The FASEB Journal article fj.07-101865. Published online May 20, 2008. The FASEB Journal • Research Communication

Incensole , an incense component, elicits psychoactivity by activating TRPV3 channels in the brain

Arieh Moussaieff,*,†,1 Neta Rimmerman,‡ Tatiana Bregman,§ Alex Straiker,‡ ʈ Christian C. Felder, Shai Shoham,¶ Yoel Kashman,# Susan M. Huang,** Hyosang Lee,** Esther Shohami,† Ken Mackie,‡ Michael J. Caterina,** J. Michael Walker,‡ Ester Fride,§ and Raphael Mechoulam*,1 *Department of Medicinal Chemistry and Natural Products, Medical Faculty, and †Department of Pharmacology, Pharmacy School, Hebrew University of Jerusalem, Jerusalem, Israel; ‡Gill Center for Biomolecular Science and Department of Psychological and Brain Sciences, Indiana University Bloomington, Bloomington, Indiana, USA; §Departments of Behavioral Sciences and of Molecular ʈ Biology, Ariel University Center of Samaria, Ariel, Israel; Eli Lilly and Co., Neuroscience Division, Indianapolis, Indiana, USA; ¶Research Department, Herzog Memorial Hospital, Jerusalem, Israel; #School of Chemistry, Raymond and Beverly Sackler Faculty of Exact Sciences, Tel-Aviv University, Ramat Aviv, Israel; and **Department of Biological Chemistry, Johns Hopkins School of Medicine, Baltimore, Maryland, USA

ABSTRACT Burning of Boswellia resin as incense has Research on plants eliciting psychoactive effects, been part of religious and cultural ceremonies for such as Cannabis sativa, Papaver species, and Nicotiana millennia and is believed to contribute to the spiritual tabacum, has provided important insights into neuro- exaltation associated with such events. Transient recep- chemical processes and diseases of the central ner- tor potential vanilloid (TRPV) 3 is an ion channel vous system (CNS) (1). Psychoactive plant drugs have implicated in the perception of warmth in the skin. also played a major role in religious customs in many TRPV3 mRNA has also been found in neurons through- ancient cultures, as they exert a profound effect on out the brain; however, the role of TRPV3 channels human consciousness, emotions and cognition. No- there remains unknown. Here we show that incensole table examples of plants that were used in religious acetate (IA), a Boswellia resin constituent, is a potent rituals are the mythological Aryan soma, Ipomoea TRPV3 agonist that causes anxiolytic-like and antide- linnaeus (the source of South American ololiuqui), pressive-like behavioral effects in wild-type (WT) mice Cannabis sativa, Salvia divinorum (the source of divi- with concomitant changes in c-Fos activation in the norin) (2), Nicotiana tabacum (3), and several Bo- brain. These behavioral effects were not noted in swellia species. ؊ ؊ TRPV3 / mice, suggesting that they are mediated via The resin of Boswellia species (Burseraceae; “frank- TRPV3 channels. IA activated TRPV3 channels stably incense” and “olibanum”) is mentioned in numerous expressed in HEK293 cells and in keratinocytes from ancient texts as incense by itself or as a major ؉ ؉ TRPV3 / mice. It had no effect on keratinocytes component of incense (4). In the ancient Middle ؊ ؊ from TRPV3 / mice and showed modest or no effect East, Boswellia resin was considered a highly precious on TRPV1, TRPV2, and TRPV4, as well as on 24 other commodity, carried in caravans from sub-Sahara re- receptors, ion channels, and transport proteins. Our gions, where it is still a major export product (5). In results imply that TRPV3 channels in the brain may play ancient Egypt, incense burning signified a manifes- a role in emotional regulation. Furthermore, the bio- tation of the presence of the gods and a gratification chemical and pharmacological effects of IA may pro- to them. In ancient Judea, it was a central ceremony vide a biological basis for deeply rooted cultural and in the temple. The ancient Greeks used incense religious traditions.—Moussaieff, A., Rimmerman, N., burning as an oblation. In Christendom, its use in Bregman, T., Straiker, A., Felder, C. C., Shoham, S., worship has continued since the fourth or fifth Kashman, Y., Huang, S. M., Lee, H., Shohami, E., century C.E (6). The psychoactivity of Boswellia was already recognized in ancient times. Dioscorides Mackie, K., Caterina, M. J., Walker, J. M., Fride, E., Mechoulam, R. Incensole acetate, an incense compo- nent, elicits psychoactivity by activating TRPV3 chan- 1 Correspondence: Department of Medicinal Chemistry nels in the brain. FASEB J. 22, 000–000 (2008) and Natural Products, Medical Faculty, Hebrew University of Jerusalem, 91120, Jerusalem, Israel. E-mail: A.M., [email protected]; or R.M., [email protected] Key Words: Boswellia ⅐ frankincense ⅐ depression ⅐ anxiety doi: 10.1096/fj.07-101865

0892-6638/08/0022-0001 © FASEB 1 (first century C.E.) writes that it causes madness (7). ethoxydiphenyl borate (2-APB; ref. 19). Shortly after- In the Jewish Talmud (300–600 C.E.), Boswellia resin ward, the monoterpene was found to be an is mentioned as a potion (in wine) given to prisoners agonist (20). In addition, , , and euge- condemned to death to “benumb the senses” (8). In nol have recently been described to activate TRPV3 Ethiopia, where Boswellia trees are indigenous, it is (21). The function of TRPV3 in the CNS remains to be believed to have a tranquilizing effect (9). For addi- elucidated. tional data on the cultural and religious customs associated with Boswellia, see Supplemental Material. In view of the prolonged use of Boswellia, its MATERIALS AND METHODS historical importance, and its significance in cultural and religious rituals along with its purported phar- Drugs and reagents macological effects, we investigated the biochemical profile and psychoactivity of purified components of Boswellia carterii resin was purchased from Pamir (Tel Aviv, Boswellia resin. We assumed that the spiritual exalta- Israel). For the extraction and isolation of IA, we followed the procedure described by us previously (11). The material ob- tion caused by incense burning in religious ceremo- tained was identified by direct comparisons [of mass spectrom- nies would be enhanced by putative pharmacological etry (MS) and nuclear magnetic resonance (NMR) data] with effects of its constituents, particularly on the conduc- the compound used previously. Diazepam was purchased from tors of the ceremonies, who presumably inhale large Elkins-Sinn (Cherry Hill, NJ, USA). Desipramine hydrochloride, amounts of smoke. We are unaware of any attempt to , and camphor were purchased from Sigma-Aldrich (St. identify constituents with an effect on sensation or Louis, MO, USA). Fura-2 AM and pluronic 127 were purchased from Molecular Probes (Eugene, OR, USA), and 2-APB and emotion. Menon and Kar (10) have reported that an ionomycin were purchased from Calbiochem (EMD Bio- ether extract of Boswellia serrata resin produces anal- sciences, San Diego, CA, USA). 4␣-Phorbol 12,13-didecanoate gesic and sedative effects in rats, but the compounds (4␣PDD) was purchased from PKC Pharmaceuticals (Woburn, causing these effects have not been isolated. MA, USA). Pental was purchased from C.T.S. Chemical Indus- We examined Boswellia extract for the presence of tries (Tel Aviv, Israel). Rabbit anti c-Fos, horseradish peroxidase, novel bioactive components and isolated incensole and diaminobenzidine were purchased from Sigma. Secondary acetate (IA) as a major active constituent of Boswellia biotinylated donkey anti-rabbit antibody was purchased from Chemicon (Temecula, CA, USA). resin. IA showed an antiinflammatory action, as re- IA and 4␣PDD were dissolved in ethanol for in vitro assays; cently reported by us (11, 12), as well as several 2-APB, ionomycin, and capsaicin were dissolved in dimethyl CNS-associated activities, described below. It is known sulfoxide (DMSO). All solvents were applied in a volume as a macrocyclic diterpenoid (13), considered to be a Ͻ1.1% of the total volume in the well, a concentration that biomarker of Boswellia species (14), but has not been did not induce effects on calcium mobilization or ion cur- previously evaluated for its psychoactivity. rents. To evaluate the activity profile of IA, we used behav- Cell culture ioral, pharmacological, and physiological assays. They included standard behavioral mice models for the Human embryonic kidney (HEK) 293 cells stably express- evaluation of anxiolytic and antidepressive drugs (ele- ing human TRPV1 were a kind gift from Merck Research vated plus maze, Porsolt’s forced swimming test, open Laboratories (Whitehouse Station, NJ, USA). Cells were field behavior, and a test for catalepsy), c-Fos immuno- cultured in minimal essential medium, Dulbecco modified staining, various receptor binding assays, and physio- Eagle medium (DMEM), modified with nonessential amino logical measurements. acids, 1 mM sodium pyruvate, 2 mM l-glutamine, and 1.5 Transient receptor potential (TRP) channels are g/l sodium bicarbonate (TACK, Manassas, VA, USA), con- taining 1% penicillin-streptomycin and 10% FBS. Cells activated by multiple stimuli in expression systems and were passaged 3 times a week using trypsin-EDTA 1ϫ mediate transmembrane flux of cations (15). The TRP (Invitrogen, Carlsbad, CA, USA) and grown under 5% CO2 vanilloid (TRPV) channel subfamily consists of six at 37°C. members; TRPV1-TRPV4 comprise the so-called HEK293 cells stably expressing mouse TRPV3-YFP, thermo-TRPVs and are activated by temperature and pcDNA3, and rat TRPV4 were acquired as described previ- chemical stimuli. Although thermo-sensitive, TRPV ously (20, 22, 23). HEK293 cells were transiently transfected with a rat TRPV2 channels are also expressed in tissues such as the brain, plasmid (24), using lipofectamine reagent (Invitrogen) ac- in which thermoregulatory homeostasis precludes dra- cording to manufacturer’s protocol. HEK293 cells expressing matic temperature swings (15). mouse TRPV3-YFP, pcDNA3, rat TRPV2, or rat TRPV4 were TRPV3 has a threshold for temperatures in the cultured in 1ϫ DMEM (high glucose) with l-glutamine 31–39°C range, and it is highly expressed in epithelial (Mediatech, Inc., Herndon, VA, USA), containing 1% peni- cells of the skin and oropharynx (16–18), where its cillin-streptomycin (Invitrogen) and 10% FBS. Cells were passaged 3 times a week using 1ϫ trypsin-EDTA (Invitrogen) activation causes a feeling of warmth. mRNA encoding and grown under 5% CO at 37°C. this channel has also been detected in the brain (17, 2 18). Owing to the recent identification of TRPV3 ion Animals and procedures channel, relatively few agonists are known for it. The first chemical agonist identified for TRPV3 was the Female Sabra mice (Harlan, Rehovot, Israel; 15–20 wk old) synthetic broad spectrum TRP modulator 2-amino- and wild-type (WT) C57BL/6 or TRPV3 knockout (KO)

2 Vol. 22 August 2008 The FASEB Journal MOUSSAIEFF ET AL. female mice (18–20 wk old; ref. 18) were used for behavioral coronal sections, 30 ␮m thick, and were collected floating assessments. Ten mice were housed in each cage. The animal and kept in a cryopreservation buffer at Ϫ18°C until immu- care and protocols met the guidelines of the U.S. National nohistochemical staining of c-Fos. For each brain region, two Institutes of Health, detailed in the Guide for the Care and Use fields were sampled containing this region from the right and of Laboratory Animals, and were applied in conformity with the left hemisphere. After the sections were rinsed in PBS (0.02 Johns Hopkins University and Ariel University Center of M, pH 7.4) and quenched of endogenous peroxidase by Samaria Institutional Ethics Committees. Female Sabra mice incubation in 0.15% for 30 min, they were were also used for the c-Fos immunostaining. Temperature in incubated with rabbit anti c-Fos diluted 1:5000. The second- the animal room was maintained between 20–22°C; the light ary antibody was diluted 1:400, and sections were incubated cycle was 12 h lights on (8 AM to 8 PM) and 12 h lights off (8 with extravidin conjugated with horseradish peroxidase di- PM to 8 AM). Mice were injected intraperitoneally with IA in luted 1:200. The final color reaction was performed with a mixture of isopropanol:cremophor:saline (1:1:18) at a diaminobenzidine. volume of 10 ␮l/g body weight. Binding assays Behavioral assays All binding assays were performed under license agreement The different behavioral assays performed were all based on at Cerep (Paris, France; refs. 31, 32). established assays. Preliminary observations indicated similar effects of IA when the time interval between injection and Calcium imaging of HEK293 cells testing was between 15 and 45 min. Therefore, mice were assayed within these time limits. Human TRPV1-, mouse TRPV3-YFP-, rat TRPV4-, and Elevated plus maze pcDNA3-expressing HEK293 cells were plated 24–48 h before imaging in 96-well black walled, clear bottom Cell- Bind plates (Corning, Corning, NY, USA), loaded with 3 ␮M The elevated plus maze assay is based on the preference of Fura-2 AM, and imaged as described previously (33). For single- mice for the closed arms of a maze, apparently due to fear of cell calcium imaging, HEK293-rat TRPV2- and -mouse TRPV3- open spaces. The assay was performed as described by others YFP-expressing cells were plated on collagen-coated glass cover (25, 26). Briefly, mice were placed in the central platform slips. Cells were loaded for 60 min with 3 ␮M Fura-2 AM in (10ϫ10 cm) between the open (10ϫ45 cm) and closed ϫ ϫ 0.05% w/v pluronic 127 in HEPES-Tyrode buffer (pH 7.4) (10 45 40 cm) arms of a plus maze. The time spent in each containing the following (in mM): 25 HEPES, 140 NaCl, 2.7 KCl, of the arms was recorded. An “antianxiety” effect was mea- 1.8 CaCl2, 0.5 MgCl2, 0.4 NaH2PO4, and 5 glucose. All experi- sured as the time spent on the aversive open arms of the maze ments were performed at room temperature (ϳ24°C). A Flex- (relative to the total time spent in both arms). station II plate reader (Molecular Devices) was programmed with the following settings: excitation wavelength, 340 and 380 Porsolt forced swimming test nm; emission wavelength, 510 nm. The initial volume per well was 175 ␮l. After 30 s of baseline recording, 75 ␮l buffer The Porsolt forced swimming test was based on previous containing drug or vehicle was added at a rate of up to ϳ52 ␮l/s designs (27, 28): mice were placed ina2Lglass beaker (11 per well. The total run time was 200 s. For single-cell calcium cm diameter) filled with water (24Ϯ1°C) up to 30 cm from imaging experiments, coverslips were mounted on an inverted the bottom (so that the mouse could not touch the bottom) microscope (Nikon TS-100, Tokyo, Japan). Cells were alternately and 8 cm from the rim (so that the mouse could not escape). excited at 340 and 380 nm. Emitted light (510 nm) was captured Immobility time was defined as the time the animal did not using a Cohu 4920 cooled charge-coupled device (CCD) camera move (except for small movements required to float). (Cohu, San Diego, CA, USA) and analyzed with the InCyt Im2 image acquisition and analysis software (Intracellular Imaging Open field behavior Inc., Cincinnati, OH, USA).

؉ ؉ ؊ ؊ Mice were placed in a transparent open field of 30 ϫ 40 cm, Calcium imaging of TRPV3 / and TRPV3 / divided into 20 squares of equal size, and their horizontal keratinocytes (ambulatory) and vertical activity was measured for 8 min by scoring the number of squares crossed (29, 30) and number Primary keratinocytes from TRPV3-deficient and TRPV3ϩ/ϩ of rears accordingly. mouse pups (days 1–4) were harvested and cultured as described previously (34). Cells were plated on glass cover- Cataleptic effect slips (105/cm2) and incubated for 48–60 h and then loaded with Fura-2 AM (20 ␮M, 0.04% pleuronic acid, 32°C for 1 h) The cataleptic effect of IA was measured as time of immobility in imaging buffer containing the following (in mM): 130 on an elevated ring of 5.5 cm diameter (during 4 min; refs. NaCl, 2.5 CaCl2, 0.6 MgCl2, 10 HEPES, 1.2 NaHCO3, and 10 2ϩ 29, 30). glucose, pH 7.45. Ratiometric Ca imaging was performed as described previously (22). Drug was added to the bath after a c-Fos immunostaining period of baseline recording. Calcium measurements were made from 30 randomly selected cells per coverslip (the cells were not chosen based on criteria such as morphology or Mice were deeply anesthetized by an intraperitoneal injection labeling). of 200 mg/kg sodium pentobarbital (Pental). Brains were fixed by transcardial perfusion with ice-cold 4% paraformal- dehyde containing 4% sucrose (pH 7.4) and by overnight Electrophysiological recording immersion in the same, refrigerated fixative. Brains were then immersed in 12% sucrose in PBS (0.02 M, pH 7.4) and kept Currents were recorded using whole-cell voltage clamp. Pi- refrigerated until sectioning. Brains were cut on a cryostat as pettes were pulled from microcapillary glass (A-M Systems,

INCENSOLE ACETATE FROM INCENSE IS PSYCHOACTIVE 3 Sequim, WA, USA). A coverslip containing cells was trans- ferred to a 300 ␮l chamber that was constantly perfused (1–2 ml/min) with external solution. Voltage protocols were gen- erated and data were digitized and recorded using Pulse (HEKA Elektronik, Lambrecht, Germany) software in con- junction with an Axopatch 200A amplifier (Axon Instru- ments, Foster City, CA, USA). The data were analyzed using an in-house Visual Basic (Microsoft, Redmond, WA, USA) analysis program. The pipette solution contained the following (in mM):

121.5 K gluconate, 10 HEPES, 17 KCl, 9 NaCl, 1 MgCl2, 0.2 EGTA, 2 MgATP, and 0.5 NaATP, pH 7.2. The external solution contained the following (in mM): 120 NaCl, 5 KCl,

1 MgCl2, 2 CaCl2, 10 glucose, and 20 HEPES, pH 7.4, with NaOH. The measured charge (pC) was defined as the charge elicited between Ϫ85 and Ϫ45 mV by a ramping voltage stimulus (Ϫ85 mV to ϩ35 mV, 0.54 mV/ms; holding potential Ϫ55 mV). Currents were sampled at 5 kHz. Experimental and control cells were alternated when- ever possible. Control values were obtained from WT HEK293 cells.

Data analysis

In general, for comparisons between two data sets we used an unpaired 2-tailed Student’s t test. For tests of statistical significance involving more than two data sets, we used a Figure 1. Structure of IA. Structure elucidation is based on 1-way ANOVA with a post hoc test, because this test is NMR and MS data (see ref. 14). appropriate for comparing three groups on one factor. Data obtained from the KO vs. WT mice in Fig. 7 were analyzed by 2-way ANOVA because in this case two factors potent anxiolytic-like effect, causing mice to spend were investigated (strain: WT vs. KO; and drug: vehicle vs. significantly more time in the aversive open arms of the IA). Behavioral responses in Sabra mice to IA vs. diazepam elevated plus maze (Fig. 2A). In the Porsolt forced or desipramine were analyzed by 1-way ANOVA with New- swimming test, a standard assay for the evaluation of man-Keuls post hoc comparisons. Motor behavior and cata- antidepressant effects, IA significantly reduced the im- lepsy data were analyzed by t tests (comparing vehicle- and IA-injected mice), and behaviors in C57BL/6 (WT) vs. mobility recorded over 9 min, thus indicating a reversal TRPV3Ϫ/Ϫ mice were analyzed using 2-way ANOVA with of an avolition response (Fig. 2B). We also observed a Bonferroni post hoc comparisons (GraphPad 4 Prism; significant reduction in open field behavior (Fig. 3A, GraphPad, San Diego, CA, USA). Analysis of calcium B), as well as increased immobility on a ring (Fig. 3C). imaging data was performed using a nonlinear regression The experiments described in Figs. 2 and 3 are curve fit (GraphPad 4 Prism). In the keratinocyte experi- representative of 4 to 7 independent experiments, and ments, drug-induced response for each cell was taken as all gave comparable results. It should be noted that the maximal postdrug measurements over time minus the average of the last 5 predrug measurements. Averaged drug while the experiments presented in Fig. 2 are designed responses over 30 randomly selected cells per coverslip to reflect the emotional state of the examined animal, were analyzed with 2-tailed unpaired t tests. the open field locomotion and the ring catalepsy tests, For analysis of c-Fos immunoreactivity, positive nuclei presented in Fig. 3, reflect the motor effects of IA. were identified based on their round form and optical density at least twice that of background. The numbers of IA modulates the expression of c-Fos in mice brains c-Fos immunoreactive nuclei from the right and left hemi- spheres were averaged to obtain a representative number in several areas, including areas involved in anxiety for the given region from each mouse. Student t tests were and depression performed comparing the control (vehicle) with the IA group. To investigate the activity of IA on different brain regions, we studied its effect on c-Fos immunoreactivity in mouse brains 60 min after administration (50 mg/kg RESULTS ip). IA significantly increased c-Fos in the lateral sep- tum, central nucleus of the amygdala, and solitary Behavioral assays nucleus, while significantly reducing c-Fos in the motor cortex, medial striatum, and hippocampal CA3 region To study the functional effects of IA on the CNS, we (Fig. 4; Table 1; Supplemental Fig. 1). first assayed IA (Fig. 1) in a panel of standard behav- ioral assays in mice (female Sabra strain, 15–20 wk old): Binding assays the elevated plus maze, the Porsolt forced swimming test, locomotion in the open field test, and cataleptic IA was assayed for its ability to displace radioligands response in a ring test. At 50 mg/kg, IA exerted a bound to an array of 24 receptors, ion channels, and

4 Vol. 22 August 2008 The FASEB Journal MOUSSAIEFF ET AL. Figure 2. Effect of IA on performance in the elevated plus maze and the Porsolt forced swimming test. Female Sabra mice (15-20 wk) were injected with vehicle isopropanol: emulphor (a commercial emulsifier):salineϭ1:1:18 or IA (50 mg/kg) in vehicle. Diazepam (5 mg/kg, dissolved in ethanol: ϭ Figure 4. IA modulates stress-induced c-Fos expression in emulphor:saline 1:1:18) was used as positive control for the several brain areas. Diagrams illustrate brain areas of plus maze; desipramine (DMI; 15 mg/kg, dissolved in saline) female Sabra mice (15–20 wk; nϭ4–5) where IA (50 was used as a positive control in the Porsolt forced swimming mg/kg) significantly changed the number of c-Fos-immu- test. Thirty to 40 min later, the mice were tested in the noreactive cells, 60 min after intraperitoneal injection of elevated plus maze for 5 min (A) and the Porsolt forced IA or vehicle. Drawings were modified from Paxinos and swimming test for 9 min (B). Data are means Ϯ se; n ϭ 4–6. Ͻ Ͻ Franklin (49), Plates 30 (A), 38 (B), 45 (C), and 89 (D). *P 0.05, **P 0.01 vs. WT vehicle-injected mice (Newman- Atlas sections are arranged from anterior (A) to posterior Keuls post hoc test). (D). Number under each section indicates its distance (mm) anterior (A) or posterior (P) from the bregma. IA significantly increased c-Fos (red fill) in the lateral septum transport proteins potentially relevant to its putative (LS), central nucleus of the amygdala (CEA), solitary emotional and behavioral effects (Supplemental Table complex (Sol), and bed nucleus stria terminalis (BNST). 1). No significant displacement of radioligand was IA significantly reduced c-Fos (light blue fill) in the motor observed for any of these targets using up to 10 ␮Mof cortex (MCtx), medial striatum (MSt), and hippocampal IA (data not shown). CA3 region (CA3). See Table 1 for quantification and Supplemental Fig. 1A, B for representative micrographs.

Figure 3. Effects of IA on motor activity and catalepsy. Female Sabra mice were injected intraperitoneally with 50 mg/kg IA dissolved in a mixture of isopropanol:cremophor:saline ϭ 1:1:18. Fifteen minutes later, the mice were observed for locomotion and rearing in an open field (30ϫ40 cm, divided into 20 squares of equal size) and for catalepsy assay on a ring (7.5 cm diameter) elevated 30 cm above the table top. Data are means Ϯ se; n ϭ 4–5. *P Ͻ 0.05; ***P Ͻ 0.001 (unpaired t test).

INCENSOLE ACETATE FROM INCENSE IS PSYCHOACTIVE 5 TABLE 1. IA modulates stress-induced c-Fos expression in several brain areas

Region Subregion/nucleus Vehicle (nϭ4) IA (nϭ5) t df P

Cortex Cingulate 29.875 Ϯ 16.265 38.2 Ϯ 20.825 Ϫ0.653 7 0.5346 Motor 12.125 Ϯ 5.36 3.7 Ϯ 3.2711 2.9258 7 0.0222* Somatic-sensory 2.875 Ϯ 4.0078 1.5 Ϯ 1.4142 0.7235 7 0.4928 Auditory 11.875 Ϯ 5.5434 45.7 Ϯ 47.228 Ϫ1.405 7 0.2028 Olfactory 20.625 Ϯ 10.291 17.000 Ϯ 7.608 0.610 7.000 0.561 Perirhinal 11.250 Ϯ 8.261 17.900 Ϯ 9.317 Ϫ1.116 7.000 0.301 Dorsal endopiriform 19.500 Ϯ 7.895 22.100 Ϯ 12.803 -0.353 7.000 0.734 Hippocampus CA1_2 1.375 Ϯ 0.750 0.600 Ϯ 0.822 1.459 7.000 0.188 CA3 13.500 Ϯ 2.483 4.700 Ϯ 3.213 4.488 7.000 0.003* Dentate gyrus 12.500 Ϯ 2.449 17.000 Ϯ 10.840 Ϫ0.803 7.000 0.448 Subiculum 0.750 Ϯ 0.957 0.300 Ϯ 0.671 0.832 7.000 0.433 Amygdala Central nucleus 17.500 Ϯ 10.400 52.100 Ϯ 15.433 Ϫ3.818 7.000 0.007* Basolateral nucleus 35.500 Ϯ 16.176 33.800 Ϯ 22.298 0.127 7.000 0.902 Corticomedial nucleus 31.500 Ϯ 19.188 41.300 Ϯ 20.216 Ϫ0.738 7.000 0.484 Bed nucleus stria terminalis Dorsal 8.500 Ϯ 1.061 20.100 Ϯ 13.081 Ϫ2.101 7.000 0.070 Ventral 19.000 Ϯ 19.092 39.400 Ϯ 31.466 Ϫ1.700 7.000 0.125 Septum Lateral 41.875 Ϯ 24.216 85.200 Ϯ 25.871 Ϫ2.565 7.000 0.037* Striatum Lateral 1.250 Ϯ 1.658 1.200 Ϯ 1.956 0.041 7.000 0.969 Medial 14.125 Ϯ 4.029 5.200 Ϯ 3.581 3.520 7.000 0.010* Thalamus Paraventricular 24.500 Ϯ 2.972 35.400 Ϯ 11.459 Ϫ1.830 7.000 0.110 Intermediodorsal 7.750 Ϯ 10.836 10.900 Ϯ 6.533 Ϫ0.543 7.000 0.604 Lateral geniculate 18.875 Ϯ 5.313 21.400 Ϯ 14.762 Ϫ0.322 7.000 0.757 Hypothalamus Lateral 34.125 Ϯ 4.854 41.600 Ϯ 24.605 Ϫ0.591 7.000 0.573 Ventromedial 10.625 Ϯ 7.146 15.500 Ϯ 2.669 Ϫ1.426 7.000 0.197 Paraventricular 16.625 Ϯ 11.884 45.400 Ϯ 26.409 Ϫ2.002 7.000 0.085 Anterior 20.750 Ϯ 10.603 26.000 Ϯ 15.096 Ϫ0.586 7.000 0.576 Mediodorsal 44.375 Ϯ 8.004 38.100 Ϯ 11.589 0.916 7.000 0.390 Brain stem Central gray 12.500 Ϯ 7.627 17.400 Ϯ 4.642 Ϫ1.197 7.000 0.270 Edinger Westphal 2.250 Ϯ 3.069 3.800 Ϯ 4.147 Ϫ0.621 7.000 0.555 Solitary nucleus 18.875 Ϯ 20.738 78.000 Ϯ 12.247 Ϫ4.910 6.000 0.003*

Results of Student t tests for the counts of Fos-immunoreactive cells in mice (female Sabra strain, aged 15–20 wk; nϭ4–5) in different brain areas 60 min after IA or vehicle was injected; values are means Ϯ sd. *Significant change; P Ͻ 0.05.

IA activates TRPV3 channels, but not several other only minimal calcium influx in HEK293 cells express- TRPV channels, as determined by calcium imaging ing human TRPV1 and rat TRPV4 (Fig. 5E, G).

Our attention then turned to the TRPV family of IA activates a TRPV3 current in HEK293 cells channels, especially as TRPV1 is assumed to mediate behavioral effects that resemble those exhibited by IA IA also activated a cation current in mouse-TRPV3-YFP- (35). We hence proceeded to examine the effect of IA expressing HEK293 cells (Fig. 6A–C) with properties on TRPV1–4. IA significantly increased calcium influx consistent with TRPV3 activation (36). This current was ϭ ␮ ϭ lower than the current activated by 2-APB, which served (EC50 16 M; Hill slope 2.2; Fig. 5A, B, D)in HEK293 cells stably expressing mouse TRPV3-YFP. as a positive control (Fig. 6D). IA-induced current was When calcium was removed from the extracellular not activated in HEK293 cells not expressing TRPV3 medium, the calcium increase in response to IA was nor in TRPV1 or TRPV4 expressing cells (Fig. 6C, E). significantly reduced (Fig. 5B), providing further evi- dence for the influx of calcium through TRPV3 chan- IA exerts anxiolytic-like and antidepressive-like nels. The effect of IA on TRPV3-induced calcium influx effects in WT but not in TRPV3 KO mice resembles the effect of the broad-spectrum TRP agonist 2-APB, which served as a positive control (Fig. 5A, D). Given the effect of IA on TRPV3 channels and the IA (500 ␮M) also induced a calcium influx in primary observation that IA does not interact with a long list Ϫ Ϫ keratinocytes from WT mice but not from TRPV3 / of receptors known to be involved in psychoactivity, mice (Fig. 5C). The effect of IA (500 ␮M) on TRPV- we investigated the possibility that its behavioral induced calcium influx in primary keratinocytes resem- effects are mediated through CNS TRPV3 channels. bles the one of camphor (10 mM), a known agonist of We repeated the panel of behavioral assays with WT Ϫ Ϫ TRPV3. Interestingly, at a concentration (100 ␮M) that and TRPV3 / mice, which were administered either was maximally effective in TRPV3 expressing cells, IA IA or vehicle. IA (75 mg/kg) exerted a significant Ϫ Ϫ did not induce calcium influx in HEK293 cells tran- anxiolytic-like effect in WT mice, while TRPV3 / siently transfected with rat-TRPV2 (Fig. 5F) and caused mice spent identical time on the open arms, regardless

6 Vol. 22 August 2008 The FASEB Journal MOUSSAIEFF ET AL. Figure 5. IA is a potent TRPV3 activator. A) IA or 2-APB evoked calcium increases in mouse HEK293 TRPV3-YFP-transfected cells compared with vehicle; #P Ͻ 0.001 (nϭ9). IA-treated HEK293-TRPV3(ϩ) cells show a significantly higher activation than HEK293-pcDNA cells; *P Ͻ 0.001 (nϭ9). B) IA dose dependently induced calcium influx in TRPV3-YFP-transfected HEK293 cells in the presence of calcium in the extracellular media (E); EC50 ϭ 16 ␮M, Hill slope ϭ 2.2 (nϭ10). In the absence of calcium (⅜), the effect of IA was markedly reduced; #P Ͻ 0.05 (nϭ5). C) IA (500 ␮M) increased intracellular calcium levels in primary keratinocytes from TRPV3ϩ/ϩ but not TRPV3Ϫ/Ϫ mice. Camphor (10 mM) showed a similar effect. *,#P Ͻ 0.005 (nϭ6), t test (2-tailed). D) Representative single-cell calcium traces of HEK293 cells stably expressing mouse-TRPV3-YFP. IA or 2-APB was applied at 60 s (after 60 s of baseline recording; arrow); the drug remained in the bath throughout the recording. E) IA induced a very small influx of calcium in human TRPV1-transfected HEK293 cells compared with vehicle; # P Ͻ 0.001, (nϭ22–29). Capsaicin induced a calcium increase significantly greater than that induced by IA; *P Ͻ 0.001 (nϭ29–35). F) IA did not induce calcium influx in HEK293 cells transiently transfected with rat-TRPV2, whereas 2-APB increased calcium in these cells; #P Ͻ 0.001 (nϭ41–51). G) IA induced a very modest calcium influx in rat-TRPV4-transfected HEK293 cells compared with vehicle; #P Ͻ 0.001 (nϭ26). 4␣PDD induced a calcium increase that was significantly larger than the effect of IA; * P Ͻ 0.001 (nϭ26). All error bars indicate se; P values in all subfigures but C represent analysis with 1-way ANOVA with Bonferroni’s post hoc test.

INCENSOLE ACETATE FROM INCENSE IS PSYCHOACTIVE 7 Figure 6. IA activates a TRPV3 current when it is stably expressed in HEK293 cells. A) Sample time course shows summed charge of current activated (Ϫ85 to Ϫ45 mV, in pC) with application of IA (200 ␮M). B) Sample current response to voltage ramp from same cell as A. C) Dose response for IA shows activation of currents at 200 ␮M in TRPV3(ϩ) HEK293 cells (f) but not in TRPV3(Ϫ) controls (Œ). *P Ͻ 0.001 vs. TRPV3(Ϫ); 1-way ANOVA with Dunnett’s post hoc test. D) TRPV agonist 2-APB (100 ␮M) activates currents in TRPV3(ϩ) cells but not in TRPV3(Ϫ) cells. **P Ͻ 0.001; unpaired 2-tailed Student’s t test. E) IA (200 ␮M) does not activate currents in TRPV1(ϩ) and TRPV4(ϩ) cells nor does vehicle in TRPV3(ϩ) cells. TRPV3(ϩ) response to IA is shown for reference. Two TRPV3(Ϫ) controls were used, one set in WT HEK293 cells and one set in nonglowing cells found among the TRPV3-YFP-expressing HEK293 cells; the results were identical. *P Ͻ 0.001 vs. TRPV3(ϩ); 1-way ANOVA with Dunnett’s post hoc test. Error bars indicate se; n ϭ 4–5. of whether they were injected with IA or only vehicle many cases, due to treatment-limiting adverse effects (40, ϭ ϭ Ͻ ϭ (Fig. 7A; Fstrain 6.3, df 1,14, P 0.05; Finteraction 5.0, 41). Finding new drugs for the amelioration of anxiety dfϭ1,14, PϽ0.05). In the Porsolt forced swimming test, and depression is thus of considerable significance and IA significantly reduced the immobility time in WT but even more so is the search for new pharmacological Ϫ/Ϫ ϭ ϭ Ͻ not in TRPV3 mice (FIA 5.5, df 1,16, P 0.04; targets that can be addressed in this quest. ϭ ϭ Ͻ Finteraction 5.9, df 1,16, P 0.03; Fig. 7B). No signifi- c-Fos transcription factor is a product of an immediate cant differences were recorded between vehicle-treated Ϫ Ϫ early gene, and changes in its expression serve as markers WT mice and vehicle-treated TRPV3 / mice in the of changes in neuronal activity. It is used in histological Porsolt forced swimming and elevated plus maze assays. sections to map out brain regions that are activated or The decreased locomotion and cataleptic effects of IA attenuated after treatment with psychoactive drugs, such Ϫ/Ϫ were maintained in TRPV3 mice (Fig. 8). as anxiolytics and antidepressants (42, 43). The central nucleus of the amygdala and the lateral septum play major roles in the expression of emotions (44, 45), and it DISCUSSION is assumed that c-Fos expression in the central nucleus of the amygdala is due to circuits that are engaged by both The quest for the discovery of active principles present anxiolytic and anxiogenic drugs (44). in psychoactive plants is based mainly on the expecta- We found that IA shows significant TRPV3-dependent tion of uncovering through them novel biochemical activity in both the elevated plus maze and the Porsolt pathways in the CNS, as well as novel molecular leads to forced swimming test. Despite minor differences, which drugs for the treatment of mental disorders. Indeed, are expected in different strains, IA maintains significant the development of new and potent psychoactive drugs, behavioral effects on both Sabra mice and WT C57BL/6 in particular plant-derived natural products, especially mice, generalizing and corroborating its putative psycho- during the 1960s and the 1970s, was a cornerstone in active effect. It should be noted that these behavioral the evolving science of psychopharmacology (2). assays are independent from the locomotion open field test. Hence, despite the significant attenuation of locomo- IA as a novel anxiolytic and antidepressive agent tion, IA significantly decreased the time mice spend immobile in water. IA also modulates the expression of Anxiety and depression are the most common forms of c-Fos in mice brains in several areas, including areas psychiatric disorders in the United States (37, 38) and involved in anxiety and depression (Fig. 4). Thus, the data pose risk factors for numerous additional diseases (39). from the behavioral assays together with the c-Fos immu- The clinical treatment of these high-prevalence disorders nostaining suggest that IA may represent a novel anxio- often relies on pharmacologic interventions. However, lytic and antidepressive agent. Our next step was to look although antianxiety and antidepressive drugs have been for a possible mechanism for these behavioral effects. developed, new types of such medicaments are needed, as Considering its effects on animal models, we were sur- the presently available treatments are not satisfactory in prised to note that IA does not bind to any of 24 related

8 Vol. 22 August 2008 The FASEB Journal MOUSSAIEFF ET AL. receptors, ion channels, and transport proteins. However, we found that it is a potent TRPV3 channel activator.

IA is a potent TRPV3 activator

In view of the critical role played by natural products in the discovery of the TRP channels and their functions (46), several groups have recently attempted to identify natural products that activate TRPV3 channel. Smith et al. (47) have screened extracts from 50 Chinese herbal plants; no TRPV3 active compounds were identified. Vogt-Eisele et al. (48) examined a library of monoter- penoids for their action on TRPV3 and found some of ␮ them to be active, the lowest EC50 reported as 370 M. As described above (Fig. 5A, B, D), we found that IA induces calcium influx in HEK293 cells expressing TRPV3 channel. This observation was confirmed by the activation of TRPV3 in primary mouse keratinocytes (Fig. 5C) and by the activation of a TRPV3 current in HEK293 cells stably expressing TRPV3 (Fig. 6A–D). The effect of IA on TRPV3-induced calcium influx resem- bles the effect of 2-APB, the most potent known TRPV3 activator (Fig. 5D). However, its effect on cell current is lower (Fig. 6). The effect elicited by IA on TRPV3 Figure 7. IA exhibits an anxiolytic effect in the elevated plus channels seems to be more specific than that by 2-APB, maze test and an antidepressant-like effect in the Porsolt as it showed no or very modest activity on cells express- forced swimming test in WT but not TRPV3Ϫ/Ϫ mice. WT ing TRPV2, TRPV1, or TRPV4 (Figs. 5E–G and 6E). and TRPV3Ϫ/Ϫ mice (18–20 wk old) were injected with ϭ Because the charge response to IA lags the calcium vehicle (isopropanol:emulphor:saline 1:1:18) or IA (75 mg/ response, it is possible that a calcium-activated current kg). Thirty to forty min later, the mice were tested in the elevated plus maze for 5 min (A) and the Porsolt forced is evoked in the cells. However, this difference can be swimming test for 9 min (B). In the elevated plus maze assay, attributed to the inherent differences between the IA caused WT mice to spend significantly more time in the calcium imaging system and an electrophysiology appa- aversive open arms of the maze (relative to the total time ratus, e.g., differences in the ions measured or in drug spent in both arms). In the Porsolt forced swimming test, IA application (manual application vs. perfusion system). significantly reduced immobility in WT mice, whereas TRPV3 Terpenoids have been used for centuries as medici- KO mice did not respond to IA. No significant difference was noted in WT and TRPV3 KO mice in response to vehicle. Data nally useful compounds, but still little is known about are means Ϯ se; n ϭ 4–5. *P Ͻ 0.05, **P Ͻ 0.01 vs. WT the mechanism of action of many of them (48). TRP vehicle-injected mice (Bonferroni post hoc test). channels as targets for terpenoids can potentially ex- plain several of their described effects. Investigations of novel ligands for members of the TRP channel family are not only relevant for the understanding of this

Figure 8. IA shows a similar effect on open field behavior in WT and TRPV3Ϫ/Ϫ mice. WT and TRPV3Ϫ/Ϫ mice (18–20 wk) were injected with vehicle (isopropanol:emulphor:salineϭ1:1:18) or IA (75 mg/kg). Fifteen min later, they were introduced into a transparent cage and tested for the effect of IA on locomotion for 8 min: horizontal, number of squares crossed (A); and vertical, number of rears (B). Mice were also tested for the cataleptic effect of IA as time of immobility on an elevated ring of 5.5 cm diameter over 4 min (C). No difference was noted in WT and TRPV3 KO mice in response to vehicle. Data are means Ϯ se; n ϭ 4–5. #P Ͻ 0.05, ***P Ͻ 0.001 vs. WT vehicle-injected mice (Bonferroni post hoc test).

INCENSOLE ACETATE FROM INCENSE IS PSYCHOACTIVE 9 protein family but may also have implications for of the Dr. L. and Dr. M. Deutch chair in psychopharmacology at target-directed drug design (48). Hebrew University.

TRPV3 channel may pose a possible novel therapeutic target in the CNS REFERENCES

1. O’Connor, K. A., and Roth, B. L. (2005) Screening the recep- Our data, in particular the observations that IA has no torome for plant-based psychoactive compounds. Life Sci. 78, effect in the elevated plus maze and the Porsolt forced 506–511 Ϫ Ϫ swimming test in TRPV3 / mice, indicate that the 2. Schultes, R. E., and Hofmann, A. (1980) The Botany and the Chemistry of Hallucinogens, 2nd Ed., Charles C. Thomas, Spring- TRPV3 channels are involved in the effects of IA in animal field, IL, USA models for antidepressant and anxiolytic effects. This 3. Elferink, J. G. (1983) The narcotic and hallucinogenic use of implies that TRPV3 channels in the brain may play a role tobacco in pre-Columbian Central America. J. Ethnopharmacol. 7, in emotional regulation, thus attributing, for the first 111–122 4. MacCulloch, J. A. (no date) Encyclopaedia of Religion and Ethics, Vol. time, a function for this channel in the CNS. These data 7 (Hastings, J., ed) p. 201, Charles Scribner’s Sons, New York, NY, also raise hopes that TRPV3 channels in the brain may be USA targeted in the efforts of ameliorating depressive and 5. Moore, P. D. (2006) Conservation biology: unkind cuts for incense. Nature 444, 829 anxiety disorders. Our observation that IA reduces activity 6. Encyclopaedia Britannica, 15th Ed. (1975) Micropaedia, Vol. 5, in the open field and the ring tests to the same extent in p. 322, University of Chicago, Chicago, IL, USA Ϫ Ϫ both TRPV3 / and WT mice implies that the sedation 7. Gunther, R. T. (1959) The Greek Herbal of Dioscorides, Hafner, produced by IA is not mediated via TRPV3 channels. New York, NY, USA 8. Shachter, J., and Epstein, I., eds. (1935) Seder Nezikin, San- IA is chemically unrelated to compounds that are hedrin Tractate. In The Babylonian Talmud, p. 43a, Suncino currently in use as anxiolytic and antidepressive drugs. Press, New York, NY, USA It belongs to an important group of common natural 9. Getahon, A. (1976) Some Common Medicinal and Poisonous Plants Used in Ethiopian Folkmedicine (mimeographed), Addis Abeba products, cembranoid diterpenes. Many of these com- University, Addis Abeba, Ethiopia pounds exhibit biological activities; however, to the best 10. Menon, M. K., and Kar, A. (1971) Analgesic and psychophar- of our knowledge, no work has been done on the CNS macological effects of the gum resin of Boswellia serrata. Planta activities of these compounds to date. The identifica- Med. 19, 333–341 11. Moussaieff, A., Shohami, E., Kashman, Y., Fride, E., Schmitz, tion of CNS effects by a cembranoid diterpene may lead M. L., Renner, F., Fiebich, B. L., Munoz, E., Ben-Neriah, Y., and to a novel group of antidepressive and anxiolytic Mechoulam, R. (2007) Incensole acetate, a novel anti-inflam- agents, originating from an ancient drug. matory compound isolated from Boswellia resin, inhibits nuclear Taken together, our data support our original con- factor-kappaB activation. Mol. Pharmacol. 72, 1657–1664 12. Moussaieff, A., Shein, N. A., Tsenter, J., Grigoriadis, S., Sime- tention, namely that Boswellia resin may affect sensation onidou, C., Alexandrovich, A., G., Trembovler, V., Ben-Neriah, and emotional states. These observations also indicate Y., Schmitz, M. L., Fiebich, B. L., Munoz, E., Mechoulam, R., that the TRPV3 channel is involved in emotional and and Shohami, E. (2008) Incensole acetate—a novel neuropro- tective agent isolated from Boswellia carterii. [E-pub ahead of behavioral processes in the CNS in addition to its print] J. Cereb. Blood Flow Metab. doi:10.1038/jcbfm.2008.28 known effects on the perception of warmth (ther- 13. Corsano, S., and Nicoletti, R. (1967) The structure of incensole. mosensation). It is possible that IA augments the eu- Tetrahedron 23, 1977–1984 phoric feeling produced during religious functions, 14. Hamm, S., Bleton, J., Connan, J., and Tchapla, A. (2005) A chemical investigation by headspace SPME and GC-MS of due to both positive, presumably mild, emotional ef- volatile and semi-volatile terpenes in various olibanum samples. fects and the sensation of warmth. Thus the neurobe- Phytochemistry 66, 1499–1514 havioral effect of IA may provide a biochemical basis for 15. Ramsey, I. S., Delling, M., and Clapham, D. E. (2006) An introduction to TRP channels. Annu. Rev. Physiol. 68, 619–647 the millennial and widespread use of Boswellia-contain- 16. Chung, M. K., Lee, H., Mizuno, A., Suzuki, M., and Caterina, ing incense. However, only direct human trials includ- M. J. TRPV3 and TRPV4 mediate warmth-evoked currents in ing the investigation of human dosage and dosage primary mouse keratinocytes. J. Biol. Chem. 279, 21569–21575 forms may give final, concrete proof. 17. Xu, H., Ramsey, I. S., Kotecha, S. A., Moran, M. M., Chong, J. A., Lawson, D., Ge, P., Lilly, J., Silos-Santiago, I., Xie, Y., DiStefano, Thus, IA joins the notable group of plant-derived P. S., Curtis, R., and Clapham, D. E. (2002) TRPV3 is a compounds that are active in the mammalian CNS. It is calcium-permeable temperature-sensitive cation channel. Nature therefore possible that IA would provide us with novel 418, 181–186 therapeutic agents and neurochemical insights. 18. Smith, G. D., Gunthorpe, M. J., Kelsell, R. E., Hayes, P. D., Reilly, P., Facer, P., Wright, J. E., Jerman, J. C., Walhin, J. P., Ooi, L., Egerton, J., Charles, K. J., Smart, D., Randall, A. D., Anand, P., We thank Dr. Ardem Patapoutian (Department of Cell Biol- and Davis, J. B. (2002) TRPV3 is a temperature-sensitive va- ogy, Scripps Research Institute, La Jolla, CA, USA) for the nilloid receptor-like protein. Nature 418, 186–190 generous gift of TRPV3 null mice. This research was supported 19. Chung, M. K., Lee, H., Mizuno, A., Suzuki, M., and Caterina, M. J. by the following U.S. National Institues of Health grants: (2004) 2-aminoethoxydiphenyl borate activates and sensitizes the DA9789 (to R.M.), NS051551 (to M.J.C.), DA11322 and heat-gated ion channel TRPV3. J. Neurosci. 24, 5177–5182 DA21696 (to K.M), and DA018244 and DA020402 (to J.M.W.). 20. Moqrich, A., Hwang, S. W., Earley, T. J., Petrus, M. J., Murray, A. N., Spencer, K. S., Andahazy, M., Story, G. M., and Patapoutian, A. This work was also supported by the Lilly Foundation Inc. (2005) Impaired thermosensation in mice lacking TRPV3, a heat (Indianapolis, IN, USA) and the Gill Center for Biomolecular and camphor sensor in the skin. Science 307, 1468–1472 Science at Indiana University in Bloomington and the Samaria 21. Xu, H., Delling, M., Jun, J. C., and Clapham, D. E. (2006) and Jordan Rift R&D Center (to E.F.). S.M.H. is a Pfizer Fellow Oregano, thyme and clove-derived flavors and skin sensitizers of the Life Sciences Research Foundation. E.S. is the incumbent activate specific TRP channels. Nat. Neurosci. 9, 628–635

10 Vol. 22 August 2008 The FASEB Journal MOUSSAIEFF ET AL. 22. Chung, M. K., Guler, A. D., and Caterina, M. J. (2005) Biphasic Egerton, J., Charles, K. J., Smart, D., Randall, A. D., Anand, P., currents evoked by chemical or thermal activation of the and Davis, J. B. (2002) TRPV3 is a temperature-sensitive va- heat-gated ion channel, TRPV3. J. Biol. Chem. 280, 15928–15941 nilloid receptor-like protein. Nature 418, 186–190 23. Guler, A. D., Lee, H., Iida, T., Shimizu, I., Tominaga, M., and 37. Sheehan, D. V., and Sheehan, K. H. (2007) Current approaches Caterina, M. (2002) Heat-evoked activation of the ion channel, to the pharmacologic treatment of anxiety disorders. Psychophar- TRPV4. J. Neurosci. 22, 6408–6414 macol. Bull. 40, 98–109 24. Caterina, M. J., Rosen, T. A., Tominaga, M., Brake, A. J., and 38. Nurnberg, H. G., Hensley, P. L., Gelenberg, A. J., Fava, M., Julius, D. A. (1999) A capsaicin-receptor homologue with a high Lauriello, J., and Paine, S. (2003) Treatment of antidepressant- threshold for noxious heat. Nature 398, 436–441 associated sexual dysfunction with sildenafil: a randomized 25. Crawley, J. N. (2000) What’s Wrong with My Mouse? Behavioral controlled trial. JAMA 289, 56–64 Phenotyping of Transgenic and Knockout Mice, Wiley-Liss, New York, 39. Keller, M. B. (2003) Past, present, and future directions for NY, USA defining optimal treatment outcome in depression: remission 26. Treit, D., and Menard, J. (1998) Animal models of anxiety and and beyond. JAMA 289, 3152–3160 depression, In In vivo Neuromethods (Boulton, A., Baker, G., and 40. Bandelow, B., Wedekind, D., and Leon, T. (2007) Pregabalin for Bateson, A., eds) p 89, Humana Press, Totowa, Canada the treatment of generalized anxiety disorder: a novel pharma- 27. Harkin, A., Connor, T. J., Burns, M. P., and Kelly, J. P. (2004) cologic intervention. Expert Rev. Neurother. 7, 769–781 Nitric oxide synthase inhibitors augment the effects of serotonin 41. Haberfellner, E. M. (2007) A review of the assessment of re-uptake inhibitors in the forced swimming test. Eur. Neuropsy- antidepressant-induced sexual dysfunction used in randomized, chopharmacol. 14, 274–281 controlled clinical trials. Pharmacopsychiatry 40, 173–182 28. Petit-Demouliere, B., Chenu, F., and Bourin, M. (2005) Forced 42. Werme, M., Ringholm, A., Olson, L., and Brene, S. (2000) swimming test in mice: a review of antidepressant activity. Differential patterns of induction of NGFI-B, Nor1 and c-fos Psychopharmacology 177, 245–255 mRNAs in striatal subregions by haloperidol and clozapine. 29. Fride, E., Perchuk, A., Hall, F. S., Uhl, G. R., and Onaivi, E. S. Brain Res. 863, 112–119 (2006) Behavioral methods in cannabinoid research. Methods 43. Dragunow, M., Robertson, G. S., Faull, R. L., Robertson, H. A., Mol. Med. 123, 269–290 and Jansen, K. (1990) D2 dopamine receptor antagonists in- 30. Fride, E., and Mechoulam, R. (1993) Pharmacological activity of duce fos and related proteins in rat striatal neurons. Neuroscience the cannabinoid receptor agonist, , a brain constit- 37, 287–294 231, uent. Eur. J. Pharmacol. 313–314 44. Thompson, B. L., and Rosen, J. B. (2006) Immediate-early gene 31. Richards, M. H. (1990) Rat hippocampal muscarinic autorecep- expression in the central nucleus of the amygdala is not specific tors are similar to the M2 (cardiac) subtype: comparison with for anxiolytic or anxiogenic drugs. Neuropharmacology 50, 57–68 hippocampal M1, atrial M2 and ileal M3 receptors. Brit. J. Phar- 45. Henry, B., Vale, W., and Markou, A. (2006) The effect of lateral macol. 99, 753–761 septum corticotropin-releasing factor receptor 2 activation on 32. Angel, I., and Bidet, S. (1991) The binding site for [3H]gliben- anxiety is modulated by stress. J. Neurosci. 26, 9142–9152 clamide in the rat cerebral cortex does not recognize K-channel 46. Dhaka, A., Viswanath, V., and Patapoutian, A. (2006) Trp ion agonists or antagonists other than sulphonylureas. Fundam. Clin. channels and temperature sensation. Annu. Rev. Neurosci. 29, Pharmacol. 5, 107–115 135–161 33. O’Dell, D. K., Rimmerman, N., Pickens, S. R., and Walker, J. M. (2007) Fatty acyl amides of endogenous tetrahydroisoquinolines 47. Smith, P. L., Maloney, K. N., Pothen, R. G., Clardy, J., and are active at the recombinant human TRPV1 receptor. Bioorg. Clapham, D. E. (2006) Bisandrographolide from Andrographis Med. Chem. 15, 6164–6169 paniculata activates TRPV4 channels. J. Biol. Chem. 28, 129897– 34. Chung, M. K., Lee, H., Mizuno, A., Suzuki, M., and Caterina, M. J. 129904 (2004) TRPV3 and TRPV4 mediate warmth-evoked currents in 48. Vogt-Eisele, A. K., Weber, K., Sherkheli, M. A., Vielhaber, G., primary mouse keratinocytes. J. Biol. Chem. 279, 21569–21575 Panten, J., Gisselmann, G., and Hatt, H. (2007) Monoterpenoid 35. Marsch, R., Foeller, E., Rammes, G., Bunck, M., Ko¨ssl, M., Hols- agonists of TRPV3. Br. J. Pharmacol. 151, 530–540 boer, F., Zieglga¨nsberger, W., Landgraf, R., Lutz, B., and Wotjak, 49. Paxinos, G., and Franklin, K. B. J. (2001) The Mouse Brain in C. T. (2007) Reduced anxiety, conditioned fear, and hippocampal Stereotaxic Coordinates, Academic Press, Oxford, UK long-term potentiation in transient receptor potential vanilloid type 1 receptor-deficient mice. J. Neurosci. 27, 832–839 36. Smith, G. D., Gunthorpe, M. J., Kelsell, R. E., Hayes, P. D., Reilly, Received for publication December 17, 2007. P., Facer, P., Wright, J. E., Jerman, J. C., Walhin, J. P., Ooi, L., Accepted for publication April 10, 2008.

INCENSOLE ACETATE FROM INCENSE IS PSYCHOACTIVE 11