 2-AG ether treatment induced PPARY activation, 3T3-Ll adipogenesis, an assay widely used to identify PPAR'y agonists, was employed. Differentiation of 3T3-L1 fibroblasts into adipocytes was assessed morphologically by their round shape and by the presence of prominent lipid vacuoles, which are readily stained by Oil Red 0. Treatment with 2-AG as well as 2-AG ether induced adipogenesis as quantified by enumeration of the total number of differentiated cells per well (Figure 40). 2-AG ether induced more adipogenesis than 2-AG, but morphologically the changes induced by both were indistinguishable from those induced by ciglitazone (Figures 38 & 39). It is notable that because 3T3-Ll fibroblast adipogenesis requires 7 days coupled with the fact that 3T3-L1 cells efficiently absorb fatty acids and store them in the form of triglycerides, higher concentrations of 2-AG and 2-AG ether were required than in other short term assays using other models (i.e. IL-2 secretion in primary splenocytes). Because mRN A expression is a more precisely quantified endpoint than adipogenesis, the effect of 2-AG and 2-AG ether upon mRNA levels of aP2, a gene regulated by PPARy, was measured. The ability of PPARy to regulate aP2 was confirmed by the induction of aP2 by ciglitazone, a PPAR'y agonist, as well as by the suppression of the induction by GW9662, a PPAR'y antagonist (Figure 41). Concordant with PPARy activation, treatment of 3T3-L1 cells with either 2-AG or 2-AG ether produced a marked increase in aP2 mRN A levels. Moreover, 2-AG ether appeared to 112 Figure 38. Differentiation of 3T3-Ll cells by 2-AG. 3T3-L1 cells were grown to confluence and either exposed to B.) vehicle or C.) 2-AG (SOuM) or A.) not exposed. Cells were cultured 7 days after exposure with fluid renewal every 2 days. Cells were then fixed in 10% formalin and stained with Oil Red 0 and hematoxylin. Images in this dissertation are presented in color. 113 «w .m mm. .8 23E 114 Figure 39. Difl'erentiation of 3T3-L1 cells by 2-AG ether, ciglitazone, and differentiation media. 3T3-Ll cells were grown to confluence and either exposed to A.) 2-AG ether (50 11M) or B.) ciglitazone (1011M) or C.) differentiation media (1011M dexamethasone, long/ml insulin, 0.5mM IBMX). Cells were cultured 7 days after exposure with fluid renewal every 2 days. Cells were then fixed in 10% formalin and stained with Oil Red 0 and hematoxylin. Images in this dissertation are presented in color. 115 .3 23E a... .... $0.0..th . ..., n 23...... ...... . 3cm... ..EA. tktttfi. . . . ....s .r. . r... a A. A \. . . t . \ . , a , L. ...,3. .- c t... . r\m . P... I i . . . . r . .. c . ., .. . . ,r, ,.. 3 L. _ 1 ..NF u . v f. 1 a a A... ... t s x . . l \l s. c t u .. . . e . e I. x a. 116 ml'm all :sented m M 1 aim 607 50- 40 - Number of differentiated cells per well 30 - 20 - 10 - 0 ... BKG VH 2-AG 2-AG CGZ ether Figure 40. Quantification of 3T3-L1 differentiation. A.) 3T3-L1 preadipocytes were cultured in plastic 2-well culture slides. Upon reaching confluency, the cells were either not treated (NA) or treated with vehicle (0.1 % ethanol), 2-AG (50 11M), 2-AG ether (50 11M), or ciglitazone (10 nM). The cells were then cultured for 7 days, after which the media was removed and the cells were washed with PBS prior to staining with Oil Red 0. B.) Cellular differentiation was quantified by counting the total number of differentiated cells per well. 117 15s r/ I Vehicle é Ciglitazone (10 nM) 2 , 2 2 2 mm. 2 2 2 (fold Inductlon / ¢ ¢ * rel. to BKG) g g g 5‘ / / / 2 2 2 2 2 2 2 2 2 ,e 2 2 2 BKG l 0 1 10 GW9662 (11M) Figure 41. Induction of aP2 by the PPAR'Y agonist, ciglitazone, is attenuated by the PPAR‘y antagonist. GW9662. 3T3 -L1 cells were cultured in 60 mm culture plates and allowed to grow to confluence. The cells were then either left untreated (BKG) or treated with GW9962 (0-25 nM) and either ciglitazone (10 nM) or vehicle (0.1% ethanol). Cells were cultured for 4 days following treatment at which time total RNA was isolated. aP2 mRN A was determined by real-time PCR using Taqman primers and probe. The results are the mean i standard error of quadruplicate cultures. *, p<0.05 as compared to the 0 + ciglitazone group. The data are representative of two separate experiments. 118 produce modestly greater induction of aP2 than 2-AG as compared to the positive control, ciglitazone, at half the concentration of 2-AG, which is likely due to the resistance of 2-AG ether to hydrolysis (Figure 42). In order to fln'ther evaluate the specificity of 2-AG and AG ether in the activation of PPARy, transient transfection experiments in 3T3-L1 cells were conducted using the PPAR'r—LBD/Gal4-DBD Gal 4 luciferase reporter which is activated by ligands for PPARy but not PPARa or PPAR5. Interestingly, 2-AG, 2-AG ether and the positive control, ciglitazone, all exhibited comparable potency on the activation of the PPAR'y—specific luciferase plasmid. These studies confirm that treatment of 3T3-L1 cells with 2-AG and/or 2-AG ether induces PPAR'y activation (Figure 43). TX. The role of PPAR’Y in the suppression of IL-2 by 2-AG, 2-AG ether, and AEA To establish a causal relationship between PPAR‘y activation and suppression of IL-2 by 2-AG and 2-AG ether, studies were conducted to ascertain whether the PPARy—specific antagonist, T0070907, attenuates 2-AG and/or 2-AG ether-mediated IL-2 suppression. Pretreatment of freshly isolated splenocytes with increasing concentrations of T0070907, attenuated 2-AG-mediated IL-2 suppression in a concentration-responsive manner (Figure 44a). Notably, T0070907 treatment alone, at the highest concentration used (10 nM) produced a marked suppression of IL-2 in the absence of 2-AG or 2-AG ether, but only in splenocytes. Identical experiments were performed in the Jurkat T cell. These studies showed that Jurkat cells were more refractory to the suppressive effects of T007 0907, while almost completely abrogating 2-AG and 2-AG ether-mediated inhibition of IL-2 secretion (Figures 44b and 45). 119 Figure 42. Effects of 2-AG and 2-AG ether upon aP2 production. 3T3-L1 cells were cultured in 60 mm culture plates and allowed to grow to confluence. The cells were then either left untreated (NA) or treated with VH (0.1% ethanol), ciglitazone (10 11M) and either A.) 2-AG (50 nM) or B.) 2-AG ether (25 nM). Cells were cultured for 4 days following treatment at which time total RNA was isolated. aP2 mRN A was determined by real-time PCR using Taqman primers and probe. The results are the mean :1: standard error of quadruplicate cultures. *, p<0.05 as compared to VH group. The data are representative of at least two separate experiments. 120 I. this numb) m i; 5mm Iata are 15 - 10 ‘ aP2 mRNA (fold induction relative to BKG) 5 _ BKG VH 2-AG CGZ (50 nM) (10 PM) B. 8 . at 6 .. aP2 mRNA * (fold induction relative to BKG) 4 . BKG VH 2-AG ether CGZ (25 nM) (10 nM) Figure 42. 121 Figure 43. Effects of 2-AG and 2-AG ether upon luciferase activity in 3T3-L1 cells transfected with PPARy-LBD/Gal 4-DBD. 3T3-L1 cells were transiently transfected with the PPAR'y-LBD/Gal4-DBD Gal4 luciferase reporter plasmid. Following transfection, the cells were trypsinized, washed, resuspended in DMEM with 2% BCS, and pooled together. The cells were then cultured in 24-well plates for 5 h prior to treatment. Cells were either lefi untreated (NA), or treated with 0.1% ethanol (V H), 10- 50 uM ciglitazone (CGZ), 1-30 uM 2-AG, or 1-30 uM 2-AG ether. Cells were then incubated for 16-20 h. The luciferase activity was quantified in relative light units (RLU) by chemiluminscence assay. The results are the mean 1- standard error of triplicate cultures. * denotes p<0.05 compared to VH group. These data are representative of two separate experiments. 122 20 3'5 2-AG ether (11M) F_~ ‘3 \‘0 S.\\\\\\\\\\\\\\\\\\\ * I-liHlllllll III II " "lllllllllllll" II” | Illlll llllllll I-1H11 I” || || ””1“" | l Illllllll III” III III III“- T T NA VHIIO 25 50,11 '5 10 20 30,.1 Z-AG (HM) CGZ (HM) 12.5“ 104 75‘ 5.. 25- 0- Luciferase Activity (RLU) Figure 43 123 Figure 44. Effect of the PPAR‘Y antagonist, T0070907, upon suppression of IL-2 by 2-AG in primary murine splenocytes and human Jurkat T cells. A.) Splenocytes (1 x 106 cells/m1) or B.) Jurkat cells (5 x 105 cells/ml) were treated with T0070907 (0.01 — 10 nM) or vehicle (0.02% DMSO) for 30 min prior to treatment with 2-AG (20 11M). Following a 30 min incubation, the cells were then stimulated with 40 nM PMA/0.5 11M ionomycin for 24 h. The supematants were harvested and IL-2 production was measured by ELISA analysis. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean :1: standard error of triplicate cultures. * denotes p<0.05 compared to the VH + 2-AG group. These data are representative of at least two separate experiments. 124 A. 1250' I Vehicle Z-AG (20 pl“) 1000- 750- IL-2 (U/ml) 500- 250- / . _ . 2 BKG p/I L 0 VB 0.1 1 5 10' T0070907 (11M) _ B- 16000- 14000- I Vehicle 12000- 2-AG (20 nM) 10000- IL-2 8000- (pg/ml) 6000- 4000‘ 2000- T0070907(uM) _ Figure. 44 125 I Vehicle '/ - 100001 2 AG ether (20 11M) 7500‘ IL-2 (pg/ml) 500“ ‘ 25001 BKG Pl! [0 VII 0.2 2 10. T0070907 (11M) + Figure 45. Effect of the PPAR? antagonist, T0070907, upon suppression of IL-2 by 2-AG ether in human Jurkat T cells. Jurkat cells (5 x 105 cells/ml) were treated with T0070907 (0.2 — 10 nM) or vehicle (0.02% DMSO) for 30 min prior to treatment with 2-AG ether (20 nM). Following a 30 min incubation, the cells were then stimulated with 40 nM PMA/0.5uM ionomycin for 24 h. The supematants were harvested and IL-2 production was measured by ELISA analysis. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean i standard error of triplicate cultures. * denotes p<0.05 compared to the VH + 2-AG ether group. These data are representative of two separate experiments. 126 Likewise, T0070907 pretreatment also blocked AEA-mediated inhibition of IL-2 production (Figure 46). X. The role of NFAT, NFKB, and AP-l in the suppression of IL-2 by 2-AG Published studies from a number of laboratories have demonstrated that suppression of IL-2 by activated PPAR'y likely involves the transrepression of the transcription factor, NFAT. As such, it is notable that 2-AG also inhibits the production of other cytokines which are regulated by NFAT, such as IL-4, W7, and TNFa (302) (Figure 47). In addition to NFAT, there is evidence to suggest that NFKB and AP-l can also be transrepressed by PPARy. The subsequent studies were designed to ascertain the effect of 2-AG upon the transcriptional activity of NFAT, NFKB, and AP-l in Jurkat T cells transfected with NFAT-luc, NFKB-luc, or AP-l-luc reporters respectively. Preliminary studies were designed to determine the kinetics of NFAT-luc activity as well as to compare different activators. The greatest level of NFAT-luc activity was induced by 80 nM PMA and 1 uM ionomycin at 12 hours and as a result was used for the subsequent experiments (Figure 48). 2-AG treatment caused a concentration-dependent decrease in NFAT transcriptional activity in Jurkat T cells transfected with NFAT-luc, which is consistent with previous studies from this laboratory in EL4 T cells transfected with NFAT-CAT (Figure 49). The magnitude of induction of NFAT-luc activity in Figure 48 is much more robust than that seen in Figure 49, which is likely due to differences in transfection efficiency between the two experiments. Consistent with the effects of 2-AG upon IL-2, the 2-AG-mediated inhibition of NFAT transcriptional activity was also blocked with the PPARy antagonist, T0070907, suggesting that PPARY inhibits NFAT 127 15007 1000 I Vehicle AEA<200M1 IL-2 (U/ml) 500- * ,, '/ . 7 2 2' I / r 2 / I ., 2 2 2 2 2 N.D. I 5 4 l. A 2 BKGPII l0 VH 0.1 1 5 10I T0070907 (nM) b Figure 46. Effect of the PPAR‘y antagonist, T0070907, upon suppression of IL-2 by AEA in primary murine splenocytes. Splenocytes (1 x 10‘5 cells/ml) were pretreated with T0070907 or VH (0.1% DMSO) for 30 min followed by AEA treatment for 30 min. Cells were then stimulated with PMA (40 nM) and ionomycin (0.5 nM) for 24 h. The supernatants were harvested and IL-2 production was measured by ELI SA analysis. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean :1: standard error of triplicate cultures. * p<0.05 compared to the VB + AEA group. These data are representative of at least three separate experiments. 128 Figure 47. Effects of 2-AG upon IFN? and IL-4 production in primary splenocytes. A.) Splenocytes (l x 106 cells/ml) were treated with 2-AG (0.1 — 20 11M) or vehicle (0. 1% ethanol) for 30 min prior to stimulation with 40 nM PMA/0.5 11M. Afier a 6 h incubation, the cells were harvested and total RNA was isolated. IFN‘y and IL-4 mRNA was determined by real-time PCR using predeveloped Taqman primers and probe fi'om Applied Biosystems (Foster City, CA). The results are the mean :t standard error of triplicate cultures. " denotes p<0.05 compared to the VH group. These data are representative of two separate experiments. 129 IFN? mRNA (fold induction rel. to BKG) IL-4 mRNA (fold induction rel. to BKG) Figure 47. 4000 '1 3000 - 2000‘ 1000- * 0.. BKG P/I VH l0.1 1 2.5 S 10 15 20l 2-Arachidonyl glycerol (11M) 4000- 3000- 2000'- 1000‘ * 0— BKG P/I VH L0.1 » a1: a1: 1 2.5 5 10 15 20I 2-Arachidonyl glycerol (11M) 130 4 Figure 48. Effect of PMA/ionomycin and PHA/PMA upon NFAT transcriptional activity. Jurkat T cells (5 x 105 cells/ml) were transiently transfected with the NFAT-luc luciferase reporter plasmid. Following transfection, the cells were washed and resuspended in RPMI with 2% BCS. The cells were then either left untreated (NA) or treated with 40 nM PMA/0.5 11M ionomycin, 80 nM PMA/1 1.1M ionomycin, or PHA/PMA and incubated for 12 hr. NFAT-Inc activity was quantified by chemiluminescence assay. J urkat T cells transfected with the pTA-luc plasmid were treated with 80 nM PMA/1 11M ionomycin and served as a negative control. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean :1: standard error of quadruplicate cultures. These data are representative of two separate experiments. 131 .wv 95w:— asefl ...—em Sefl Na m . k m . .. W W rem -cefi va—m— no.5 1:3 ”8:365 29: o: r 42%?” m :32 E 2.23:3 I .53 5 2.8225... n { ; «z m emu 2.3.9.2 I 18m 132 50'- NFAT-luc (fold induction over BKG) BKGP/IVHII 5 10 20 501 2-Arachidonyl glycerol (11M) > Figure 49. Effect of 2-AG upon NFAT transcriptional activity. Jurkat T cells (5 x 105 cells/ml) were transiently transfected with the NFAT-Inc luciferase reporter plasmid. Following transfection, the cells were washed and resuspended in RPMI with 2% BCS. The cells were then treated with 2-AG (1-50 nM) for 30 min prior to treatment with 80 nM PMA/1 11M ionomycin. The cells were then incubated for 12 hr. NFAT-luc activity was quantified by chemiluminescence assay. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean :1: standard error of quadruplicate cultures. *, p<0.05 as compared to the VH group. These data are representative of two separate experiments. 133 activation resulting in IL-2 suppression (Figure 50). Likewise, 2-AG also repressed the transcriptional activity of both NFKB-luc and to a lesser degree, APl-luc (Figures 51 & 53). Similar to the effects of 2-AG upon NFAT, 2-AG-mediated inhibition of NFKB activity was also antagonized with T0070907, suggesting that transrepression of NFch by PPAR'y also plays a role in the inhibition of IL-2 by 2-AG (Figure 54). Conversely, T0070907 alone causes modest suppression of AP-l-luc activity by itself, which is comparable to that produced by 2-AG, confounding the interpretation of these results (Figure 52). Although the inhibition of AP-l by 2-AG appears to be more robust in Figure 52 than 51, the difference in vehicle effects between the two experiments may contribute to the disparity as 2-AG-mediated inhibition appears more consistent when compared to the PMA/Io controls without vehicle. Collectively, the aforementioned observations suggest that while AP-l is only modestly inhibited by 2-AG, NFAT and NFKB are strongly inhibited by 2-AG and that PPAR'y plays a role in the inhibition of NFAT and NFch. 134 Figure 50. Effect of the PPAR? antagonist, T0070907, upon 2-AG—mediated suppression of NFAT transcriptional activity. Jurkat T cells (5 x 105 cells/ml) were transiently transfected with the NFAT-luc luciferase reporter plasmid. Following transfection, the cells were washed and resuspended in RPMI with 2% BCS. The cells were then either treated with T0070907 (1-10 11M) or VH (0.05% DMSO) and 2-AG (20 nM). Following a 30 min incubation, the cells were then treated with PMA (80 nM) and ionomycin (1 1.1M) and incubated for 12 hr. NFAT-Inc activity was quantified by chemiluminescence assay. J urkat T cells transfected with the pTA-luc plasmid (pTA) were treated with 80 nM PMA/1 11M ionomycin and served as a negative control. NFAT-luc activity was quantified by chemiluminescence assay. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean :t standard error of quadruplicate cultures. These data are representative of two separate experiments. 135 .1 O 2 93 ° CD 0- -‘= < Q I .ediated > N . . I A‘ :ells"m1)wer _ W 3 llloxting ”S The eels * QCQQQ‘; k\\\\\\\\\\\\\\\ In V ' — and 22161313 ) —C{Q u\\\\\\V _, 'A (801011101 1116de “53 5mid(p1.11 Q :ontrol. — (130111011352 1 . as. 1 7: E sultsaffthc . CD 11.20me M m <1 é e': .'= g E In G In F4 —1 N FAT-luc (fold induction over BKG) 136 T0070907 (nM) Figure 50. 12.5- AP-l-luc (fold induction over BKG) 2.5- 0- BKGP/I VHI l 2.5 5 10 15 20I L 2-Arachidonyl glycerol (11M) D Figure 51. Effect of 2-AG upon AP-l transcriptional activity. Jurkat T cells (5 x 105 cells/ml) were transiently transfected with the AP-l-luc luciferase reporter plasmid. Following transfection, the cells were washed and resuspended in RPMI with 2% BCS. The cells were then treated with 2-AG (1-20 nM) for 30 min prior to treatment with 80 nM PMA/1 11M ionomycin. The cells were then incubated for 12 hr. AP-l-luc activity was quantified by chemiluminescence assay. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean :h standard error of quadruplicate cultures. *, p<0.05 as compared to the VH group. These data are representative of two separate experiments. 137 15 n 10 - AP-l-luc (Fold induction over BKG) 5 —r a: 0 l BKG P/I VH L1 5 10l 2-AG (20 nM) T0070907 (M) :> Figure 52. Effect of T0070907 and 2-AG upon AP-l transcriptional activity. Jurkat T cells (5 x 105 cells/ml) were transiently transfected with the AP-l-luc luciferase reporter plasmid. Following transfection, the cells were washed and resuspended in RPMI with 2% BCS. The cells were then treated with 2-AG (1-20 11M) or T0070907 (1 - 10 nM) for 30 min prior to treatment with 80 nM PMA/1 1.1M ionomycin. The cells were then incubated for 12 hr. AP-l-luc activity was quantified by chemiluminescence assay. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean :1: standard error of quadruplicate cultures. *, p<0.05 as compared to the VH group. These data are representative of two separate experiments. 138 NFKB-luc (fold induction over BKG) 0 pTABKP/I VHll 2.5 5 10 15 20I 2-Arachidonyl glycerol (nM) p Figure 53. Effect of 2-AG upon NFKB transcriptional activity. Jurkat T cells (5 x 105 cells/ml) were transiently transfected with the NFKB-luc luciferase reporter plasmid. Following transfection, the cells were washed and resuspended in RPMI with 2% BCS. The cells were then treated with 2-AG (1-20 nM) for 30 min prior to treatment with 80 nM PMA/1 uM ionomycin. The cells were then incubated for 12 hr. NFKB-luc activity was quantified by chemiluminescence assay. Cellular viability was .>_ 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean a: standard error of quadruplicate cultures. *, p<0.05 as compared to the VH group. These data are representative of two separate experiments. 139 Figure 54. Effect of the PPAR! antagonist, T0070907, upon 2-AG-mediated inhibition of NFKB transcriptional activity. Jurkat T cells (5 x 105 cells/ml) were transiently transfected with the NFKB-luc luciferase reporter plasmid. Following transfection, the cells were washed and resuspended in RPMI with 2% BCS. The cells were then either treated with T0070907 (1-10 nM) or VH (0.05% DMSO) and 2-AG (20 11M). Following a 30 min incubation, the cells were treated with PMA (80 nM) and ionomycin (1 11M) and incubated for 12 hr. NFKB-luc activity was quantified by chemiluminescence assay. Jurkat T cells transfected with the pTA-luc plasmid (pTA) were treated with 80 nM PMA/1 11M ionomycin and served as a negative control. Cellular viability was 2 85% for all treatment groups as assessed by trypan blue exclusion. The results are the mean 1 standard error of quadruplicate cultures. These data are representative of two separate experiments. 140 I Vehicle '2 2-AG (20 pM) w R\‘{\\\\\\\\\\\\\\\\V % t:A\\\\\\\\\\\\\\\\\V a; E ..re A\\\\\\\\ _, : s\\\\\\\\V E :A\\\\\' a .... n..— .. CD i: M m . <1 H l l I Q: ‘1' N C E O ’5 g u U o :21 E :4 E 99 1: ‘2 i‘.’ E: 75 6; Ea Z a e E H 41 DISCUSSION 1. Effects of AEA, 2-AG, and 2-AG ether upon IL-2 secretion The putative endogenous cannabinoids, AEA, 2-AG and 2-AG ether, suppress IL-2 secretion in both murine splenocytes as well as human Jurkat T cells, activated with PMA and ionomycin. While 2-AG and 2-AG ether are somewhat more potent in Jurkat T cells than murine splenocytes, AEA is markedly more potent in Jurkat cells compared to splenocytes. Although the reason for the observed differences between the two models is unclear, species differences could result in increased sensitivity of hmnan T cells to 2-AG and AEA. Alternatively, metabolism of ABA, and to a lesser extent 2-AG, by macrophages in the primary splenocyte population, may diminish the activity of ABA and 2-AG and thereby reduce their potency in the mixed cell preparation. In activated Jurkat cells, suppression of IL-2 by ABA and 2-AG is observed at concentrations as low as 100 nM and 2.5 nM, respectively. While this is slightly higher than the calculated endogenous levels, concentrations of ABA and 2-AG exceeding 50 nM and 250 nM, respectively, have been detected in human plasma (381). Interestingly, it has been demonstrated that AEA can be sequestered in cells such that the intracellular concentrations of ABA can be as much as 3 orders of magnitude higher than extracellular concentrations (167). Furthermore, intracellular levels of ABA have been calculated to be as high as 2 uM in Hep2 human laryngeal cells (166). Although cellular accumulation of 2-AG is not nearly as well characterized as that of ABA, there is evidence to suggest that 2-AG is also sequestered intracellularly (244). Measurements of 2-AG and AEA are further confounded by the lability of the compounds. Due to the chemical and metabolic 142 instability of 2-AG and AEA, it seems likely that they are synthesized de novo from membrane components in the locale of their targets. In the case of 2-AG, this is supported by studies showing that diacylglycerol (DAG) is efficiently converted to 2-AG, presumably by a DAG lipase (231). Given the high concentrations of DAG produced in T cells as well as other immune cell types upon activation, it seems likely that the local concentrations of newly synthesized 2-AG are quite high within the local environment of the target in many activated immune cell types. 11. Role of the cannabinoid receptors in the suppression of IL-2 by 2-AG, 2-AG ether and AEA While a variety of immune effects have been attributed to the cannabinoid receptors, the majority of these have relied solely upon cannabinoid receptor antagonists to ascertain the roles of CB1 and CBZ. Studies from our laboratory have demonstrated that Ag-THC-mediated calcium influxes can be attenuated with both SR141716A and SR144528 in splenocytes derived from both wild-type and CB1/CB2 null mice, demonstrating that SR141716A and SR144528 can act at targets other than CB1 and CB2 (unpublished observations). Consequently, studies that report effects of CB] and/or CB2 solely through the use of SR141716A and SR144528 may need to be reevaluated. Although CB2 is expressed at higher levels than C81 in immune cells, several immune effects have been correlated with CB1 activation in myeloid cells. Similarly, the majority of immune effects associated with CB2 activation have also been in myeloid cells. Nonetheless, there have been a few reports of CB2-mediated effects in lymphoid cells, including increased proliferation of CD40-activated B cells, decreased CD8+ T cell migration, induction of TGFB in human peripheral blood lymphocytes, and suppression 143 of macrophage-dependent T cell activation (113, 116, 123, 297). The involvement of CBZ in all of the aforementioned effects, with the exception of the inhibition of macrophage-dependent T cell activation, was determined through the use of the CBZ antagonist, SR144528, or the CB2-specific agonist, JWH133. While it has been demonstrated that the cannabinoid-mediated enhancement of CD40-activated B cell proliferation is blocked with SR144528, the magnitude of the effect is modest (approximately 15% and 25% increase in virgin B cells and germinal center B cells, respectively) and as such, it is unclear if the observed effects are physiologically relevant (113). In human CD8+ T cells, migration induced by stromal cell-derived factor 1 (SDF -1) is inhibited 50% by ABA (40 nM) treatment (297). Likewise, the CBZ-specific agonist, JWH133 (10 nM) produced a similar level of inhibition, which led the authors to believe that the effect was mediated by CB2. This assertion would be greatly strengthened through the use of CBZ null mice or multiple CB2 antagonists. Although the role of CB2 in CD8+ T cell migration is not yet entirely clear, evidence for the role of CB2 in the induction of TGFB production is somewhat stronger. It has been demonstrated that 159-THC induces TGFB production in human peripheral blood lymphocytes, an effect that is completely blocked with SR144528 (10 nM) (123). Although SR144528 was the sole tool used to ascertain CB2 involvement, the concentrations were sufficiently low to support the supposition that SR144528 is blocking CB2 specifically. Nonetheless, the role of CB2 in the induction of TGFB production by 139-THC should be confirmed through alternative approaches. Additionally, the impaired macrophage-dependent T cell activation by Ag-THC has been 144 attributed to CB2 as determined through the use of CB2 null mice (116). As reported by McCoy et al., the impaired macrophage-dependent T cell activation was determined to be due to altered antigen processing by macrophages rather than the direct inhibition of T cells by A9-THC (115). Curiously, A9-THC-mediated inhibition of macrophage- dependent T cell activation was not concentration-dependent, such that low nanomolar concentrations of 159-THC suppressed T cell activation, whereas higher concentrations had no effect. In contrast to the aforementioned findings, studies from this laboratory and others have determined that IL-2 production by splenic T cells and the EL4 T cell line is suppressed by 159-THC, independent of macrophage antigen processing (292, 382). There are a number of differences between the two studies, however, including the concentrations of Ag-THC and the models used. Additionally, the T cells in the study by McCoy et al. were suboptimally activated, which is significant in that suboptimally activated T cells have been shown to exhibit enhanced IL-2 production with cannabinoid treatment (332, 333). CB2 activation also produces effects in mixed immune cell populations. It has been demonstrated by McKallip et al. that apoptosis induced by A9-THC treatment can be blocked with the CBZ antagonist, SR144528, in mixed cell populations, such as splenocytes and thymocytes (114). This is in contrast to the findings of this laboratory, in which no effects upon viability were observed with A9-THC-treated splenocytes. A number of factors could account for these seemingly contradictory observations, including differences in when various responses were measured in experiments, the assays used to assess viability and apoptosis, and the strains and/or stocks of mice used. Importantly, the mode of activation typically used by this laboratory (PMA/Io) induces a 145 particularly robust activation of T cells, which may not necessarily be comparable to the milder modes of T cell activation employed by McKallip et al. Robust activation likely provides strong growth signals, which may make T cells more refractory to apoptosis. Furthermore, because the involvement of CB2 was ascertained solely with SR144528, used in the micromolar range, it is unclear whether the apoptotic effects of A9-THC observed by McKallip et al. are the result of the activation of CB2 or some other target. Given the ambiguity of the role of CB1 and CB2 in cannabinoid-mediated immune effects, a major objective of the current studies was to examine rigorously the role of C81 and CBZ in 2-AG/AEA-mediated suppression of IL-2, a critical cytokine for T cell growth and development. The suppression of IL-2 by 2-AG and AEA in splenocytes derived from CB1/CB2 null mice coupled with the failure of the CB 1/CB2 antagonists to block the decrease in IL-2 by 2-AG and AEA demonstrates that the cannabinoid receptors are not involved. Likewise, the decrease in IL-2 secretion by 2-AG ether is also independent of CB1/CB2, as determined by splenocytes derived from CBl/CB2 null mice. III. Role of the vanilloid receptor, VRl, in the suppression of IL-2 by AEA Although there have been a number of effects by 2-AG and AEA attributed to CB1/C82, there are also a growing number of reports of endocannabinoid activities which are independent of CB 1/CB2 in keeping with the studies reported here (206, 214). One potential mechanism for cannabinoid receptor-independent activity is the vanilloid receptor, VRl. VRl is a ligand-gated cation channel, which is activated by heat, low pH, or capsaicin and other agonists (383, 384). While the efficacy of ABA for VRl is 146 currently the focus of much debate, it is clear that AEA is significantly less potent than capsaicin in the activation of VRl (214, 215, 385, 386, 387). Although VRl is highly expressed in central and peripheral neurons, it is also expressed in many other cell types, including T cells (388). It is notable that the VRl antagonist, capsazepine, at concentrations up to 1 uM did not attenuate AEA-mediated IL-2 suppression in splenocytes. Higher concentrations of capsazepine suppress IL-2 secretion in the absence of ABA and therefore were not used. While the majority of published studies report IC50 values of capsazepine for capsaicin-induced calcium influx from 0.04 — 1 11M, there have been reports of higher ICso values (34 nM), which necessitated the investigation of the effects of the prototypical VRl agonist, capsaicin, upon IL-2 secretion (3 89-393). Despite a published report which demonstrates that capsaicin suppresses IL-2 in activated T cells in a concentration-dependent manner, capsaicin does not decrease IL-2 production in splenocytes under the experimental conditions reported here (3 94). In fact the two studies are not entirely contradictory as capsaicin was not reported to suppress IL-2 production at concentrations up to 20 nM, such that the effects upon IL-2 were only observed at concentrations of 50 uM and higher. The authors assume that the suppression of IL-2 by capsaicin is not mediated by VRl because of the high concentrations of capsaicin required as well as the lack of antagonism by capsazepine (394). There have been reports of other cellular targets for AEA, including the putative endothelial AEA receptor (also called the abnormal cannabidiol receptor), L-type calcium channels, TASK-1 channels, and a G-protein coupled receptor in mouse brain that has not yet been characterized (84, 206, 395-398). While the endothelial AEA receptor has not 147 been cloned and therefore its distribution is unknown, this receptor has only been characterized pharmacologically in endothelial cells and therefore has not been described in lymphoid tissue thus far. In addition to AEA, abnormal cannabidiol, a synthetic cannabidiol analogue, has also been identified as an agonist of the putative endothelial AEA receptor, whereas cannabidiol acts as a partial agonist/antagonist (84, 228, 229, 399). Studies from this laboratory have demonstrated that cannabidiol is more potent in the suppression of IL-2 secretion than AEA, suggesting that the endothelial AEA receptor is not involved (28). Moreover, WIN55212-2, 2-AG and Ag-THC also suppress IL-2 secretion, but are not agonists of the putative endothelial AEA receptor (28, 229, 292, 303) IV. Effects of AEA and 2-AG upon calcium influx Activation of T cells requires the induction of a number of different signal transduction pathways, including MAP kinase, NFKB, and NFAT through a rise in intracellular calcium. The elevation of intracellular calcium levels is the result of the release of internal stores followed by the opening of calcium release-activated channels (CRAC) (400). The rise in intracellular calcium is oscillatory in nature and must be sustained for 30 to 45 min for commitment of the T cell to cellular activation, as evidenced by transcription of IL-2 mRNA (401-404). The induction of IL-2 transcription is achieved by calcium-calmodulin activation of calcineurin, which subsequently dephosphorylates NF AT (404). In the dephosphorylated state, NFAT translocates across the nuclear membrane and binds to the IL-2 promoter (310). Immunosuppressive drugs, such as cyclosporin, inhibit calcineurin activation and thus suppress the activation and translocation of NFAT (309, 405, 406). The kinetics of the calcium signal in relation to 148 the activation of MAP kinase and other signal transduction pathways through the T cell receptor are critical to cell fate. Elevation of intracellular calcium levels in the absence of the full array of signaling pathways necessary for T cell activation results in the induction of an anergic state (3 75). Anergy is defined as a state characterized by the inability of the T cells to become activated, regardless of antigen presentation and costimulation (407). Conversely, T cell activation also cannot occur in the absence of a sustained rise in intracellular calcium (408-410). Because it has been demonstrated that both 2-AG and AEA produce calcium influx in a variety of different cell types, including leukocytes, the effect of 2-AG and AEA upon intracellular calcium levels in splenocytes was investigated (23, 25, 27). Unlike other cannabinoids, neither 2-AG nor AEA induce a rise in intracellular calcium in splenocytes. Previously published studies from this laboratory have determined that suppression of IL-2 transcription by 2-AG is the result of the inhibition of NFAT binding to the IL-2 promoter (303). Additionally, it has been demonstrated that 2-AG is able to inhibit certain ion channels, such as voltage-gated calcium channels (249). Consequently, the effects of 2-AG upon intracellular calcium were investigated in activated splenocytes. 2-AG had little effect upon calcium influx mediated by ionomycin. Because ionomycin induces elevated calcium levels through a variety of different mechanisms, including disruption of the plasma membrane, the effects of 2-AG upon concanavalin A-mediated calcium rise were also studied. The intracellular calcium rise produced by concanavalin A is modest, but adequate for full activation of T cells. While it may appear that 2-AG has a modest suppressive effect upon the initial phase of calcium influx induced by concanavalin A, this observation was not consistent in 149 subsequent experiments. Collectively, this suggests 2-AG does not inhibit calcium rise produced by either ionomycin or concanavalin A and that 2-AG—mediated inhibition of NFAT occurs downstream of the initial calcium influx produced fi'om T cell activation. V. Role of hydrolysis in the suppression of IL-2 by AEA and 2-AG Because arachidonic acid inhibits IL-2 production with comparable potency to that of ABA and 2-AG, the role of hydrolysis in the suppression of IL-2 by 2-AG and AEA was investigated. The hydrolysis of ABA is believed to occur through the combined actions of the putative AMT and FAAH, therefore the individual roles of both were considered (143, 152, 172). Likewise, it has also been demonstrated that the AMT may also play a role in the transport of 2-AG across the membrane (244). The inability of inhibitors of the putative AMT to affect the suppression of IL-2 by AEA or 2-AG suggests that the AMT represents neither a mode of activation nor inactivation in this particular system. Likewise, 2-AG/AEA-mediated suppression of IL-2 is unaffected by pretreatment with MAFP, an inhibitor of all enzymes identified thus far which hydrolyze ABA and 2-AG. Furthermore, 2-AG ether, the non-hydrolyzable analogue of 2-AG, produces a concentration-dependent suppression of IL-2, which is similar to that observed with 2-AG. Moreover, studies from this laboratory have also demonstrated that fluoro-methanandamide, an analogue of AEA that is resistant to hydrolysis, also results in decreased IL-2 secretion similar to that observed with AEA (411). Collectively, these results suggest that hydrolysis is not necessary for the suppression of IL-2 by 2-AG and AEA. 150 There have been a number of reports that AEA mediates arachidonic acid release from various cell types, including human peripheral blood mononuclear cells (296, 412). Therefore, an alternative explanation for the observations of the current investigation is that AEA causes release of arachidonic acid from primary splenocytes. The released arachidonic acid is then subsequently metabolized into an eicosanoid, which mediates the inhibition of IL-2 secretion. While the current studies do not negate the possibility that AEA causes arachidonic acid release, the striking similarity in the concentration responses of ABA and arachidonic acid in addition to their ICso values (11.4 11M and 10.3 nM, respectively) suggests that the same moiety is responsible for the activity of both. VI. Role of COX in the suppression of IL-2 by AEA and 2-AG While the vast majority of research on AEA and 2-AG metabolism has focused upon hydrolysis, there has also been considerable interest in the role of the COX enzymes. There are two isoforms of the COX enzyme, COX-1 and COX-2. COX-1 is constitutively expressed in most cell types, whereas COX-2 expression is induced in response to a stimulus in certain cell types (336). Both COX enzymes have been found in most cell types within the immune system, including T cells (334). It has been reported that ABA and 2-AG can bind directly to COX-2 and be metabolized into PGEz- glyceryl ester or PGEz—ethanolamine as well as a wide array of other COX products (176, 177, 245). The same studies have also demonstrated, however, that ABA and 2-AG neither bind directly to, nor are oxygenated by, COX-l. The attenuation of AEA-mediated IL-2 suppression by the nonspecific COX inhibitors, flurbiprofen and piroxicam, implicates COX metabolism in the observed 151 effects of ABA upon IL—2. Unlike the COX-l-specific inhibitors, SC560 and FR122047, the COX-2-specific inhibitor, NS398, attenuates the inhibitory effects of ABA upon IL-2 secretion in splenocytes. Likewise, flurbiprofen and NS398, but not the COX-1 specific inhibitors, are effective in blocking the suppression of IL-2 by 2-AG in Jurkat T cells. Collectively, this suggests that COX-2 metabolites of 2-AG and AEA rather than the parent molecules themselves, are responsible for the observed effects upon lL-2 production in activated T cells. Interestingly, the COX-1 selective inhibitor, piroxicam, attenuates the suppression of IL-2 by ABA and arachidonic acid in murine splenocytes, but is not particularly effective in blocking the suppression of IL-2 by 2-AG in Jurkat T cells. This is likely due to species differences in the COX-2 protein as other laboratories have demonstrated that piroxicam is markedly less effective than other COX inhibitors in the inhibition of human COX-2. It is curious, however, that piroxicam is a potent and effective analgesic and anti-inflammatory agent in humans. While the reason for the discrepancy is unknown, it may be that piroxicam also acts at other targets in addition to the COX enzymes. There is evidence to support that a number of different prostanoids can inhibit IL-2 secretion, including Per—:2, 15-deoxy-A'2’” PGJ2, and PGI; (338-340). Of the aforementioned eicosanoids, lS-deoxy-A'z’l4 PGJ; (15d-PGJ2) is the most recent to be recognized to suppress IL-2 secretion. There has been growing interest in 15d-PGJ2 in a number of different research areas due to its identification as one of the most potent endogenous ligands of PPAR’y (341, 413). While the lack of IL-2 suppression by PGEz- ethanolamine eliminates it as the ABA metabolite responsible for the effects upon IL-2 secretion, the potent immunosuppressive effects of 15d- PG]: and PG]; upon IL-2 suggest 152 a possible role for the ethanolamine and glycerol ester analogues of the J-series of prostaglandins in the suppression of IL-2. 15d-PGJz is produced through the sequential metabolism of arachidonic acid by COX and PGD synthase, followed by a series of nonenzymatic transformations (414). Whether 15d-PGJ2 is produced in vivo has been somewhat of a controversy, however recent studies have identified elevated levels of 15d-PGJ2 in LPS-stimulated RAW264.7 macrophages (415). Future studies will determine whether 15d-PGJz is produced in other cell types, but it is notable that both enzymes required for 15d—PGJ2 formation, COX and PGD synthase, are expressed in T cells (3 34). While lSd—PGJ; is one of the most potent activators of PPARy identified thus far, other COX products are also able to activate PPAR'y (341). More studies will be needed to identify the products resulting from the COX-2 metabolism of ABA and 2-AG, which are responsible for the observed decrease in IL-2 production. While in the majority of cases studied thus far the metabolism of ABA and 2-AG has been associated with the cessation of their physiological activity, the present studies are significant because they show that metabolism of ABA and 2-AG in certain cases may yield biologically active products. It has been demonstrated here that the metabolism of AEA and 2-AG by COX-2 yields products which suppress IL-2 secretion. VII. Role of PPAR? in the inhibition of IL-2 secretion by 2-AG, 2-AG ether, and AEA Structurally related to the hormone receptors, PPARs are also members of the nuclear receptor superfamily. Currently, there are three subtypes of PPARs, which have 153 been identified: PPARa, PPARS, and PPAR'Y (413, 416-418). While the function of PPAR5 remains unclear, PPARa and PPAR‘y have been best characterized for their roles in lipid metabolism and have only recently been discovered to be involved with immune regulation as well (419). Although both PPARa and PPAR'y are expressed in T cells, only activated PPAR'y has been shown to suppress IL-2 secretion (340, 419). Physical association of activated PPAR'Y with the transcription factor, NFAT, results in transrepression of NFAT binding to the IL-2 promoter, which is thought to be the mechanism for the decrease in IL-2 production by PPAR‘Y agonists (340). Similarly, activated PPAR'y has also been shown to cause a decrease in cytokine production in activated macrophages and T cells through direct protein-protein interaction with NFKB, a key transcription factor for a number of different cytokines (344, 345, 362). Because 2-AG closely resembles the structures of known PPAR'Y agonists and both 2-AG and activated PPAR’y suppress IL-2 secretion through inhibition of NFAT, the role of PPARy in 2-AG-mediated IL-2 suppression was investigated. The initial studies in this line of investigation were designed to determine whether 2-AG and 2-AG ether activate PPAR‘y. The induction of adipogenesis and aP2 mRN A transcription suggested that indeed 2-AG/2-AG ether treatment results in PPAR'y activation. In order to determine more conclusively that 2-AG and/or 2-AG ether treatment activates PPARy specifically, subsequent experiments employing a PPARy trans-acting reporter gene construct were performed. The plasmid construct contains sequence for two components: a fusion protein containing the ligand binding domain (LBD) of PPARY fused to the DNA binding domain (DBD) of Gal4, as well as the luciferase reporter, which is activated by the binding of the fusion protein to Gal4 response elements. Activation of 154 the PPARy-LBD/Gal4-DBD luciferase plasmid demonstrates that treatment of 3T3 -L1 cells with 2-AG and/or 2-AG ether results in PPAR'y activation. While 2-AG and 2-AG ether suppress IL-2 in primary splenocytes in the low micromolar range, both compounds appear to be less potent in the 3T3-Ll model. This is likely due to several factors including, the efficient processing of fatty acids in the 3T3 -L1 fibroblasts, which is well established, and the labile properties of 2-AG in long- terrn culture assays (420). Additionally, 2-AG ether is more potent than 2-AG in assays of longer duration, including aP2 induction and adipogenesis, which may be due to its resistance to hydrolysis. Although the induction of PPAR'y-LBD/Gal4-DBD luciferase activity by 2-AG and 2-AG ether suggests that PPAR'y is activated, the role of PPAR'y in the suppression of IL-2 by 2-AG and 2-AG ether was still unknown. While studies from this laboratory and others have demonstrated that PPAR}! agonists, such as ciglitazone and 15d-PG12, suppress IL-2 secretion, the PPARY antagonist, T0070907, was used to establish causality between activation of PPAR'y by 2-AG treatment and suppression of IL-2. Although the PPAR'Y antagonist, T0070907, attenuates suppression of IL-2 by 2-AG and AEA in primary splenocytes in a concentration-dependent manner, T0070907 was unable to completely abrogate the effects of 2-AG and AEA in this model. The suppression of IL-2 by T0070907 in the absence of 2-AG or AEA is likely the reason why it does not completely block the effects of 2-AG and AEA in primary splenocytes. Unlike murine splenocytes, Jurkat T cells are more refiactory to the suppressive effects of T0070907 upon IL-2 secretion. As such, T007 0907 pretreatment results in full reversal of the IL-2 155 suppression by 2-AG and 2-AG ether in Jurkat cells. Collectively, these results suggest a role for PPARy in the suppression of IL-2 production by 2-AG, 2-AG ether, and AEA. While PPAR'y-specific luciferase induction by ciglitazone, 2-AG, and 2-AG ether may not appear to correlate with the magnitude of IL-2 suppression observed with the same compounds, this is likely due to major differences between the two assays. The inhibition of IL-2 secretion by PPAR] agonists is dependent upon endogenous PPARy and RXR levels, whereas luciferase induction is dependent upon the expression of the PPARy/Gal fusion protein from the exogenous plasmid in the transfected cells. Furthermore, endogenous PPAR'Y competes with the PPAR‘y/Gal fusion protein for agonists in the luciferase assay. Additionally, other laboratories have shown a . comparable level of luciferase activity with a similar type of plasmid (trans-acting PPARY plasmid) in NIH3T3 cells (247). Although several studies have emerged which suggest that under certain conditions, activated PPARY may cause apoptosis in activated T cells, there are also a number of studies suggesting that PPARy activation may be protective against apoptosis or have no effect upon cell viability (354, 356, 357, 359, 419). While the cause of the differential effects of activated PPAR'Y upon T cell viability is unclear, it is likely that the concentration of 15d-PGJ2 or other agonists, the kinetics of cell treatment, and/or the level/mode of activation may be contributing factors. Under the specific conditions used in the present studies, there was no detectable effect upon cell viability. In addition, it is noteworthy that ABA and 2-AG may be metabolized into a number of different products, some of which may be protective against apoptosis. 156 Early studies examining the role of PPARy in immune cells indicated that PPAR'y agonists cause a variety of immune effects, which tend to be suppressive in nature (419). The initial research on immune effects by PPAR'y was generally limited to the use of PPAR'y agonists due to the lack of potent PPAR'y-specific antagonists at that time as well as to the lack of PPAR'y knock-out mice, which were found to be embryonic lethal. The interpretation of the aforementioned early studies has become controversial due to a recent report showing that rosiglitazone and 15-deoxy PGJz have anti-inflammatory effects in macrophages which lack PPAR'y (346). As a result, the role of PPAR'y in many of the observed inhibitory immune effects by PPAR'Y agonists has been called into question and may need to be confirmed. As such, the current studies are important because the use of a reporter construct and an antagonist which are both highly specific for PPAR'y, provide evidence that the observed effects of 2-AG and 2-AG ether are in fact mediated by PPAR'y. While initial studies of PPAR'y centered around lipid metabolism and glucose homeostasis, the breadth of knowledge concerning the role of PPARy in immune regulation is rapidly increasing. Recently it has been reported that activation of PPAR'y can ameliorate autoimmune disease in a number of different animal models, including allergic asthma, experimental crescentic glomerulonephritis, as well as experimental allergic encephalitis, an animal model of multiple sclerosis (363, 364, 367). Furthermore, PPAR'Y activation causes anti-inflammatory and anti-proliferative effects in T cells derived from human multiple sclerosis patients (371). Similarly, the modulation of a variety of immune responses by 2-AG and AEA, including IL-2 secretion, suggests that 2-AG and AEA may also play an important role in immune regulation. Further evidence 157 for the regulatory role of 2-AG and AEA in the immune system comes from a number of published studies showing that 2-AG and AEA levels are markedly increased upon the activation of various immune cell types (234-236). Additionally, 2-AG and AEA levels are much higher in the sera of patients suffering from LPS-induced shock as compared to healthy control subjects suggesting that 2-AG and AEA may also play a regulatory role in sepsis (164). Collectively, this suggests that PPARy may play an important role in the control of exaggerated or inappropriate immune responses, and that activation of PPAR‘y by 2-AG and AEA may be contribute to the maintenance of immune homeostasis. VIII. Role of NFAT, NFKB, and AP-l in the suppression of IL-2 secretion by 2-AG Although PPAR'y is best known to regulate gene expression through binding to PPAR response elements (PPREs) as a heterodimer with the RXR, a number of alternative modes of gene regulation by PPAR'y have been described (342). Like other nuclear receptors, activated PPAR'y has been shown to coassociate physically with transcription factors and thus to sequester these transcription factors from binding to their response elements in the regulatory regions of target genes. In macrophages, activated PPARY has been shown to cause a decrease in AP-l, STAT-1, and NFKB DNA-binding, which ultimately results in a repression of gene induction (344). Moreover, two separate laboratories have shown that activated PPARy physically associates with NFAT and NFKB, which ultimately results in the suppression of IL-2 and IL-4 transcription (340, 345,362) Previously published studies from this laboratory have shown that suppression of IL-2 by 2-AG occurs at the transcriptional level through inhibition of NFAT, an essential 158 transcription factor for IL-2 mRN A production (303). The aforementioned studies showed that 2-AG causes suppression of NFAT DNA-binding and transcriptional activity in splenocytes and EL4 cells. Additionally, NFKB DNA-binding and transcriptional activity were also decreased by 2-AG treatment, albeit to a lesser extent than that of NFAT. The present studies demonstrate that 2-AG produces marked inhibition of NFAT and NFKB, but only modest inhibition of AP-l, in Jurkat T cells. Moreover, the PPAR'y antagonist, T0070907, blocks the inhibition of NFAT and NFKB by 2-AG. Collectively, this suggests that the suppression of NFAT and NFKB activation by 2-AG is mediated through activation of PPAR'y. IX. Summary The studies reported in this dissertation demonstrate that 2-AG, 2-AG ether, and AEA suppress IL-2 secretion in both activated murine splenocytes and human Jurkat T cells. The decrease in IL-2 production by 2-AG and AEA in splenocytes derived from CB1/C82 null mice coupled with the failure of the CB1/CB2 antagonists to block inhibition of IL-2 secretion by 2-AG and AEA demonstrates that the cannabinoid receptors are not involved. Likewise, 2-AG ether also suppresses lL-2 secretion independent of CB1/CB2, as determined by splenocytes derived from CB1/CB2 null mice. With the elimination of the cannabinoid receptors as the mediators of the suppression of lL-2 by 2-AG, 2-AG ether, and AEA, other targets were considered. The inability of capsazepine to block AEA-mediated IL-2 suppression coupled with the lack of effect by capsaicin upon IL-2 secretion, suggests that VRl is not involved. Additionally, the inability of 2-AG to modulate calcium influx in resting or activated 159 splenocytes, suggests that 2-AG does not disrupt the calcium signal initiated upon T cell activation. Similarly, AEA does not induce calcium influx in resting splenocytes. Because arachidonic acid inhibits IL-2 production with comparable potency to that of ABA and 2-AG, the role of hydrolysis in the suppression of IL-2 by 2-AG and AEA was investigated. Neither inhibitors of the putative AMT nor MAFP blocked suppression of IL-2 by ABA and 2-AG, suggesting that the effects of 2-AG and AEA are not dependent upon uptake and hydrolysis. Conversely, flurbiprofen pretreatment significantly attenuated suppression of IL-2 by ABA and 2-AG, suggesting a role for COX metabolism. Unlike the COX-1 specific inhibitors, SC560 and FR122047, the COX-2 specific inhibitor, NS398, attenuated AEA/2-AG-mediated IL-2 suppression, which implicates COX-2 rather than COX-1 in the metabolism of ABA and 2-AG. Because a number of COX metabolites are known agonists of PPAR'Y and it has been demonstrated that activated PPAR'Y suppresses IL-2 secretion, the role of PPAR'y in the suppression of IL-2 by 2-AG and AEA was investigated. Initially it was determined that treatment of 3T3-L1 cells with 2-AG or 2-AG ether results in PPARy activation as evidenced by induction of adipogenesis, increased aP2 transcription, and induction of PPAR'y-LBD/Ga14-DBD luciferase transcriptional activity. Furthermore, the PPAR‘y antagonist, T0070907, attenuated suppression of IL-2 by 2-AG, 2-AG ether, and AEA. Moreover, T0070907 also blocked inhibition of NFAT and NFKB transcriptional activity by 2-AG. Collectively, the aforementioned observations show that suppression of IL—2 by 2-AG and AEA occurs independently of CB1/CB2, VRl as well as the putative AMT and hydrolytic enzymes. Moreover, our findings suggest that suppression of IL-2 by 160 2-AG and AEA is mediated by a COX-2 metabolite, which activates PPAR'y resulting in the transrepression of the transcription factors, NFAT and NFKB (Figure 55). The findings of this dissertation are relevant in that they challenge the pervasive assumption that the immune effects of 2-AG and AEA are mediated through activation of CB2, and to a lesser extent, CB1. While the cannabinoid receptors may mediate some of the immune effects produced by 2-AG and AEA, the present studies demonstrate that other mechanisms are involved in the suppression of IL-2. Additionally, the present studies are also significant in that they demonstrate that the metabolism of 2-AG and AEA does not always result in the cessation of physiological activity, but in certain circumstances may be responsible for the observed biological effects. Furthermore, the current studies are unique in that they suggest a novel mechanism of action in which metabolites of 2-AG/AEA activate PPAR'y resulting in the inhibition of NFAT and NFKB and consequently the suppression of IL-2 production. In addition to IL-2 suppression, activation of PPAR'Y by 2-AG/AEA metabolites may also play a role in other observed physiological effects of 2-AG and AEA that are independent of CBl/CB2. Moreover, it has been demonstrated that the levels of 2-AG and AEA are elevated in a number of different immune cell types upon activation and that both PPAR'y and 2-AG/AEA ameliorate the symptoms of a variety of different animal models of autoimmune diseases. Collectively, this suggests that activation of PPARY by 2-AG and AEA may play an important role in the control of exaggerated or inappropriate immune responses, thereby maintaining immune homeostasis. 161 i=3 R. @3953 5 $3. was warm 3 as: .3 53.5.5 we 5555830.. uuanAom .mm 953% .5389..— and _ a: 33.. as 5.5.. as 5.5.. _ biz :22 biz mDmAUDZ / 3.5.8.55.— 380 1111' 5.2.3.0.. V ES: a.ease...asee?.e.”2.22.21.22.42 O O O O O O O O