A Molecular Link between the Active Component of Marijuana and Alzheimer’s Disease Pathology Lisa M. Eubanks,

A Molecular Link between the Active Component ofMarijuana and Alzheimer’s Disease PathologyLisa M. Eubanks,†Claude J. Rogers,†Albert E. Beuscher IV,‡George F. Koob,§Arthur J. Olson,‡Tobin J. Dickerson,†and Kim D. Janda*,†Departments of Chemistry, Immunology, and Molecular Biology, Molecular andIntegrated Neurosciences Department, The Skaggs Institute for Chemical Biology, andWorm Institute for Research and Medicine, The Scripps Research Institute,10550 North Torrey Pines Road, La Jolla, California 92037Received June 11, 2006Abstract: Alzheimer’s disease is the leading cause of dementia among the elderly, and withthe ever-increasing size of this population, cases of Alzheimer’s disease are expected to tripleover the next 50 years. Consequently, the development of treatments that slow or halt the diseaseprogression have become imperative to both improve the quality of life for patients and reducethe health care costs attributable to Alzheimer’s disease. Here, we demonstrate that the activecomponent of marijuana, ¢9-tetrahydrocannabinol (THC), competitively inhibits the enzymeacetylcholinesterase (AChE) as well as prevents AChE-induced amyloid â-peptide (Aâ)aggregation, the key pathological marker of Alzheimer’s disease. Computational modeling ofthe THC-AChE interaction revealed that THC binds in the peripheral anionic site of AChE, thecritical region involved in amyloidgenesis. Compared to currently approved drugs prescribedfor the treatment of Alzheimer’s disease, THC is a considerably superior inhibitor of Aâaggregation, and this study provides a previously unrecognized molecular mechanism throughwhich cannabinoid molecules may directly impact the progression of this debilitating disease.Keywords: Cannabinoids; Alzheimer’s disease; acetylcholinesteraseIntroductionSince the characterization of the Cannabis satiVa producedcannabinoid, ¢9-tetrahydrocannabinol (THC) (Figure 1), inthe 1960s,1this natural product has been widely explored asan antiemetic, anticonvulsive, anti-inflammatory, and analgesic.2In these contexts, efficacy results from THC bindingto the family of cannabinoid receptors found primarily oncentral and peripheral neurons (CB1) or immune cells (CB2).3More recently, a link between the endocannabinoid systemand Alzheimer’s disease has been discovered4which hasprovided a new therapeutic target for the treatment of patientssuffering from Alzheimer’s disease.5New targets for this* Author to whom correspondence should be addressed. Mailingaddress: Department of Chemistry, The Scripps ResearchInstitute and the Skaggs Institute for Chemical Biology, 10550North Torrey Pines Rd., La Jolla, CA 92037. Tel: 858-784-2515. Fax: 858-784-2595. E-mail: kdjanda@scripps.edu.†Departments of Chemistry and Immunology, The Skaggs Institutefor Chemical Biology, and Worm Institute for Research andMedicine (WIRM).‡Department of Molecular Biology.§Molecular and Integrated Neurosciences Department (MIND).(1) Gaoni, Y.; Mechoulam, R. Isolation, structure, and partial synthesisof an active constituent of hashish. J. Am. Chem. Soc. 1964, 86,1646-1650.(2) Carlini, E. A. The good and the bad effects of (-) trans-delta-9-tetrahydrocannabinol (¢9-THC) on humans. Toxicon 2004, 44,461-467.Figure 1. Chemical structure of ¢9-tetrahydrocannabinol(THC).brief articles10.1021/mp060066m CCC: $33.50 © XXXX American Chemical Society VOL. XXXX NO. XXXX MOLECULAR PHARMACEUTICS APublished on Web 08/09/2006 PAGE EST: 4.6debilitating disease are critical as Alzheimer’s disease afflictsover 20 million people worldwide, with the number ofdiagnosed cases continuing to rise at an exponential rate.6,7These studies have demonstrated the ability of cannabinoidsto provide neuroprotection against â-amyloid peptide (Aâ)toxicity.8-10Yet, it is important to note that, in these reports,cannabinoids serve as signaling molecules which regulatedownstream events implicated in Alzheimer’s disease pathology and are not directly implicated as effecting Aâ at amolecular level.One of the primary neuropathological hallmarks of Alzheimer’s disease is deposition of Aâ into amyloid plaquesin areas of the brain important for memory and cognition.11Over the last two decades, the etiology of Alzheimer’sdisease has been elucidated through extensive biochemicaland neurobiological studies, leading to an assortment ofpossible therapeutic strategies including prevention of downstream neurotoxic events, interference with Aâ metabolism,and reduction of damage from oxidative stress and inflammation.12The impairment of the cholinergic system is themost dramatic of the neurotransmitter systems affected byAlzheimer’s disease and, as a result, has been thoroughlyinvestigated. Currently, there are four FDA-approved drugsthat treat the symptoms of Alzheimer’s disease by inhibitingthe active site of acetylcholinesterase (AChE), the enzymeresponsible for the degradation of acetylcholine, therebyraising the levels of neurotransmitter in the synaptic cleft.13In addition, AChE has been shown to play a further role inAlzheimer’s disease by acting as a molecular chaperone,accelerating the formation of amyloid fibrils in the brain andforming stable complexes with Aâ at a region known as theperipheral anionic binding site (PAS).14,15Evidence supporting this theory was provided by studies demonstrating thatthe PAS ligand, propidium, is able to prevent amyloidacceleration in vitro, whereas active-site inhibitors had noeffect.16Due to the association between the AChE PAS andAlzheimer’s disease, a number of studies have focused onblocking this allosteric site.17Recently, we reported acombined computational and experimental approach toidentify compounds containing rigid, aromatic scaffoldshypothesized to disrupt protein-protein interactions.18-20Similarly, THC is highly lipophilic in nature and possessesa fused tricyclic structure. Thus, we hypothesized that thisterpenoid also could bind to the allosteric PAS of AChE withconcomitant prevention of AChE-promoted Aâ aggregation.Experimental SectionDocking Procedures. THC was docked to the mouseAChE structure (PDB ID code 1J07) using AutoDock 3.0.5.21Twenty docking runs (100 million energy evaluations each)were run with a 26.25 Å 18.75 Å 26.25 Å grid box(3) Howlett, A. C.; Barth F.; Bonner, T. I.; Cabral, G.; Casellas P.;Devane, W. A.; Felder, C. C.; Herkenham, M.; Mackie, K.;Martin, B. R.; Mechoulam, R.; Pertwee, R. G. International Unionof Pharmacology. XXVII. Classification of cannabinoid receptors.Pharmacol. ReV. 2002, 54, 161-202.(4) Benito, C.; Nunez, E.; Tolon, R. M.; Carrier, E. J.; Rabano, A.;Hillard, C. J.; Romero, J. Cannabinoid CB2 receptors and fattyacid amide hydrolase are selectively overexpressed in neuriticplaque-associated glia in Alzheimer’s disease brains. J. Neurosci.2003, 23, 11136-11141.(5) Pazos, M. R.; Nu´n˜ez, E.; Benito, C.; Tolo´n, R. M.; Romero, J.Role of the endocannabinoid system in Alzheimer’s disease: newperspectives. Life Sci. 2004, 75, 1907-1915.(6) Ritchie, K.; Kildea, D. Is senile dementia “age-related” or “ageingrelated”?sevidence from meta-analysis of dementia prevalencein the oldest old. Lancet 1995, 346, 931-934.(7) Evans, D. A. Estimated prevalence of Alzheimer’s disease in theUnited States. Milbank Q. 1990, 68, 267-289.(8) Milton, N. G. Anandamide and noladin ether prevent neurotoxicityof the human amyloid-â peptide. Neurosci. Lett. 2002, 332, 127-130.(9) Iuvone, T.; Esposito, R.; Santamaria, R.; Di Rosa, M.; Izzo, A.A. Neuroprotective effect of cannabidiol, a non-psychoactivecomponent from Cannabis sativa, on beta-amyloid-induced toxicity in PC12 cells. J. Neurochem. 2004, 89, 134-141.(10) Ramı´rez, B. G.; Bla´zquez, C.; Go´mez del Pulgar, T.; Guzma´n,M.; de Ceballos, M. L. Prevention of Alzheimer’s diseasepathology by cannabinoids: neuroprotection mediated by blockadeof microglial activation. J. Neurosci. 2005, 25, 1904-1913.(11) Rozemuller, J. M.; Eikelenboom, P.; Stam, F. C.; Beyreuther, K.;Masters, C. L. A4 protein in Alzheimer’s disease: primary andsecondary cellular events in extracellular amyloid deposition. J.Neuropathol. Exp. Neurol. 1989, 48, 674-691.(12) Bachurin, S. O. Medicinal chemistry approaches for the treatmentand prevention of Alzheimer’s disease. Med. Res. ReV. 2003, 23,48-88.(13) Racchi, M.; Mazzucchelli, M.; Porrello, E.; Lanni, C.; Govoni,S. Acetylcholinesterase inhibitors: novel activities of old molecules. Pharmacol. Res. 2004, 50, 441-451.(14) Inestrosa, N. C.; Alvarez, A.; Pecez, C. A.; Moreno, R. D.;Vicente, M.; Linker, C.; Casanueva, O. I.; Soto, C.; Garrido, J.Acetylcholinesterase accelerates assembly of amyloid-â-peptidesinto Alzheimer’s fibrils: possible role of the peripheral site ofthe enzyme. Neuron 1996, 16, 881-891.(15) Alvarez, A.; Alarcon, A.; Opazo, C.; Campos, E. O.; Munoz, F.J.; Calderon, F. H.; Dajas, F.; Gentry, M. K.; Doctor, B. P.; DeMello, F. G.; Inestrosa, N. C. Stable complexes involvingacetylcholinesterase and amyloid-â peptide change the biochemical properties of the enzyme and increase the neurotoxicity ofAlzheimer’s fibrils. J. Neurosci. 1998, 18, 3213-3223.(16) Bartolini, M.; Bertucci, C.; Cavrini, V.; Andrisano, V. â-Amyloidaggregation induced by human acetylcholinesterase: inhibitionstudies. Biochem. Pharmacol. 2003, 65, 407-416.(17) Johnson, G.; Moore, S. W. The peripheral anionic site ofacetylcholinesterase: structure, functions and potential role inrational drug design. Curr. Pharm. Des. 2006, 12, 217-225.(18) Dickerson, T. J.; Beuscher, A. E., IV; Rogers, C. J.; Hixon, M.S.; Yamamoto, N.; Xu, Y.; Olson, A. J.; Janda, K. D. Discoveryof acetylcholinesterase peripheral anionic site ligands throughcomputational refinement of a directed library. Biochemistry 2005,44, 14845-14853.(19) Xu, Y.; Shi, J.; Yamamoto, N.; Moss, J. A.; Vogt, P. K.; Janda,K. D. A credit-card library approach for disrupting protein-proteininteractions. Bioorg. Med. Chem. 2006, 14, 2660-2673.(20) Xu, Y.; Lu, H.; Kennedy, J. P.; Yan, X.; McAllister, L. A.;Yamamoto, N.; Moss, J. A.; Boldt, G. E.; Jiang, S.; Janda, K. D.Evaluation of “credit card” libraries for inhibition of HIV-1 gp41fusogenic core formation. J. Comb. Chem. 2006, 8, 531-539.brief articles Eubanks et al.B MOLECULAR PHARMACEUTICS VOL. XXXX NO. XXXXwith 0.375 Å grid spacing. This grid box was designed toinclude regions of both the catalytic site and the peripheralanionic site. Otherwise, standard docking settings were usedfor the AutoDock calculations, as previously detailed.18Acetylcholinesterase Inhibition Studies. All assays wereperformed using a Cary 50 Bio UV-visible spectrophotometer using an 18-cell changer, and conducted at 37 °C, usinga Cary PCB 150 water Peltier system. Solutions of acetylthiocholine iodide (ATCh iodide) and 5,5′-dithiobis(2-nitrobenzoic) acid (DTNB) were prepared according to themethod of Ellman et al.22Stock solutions of acetylcholinesterase from Electrophorus electricus were prepared bydissolving commercially available enzyme in 1% gelatin.Prior to use, an aliquot of the gelatin solution was diluted1:200 in water. For the assay, the solution was diluted untilenzyme activity between 0.1 and 0.13 AU/min at 500 íMACTh iodide was obtained. Compounds were prepared assolutions in methanol.Assays were performed by mixing AChE, THC, and 340íM DTNB in 100 mM phosphate buffer, pH 8.0, containing5% methanol. Solutions were incubated at 37 °C for 5 minbefore the reaction was initiated by the addition of ATChiodide (75-300 íM). The increase of absorbance at 412 nmwas monitored for 2-5 min. All assays were run in triplicate.Initial rates were determined by subtracting the averageobserved initial rate from the nonenzymatic reaction.Linear regression analysis of reciprocal plots of 1/Vo versus1/[S] for four THC concentrations was performed usingMicrosoft Excel software. The slope was plotted against [I]to give Ki values. Propagation of error was performed todetermine the error, ¢Ki.For studies to determine the mutual exclusivity of THCand propidium iodide, experiments were performed identically to THC inhibition studies with a fixed concentrationof ACTh iodide (125 íM), and varied concentrations ofpropidium iodide (0-25 íM) and THC (0-15 íM).AChE-Induced â-Amyloid Peptide Aggregation in thePresence of AChE Ligands. The aggregation of the â-amyloid peptide was measured using the thioflavin T basedfluorometric assay as described by LeVine23and Bartolini.16Assays were measured using a SpectraMAX Gemini fluorescence plate reader with SOFTmax PRO 2.6.1 software.Aâ1-40 stock solutions were prepared in DMSO and HuAChEstocks prepared in distilled water. All stock solutions of Aâand HuAChE were used immediately after preparation.In a 96-well plate, triplicate samples of a 20 íL solutionof 23 nM Aâ, 2.30 íM HuAChE, and various concentrationsof THC in 0.215 M sodium phosphate buffer, pH 8.0, wereprepared. These solutions were incubated at room temperature along with triplicate solutions of Aâ alone, Aâ withAChE, and Aâ with varying concentrations of THC. After48 h, a 2 íL aliquot was removed from each well, placed ina black-walled, clear-bottomed 96-well plate, and dilutedwith 50 mM glycine-NaOH buffer, pH 8.5, containing 1.5íM thioflavin T to a total volume of 200 íL. After incubationfor 5 min, the fluorescence was measured using ìexc ) 446nm and ìem ) 490 nm with excitation and emission slits of2 nm. The fluorescence emission spectrum was recordedbetween 450 and 600 nm, with excitation at 446 nm.The fluorescence intensities were averaged, and theaverage background fluorescence of buffer only, or bufferand THC, was subtracted. The corrected fluorescence valueswere plotted with their standard deviation. The equation, Fi/Fo 100%, where Fiis the fluorescence of AChE, Aâ, andTHC, and Fo is the fluorescence of AChE and Aâ, was usedto quantify the extent to which each compound inhibits Aâaggregation. The Student’s t-test function of Microsoft Excelwas used to determine p values and assess statisticalsignificance between reactions.Control experiments containing AChE, THC, and thioflavin T or AChE and thioflavin T alone were also performedto ensure that any observed fluorescence decrease was notattributable to the molecular rotor properties of thioflavin Tupon binding to AChE. For these reactions, all concentrationswere identical to those used in the described Aâ aggregationassays (vide supra).Results and DiscussionTHC binding to AChE initially was modeled in silico usingAutoDock 3.0.5.21Twenty docking runs with 100 millionenergy evaluations each were performed with a 26.25 Å 18.75 Å 26.25 Å grid box with 0.375 Å grid spacing,which included regions of both the catalytic site and the PAS.Examination of the docking results revealed that THC was(21) Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.; Hart,W. E.; Belew, R. K.; Olson, A. J. Automated docking using aLamarckian genetic algorithm and an empirical binding freeenergy function. J. Comput. Chem. 1998, 19, 1639-1662.(22) Ellman, G. L.; Courtney, K. D.; Andres, Jr., V.; Featherstone, R.M. A new and rapid colorimetric determination of acetylcholinesterase activity. Biochem. Pharmacol. 1961, 7, 88-95.(23) LeVine, H., III. Thioflavine T interaction with synthetic Alzheimer’s disease â-amyloid peptides: detection of amyloidaggregation in solution. Protein Sci. 1993, 2, 404-410.Figure 2. Predicted binding mode of THC (gray) to AChE(orange ribbon). The catalytic triad residues of AChE (green)and water molecules included in the docking calculations (lightblue spheres) are shown.Marijuana and Alzheimer’s Disease Pathology brief articlesVOL. XXXX NO. XXXX MOLECULAR PHARMACEUTICS Cpredicted to bind to AChE with comparable affinity to thebest reported PAS binders, with the primary binding interaction observed between the ABC fused ring of the THCscaffold and the Trp86 indole side chain of AChE (Figure2). Further interactions were also evident between THC andthe backbone carbonyls of Phe123 and Ser125. Encouragedby these results, we tested the ability of THC to inhibit AChEcatalytic activity. Steady-state kinetic analysis of THCinhibition revealed that THC competitively inhibits AChE(Ki ) 10.2 íM) (Figure 3A). This level of inhibition isrelatively modest, yet it is important to note that inhibitionof acetylcholine cleavage is not a prerequisite for effectivereduction of Aâ aggregation; indeed, most PAS binders aremoderate AChE inhibitors displaying either noncompetitiveor mixed-type inhibition.16While THC shows competitiveinhibition relative to the substrate, this does not necessitatea direct interaction between THC and the AChE active site.In fact, given the proximity of the PAS to the protein channelleading to the catalytic triad active site, it is possible to blocksubstrate entry into the active site while bound to the PAS,thus preventing the formation of an ESI complex.18,24In orderto test this hypothesis, additional kinetic experiments wereperformed to determine the mutual exclusivity of THC andpropidium, a well-characterized purely noncompetitive AChEinhibitor and PAS binder. Dixon plots of V-1versus THCconcentration at different fixed concentrations of propidiumreturned a series of parallel lines, indicating that THC andpropidium cannot bind simultaneously to AChE (Figure 3B).Thus, these studies verify our docking results and demonstrate that THC and propidium are mutually exclusive PASinhibitors. Additionally, recent reports have suggested thatthe selectivity of a given inhibitor for AChE over butyrylcholinesterase (BuChE) can be correlated with the ability ofa compound to block AChE-accelerated Aâ aggregation.25,26Kinetic examination of BuChE inhibition revealed a slightreduction in enzymatic activity at high concentrations of THC(IC50 g 100 íM); however, these experiments were limitedby the poor solubility of THC in aqueous solution.The activity of THC toward the inhibition of Aâ aggregation was then investigated using a thioflavin T (ThT) basedfluorometric assay to stain putative Aâ fibrils.23Using thisassay, we found that THC is an effective inhibitor of theamyloidogenic effect of AChE (Figure 4). In fact, at aconcentration of 50 íM, propidium does not fully preventAChE-induced aggregation (p ) 0.03, Student’s t test), whileTHC completely blocks the AChE effect on Aâ aggregation,with significantly greater inhibition than propidium (p )0.04, Student’s t test), one of the most effective aggregationinhibitors reported to date.16However, the observed decreasein fluorescence could also be rationalized as a result of acompetition between THC and ThT for the same site onAChE. It has been shown that ThT also can bind to the PASand that this binding leads to an increase in fluorescence.Presumably, this phenomenon results from ThT serving asa molecular rotor in which fluorescence quantum yield issensitive to the intrinsic rotational relaxation; thus, whenmolecular rotation is slowed by protein binding, the quantum(24) Szegletes, T.; Mallender, W. D.; Rosenberry, T. L. Nonequilibriumanalysis alters the mechanistic interpretation of inhibition ofacetylcholinesterase by peripheral site ligands. Biochemistry 1998,37, 4206-4216.(25) Piazzi, L.; Rampa, A.; Bisi, A.; Gobbi, S.; Belluti, F.; Cavalli,A.; Bartolini, M.; Andrisano, V.; Valenti, P.; Recanatini, M. 3-(4-{[Benzyl(methyl)amino]methyl}-phenyl)-6,7-dimethoxy-2H-2-chromenone (AP2238) inhibits both acetylcholinesterase andacetylcholinesterase-induced â-amyloid aggregation: A dualfunction lead for Alzheimer’s disease therapy. J. Med. Chem.2003, 46, 2279-2282.(26) Belluti, F.; Rampa, A.; Piazzi, L.; Bisi, A.; Gobbi, S.; Bartolini,M.; Andrisano, V.; Cavalli, A.; Recanatini, M.; Valenti, P.Cholinesterase inhibitors: Xanthostigmine derivatives blockingthe acetylcholinesterase-induced â-amyloid aggregation. J. Med.Chem. 2005, 48, 4444-4456.Figure 3. (A) Kinetic analysis of AChE inhibition by THC: 0 (b), 6.25 (2), 12.5 ([), and 25.0 íM (9). Steady-state kineticanalysis was performed using acetylthiocholine (75-300 íM) and Ellman’s reagent (340 íM) at 37 °C. (B) Dixon plots of 1/vversus [THC] at different fixed concentrations of propidium iodide: 0 (b), 6.25 (2), 12.5 ([), and 25 íM (9).brief articles Eubanks et al.D MOLECULAR PHARMACEUTICS VOL. XXXX NO. XXXXyield of the molecule can increase dramatically.27,28In orderto ensure that the observed fluorescence decrease was dueto fibril inhibition, control experiments were performed usingAChE, THC, and ThT. Reactions containing AChE and ThTalone showed the same fluorescence output as those containing AChE, THC, and ThT, providing convincing evidencethat any observed reduction in fluorescence can be attributedto fewer Aâ fibrils.ConclusionWe have demonstrated that THC competitively inhibitsAChE and, furthermore, binds to the AChE PAS anddiminishes Aâ aggregation. In contrast to previous studiesaimed at utilizing cannabinoids in Alzheimer’s diseasetherapy,8-10our results provide a mechanism whereby theTHC molecule can directly impact Alzheimer’s diseasepathology. We note that while THC provides an interestingAlzheimer’s disease drug lead, it is a psychoactive compoundwith strong affinity for endogenous cannabinoid receptors.It is noteworthy that THC is a considerably more effectiveinhibitor of AChE-induced Aâ deposition than the approveddrugs for Alzheimer’s disease treatment, donepezil andtacrine, which reduced Aâ aggregation by only 22% and 7%,respectively, at twice the concentration used in our studies.7Therefore, AChE inhibitors such as THC and its analoguesmay provide an improved therapeutic for Alzheimer’sdisease, augmenting acetylcholine levels by preventingneurotransmitter degradation and reducing Aâ aggregation,thereby simultaneously treating both the symptoms andprogression of Alzheimer’s disease.Acknowledgment. This work was supported by theSkaggs Institute for Chemical Biology and a NIH KirschsteinNational Research Service Award to L.M.E.MP060066M(27) De Ferrari, G. V.; Mallender, W. D.; Inestrosa, N. C.; Rosenberry,T. L. Thioflavin T is a fluorescent probe of the acetylcholinesteraseperipheral site that reveals conformational interactions betweenthe peripheral and acylation sites. J. Biol. Chem. 2001, 276,23282-23287.(28) Viriot, M. L.; Carre, M. C.; Geoffroy-Chapotot, C.; Brembilla,A.; Muller, S.; Stoltz, J. F. Molecular rotors as fluorescent probesfor biological studies. Clin. Hemorheol. Microcirc. 1998, 19, 151-160.Figure 4. Inhibition of AChE-induced Aâ aggregation by THCand propidium ((*) p < 0.05 versus Aâ only; (#) p < 0.05versus Aâ + propidium).Marijuana and Alzheimer’s Disease Pathology brief articlesPAGE EST: 4.6 VOL. XXXX NO. XXXX MOLECULAR PHARMACEUTICS E