Cannabis Health Search

Kidney's & Cannabis studies completed


Cannabis "marijuana" has not been shown to cause changes or damage to the kidneys. In fact, it has never been shown to cause any kind of permanent damage to the lungs, heart, liver, kidneys, brain or any of the other areas of the body it is said to affect.

 As a matter of fact, some studies have actually shown it to help decrease the spread of lung and glioma (brain cancer), and it has been shown to increase sperm production and fertility.

Science and Research

2013 - Report ~ Acute Kidney Injury Associated with Synthetic Cannabinoid Use — Multiple States,

2013 - Study ~ Cannabinoid Receptors as Therapeutic Targets for Dialysis-Induced Peritoneal Fibrosis.

2013 - Study ~ β-Caryophyllene ameliorates cisplatin-induced nephrotoxicity in a cannabinoid 2
receptor-dependent manner

2013 - Study ~ Cannabinoid Receptor 2 Expression in Human Proximal Tubule Cells is Regulated by Albumin Independent of ERK1/2 Signaling.

2013 - Study ~ First Metabolic Profile of XLR-11, a Novel Synthetic Cannabinoid, Obtained by Using Human Hepatocytes and High-Resolution Mass Spectrometry.

2013 - News ~ Synthetic Marijuana Dangerous for Kidneys

2013 - News ~ Synthetic Marijuana Harms Kidneys of 16 Users, CDC Reports

2013 - News ~ Synthetic drugs carry risk of kidney damage

2012 - Study ~ β-Caryophyllene ameliorates cisplatin-induced nephrotoxicity in a cannabinoid 2 receptor-dependent manner.

2012 - Study ~ Cannabidiol treatment ameliorates ischemia/reperfusion renal injury in rats.

2012 - Study ~ AKI Associated with Synthetic Cannabinoids: A Case Series.

2012 - News ~ Outbreak of kidney failure in Wyoming linked to "Spice"

2012 - News ~ Wyoming kidney failure outbreak linked to designer 'blueberry spice' drug, aka 'legal marijuana'

2012 - News ~ Blueberry “spice” in Wyoming linked to cases of renal failure

2012 - News ~ New health concerns about 'fake pot' in US

2011 - Study ~ Distinctive effects of plant protein sources on renal disease progression and associated cardiac hypertrophy in experimental kidney disease.

2011 - Study ~ Is there a legitimate role for the therapeutic use of cannabinoids for symptom management in chronic kidney disease?

2011 - Study ~ Cannabinoid hyperemesis syndrome inducing acute prerenal failure and electrolyte disturbance.
2005 - News - Cream with endocannabinoids effective in the treatment of pruritus due to kidney disease.

Modulation of P-glycoprotein activity by cannabinoid molecules in HK-2 renal cells (full - 2006)

Br J Pharmacol. 2006 July; 148(5): 682–687.
Published online 2006 May 22. doi: 10.1038/sj.bjp.0706778.
PMCID: PMC1751877
Modulation of P-glycoprotein activity by cannabinoid molecules in HK-2 renal cells
Paola Nieri,* Nadia Romiti, Barbara Adinolfi, Andrea Chicca,1 Ilaria Massarelli, and Elisabetta Chieli
Dipartimento di Psichiatria, Neurobiologia, Farmacologia e Biotecnologie, Università di Pisa, Via Bonanno, 6, Pisa 56126, Pisa, Italia
Dipartimento di Patologia Sperimentale, Biotecnologie Mediche, Infettivologia ed Epidemiologia, Università di Pisa, Pisa, Italia
Dipartimento di Chimica e Chimica Industriale, Università di Pisa, Pisa, Italia
Author for correspondence:
Received January 9, 2006; Revised March 7, 2006; Accepted April 6, 2006.
  • Endogenous and synthetic cannabinoid molecules have been investigated as possible MDR-1/P-glycoprotein (P-gp) modulators in HK-2-immortalized renal cells, using calcein acetoxymethylester (calcein-AM) as a P-gp substrate.
  • Among the endocannabinoid molecules tested, anandamide (AEA), but not 2-arachidonoyl-glycerol (2-AG) or palmitoyl-ethanolamide (PEA), increased the intracellular fluorescence emitted by calcein, a metabolic derivative of the P-gp substrate calcein-AM, indicative of a reduction in transport capacity.
  • All the three synthetic cannabimimetics tested, that is, R-(+)-methanandamide (R(+)-MET), AM 251 and CP55,940 significantly increased calcein accumulation in the cytosol.
  • RT–PCR demonstrated that HK-2 cells do not express CB1 or CB2 cannabinoid receptors.
  • R(+)-MET, AM251 and CP55,940 were also evaluated as modulators of P-gp expression, by Western blot analysis. Only AM251 weakly enhanced the protein levels (by 1.2-fold) after a 4-day-long incubation with the noncytotoxic drug concentration 2 μM.
  • The present data provide the first evidence that the endocannabinoid AEA and different synthetic cannabinoids may inhibit the P-gp activity in vitro via a cannabinoid receptor-independent mechanism.
Keywords: HK-2 cells, proximal tubules, kidney, MDR-1, P-glycoprotein, cannabinoid

P-glycoprotein, (P-gp), encoded by the MDR1 gene, is a membrane protein belonging to the ATP-binding cassette superfamily and it is a transmembrane efflux pump. Initially discovered in tumour cells while investigating the multidrug resistance mechanism, it has been also found and studied in many nontumour cells/tissues (Montano et al., 1996; Smit et al., 1998; Dautrey et al., 1999; Jonker et al., 1999; Gutmann et al., 2000; Florea et al., 2001; Batetta et al., 2003; Cisternino et al., 2003; Goralski et al., 2003; Elliott et al., 2004).
P-gp indeed is involved in all pharmacokinetic events (Varma et al., 2003) and many single-nucleotide polymorphisms in the MDR1 gene have been correlated with differences in individual drug responsiveness (Marzolini et al., 2004).
P-gp shows a broad substrate specificity with a prevalent affinity for hydrophobic compounds; many natural and synthetic xenobiotics and endogenous molecules are P-gp substrates or modulators (Chan et al., 2004). Cannabis sativa derivatives are among the natural compounds influencing P-gp activity. In multidrug-resistant mouse lymphoma cells, cannabinol, cannabispirol and cannabidiol increase cytotoxic drug accumulation, whereas cannabidiolic acid, tetrahydrocannabidiolic acid and Δ9-tetrahydrocannabinol (THC) reduce it (Molnar et al., 2000).
In humans, many actions of THC are mimicked by endogenous molecules called endocannabinoids (De Petrocellis et al., 2004). N-arachidonoyl-ethanolamide (anandamide (AEA)) and 2-arachidonoyl-glycerol (2-AG) are the main endogenous agonists at cannabinoid receptors. Palmitoyl-ethanolamide (PEA) and other molecules are candidate members of the endocannabinoid mediator family (Facci et al., 1995; De Petrocellis et al., 2004; Bradshaw & Walker, 2005). Endocannabinoids are produced under physiological and pathological conditions by central and peripheral tissues, suggesting a possible therapeutic use of cannabimimetics or cannabinoid receptor antagonists in different pathologies (Goutopoulos & Makriyannis, 2002).
As no study has aimed to investigate the interaction between endocannabinoids or synthetic cannabinoids and P-gp until now, we tested the effect of some endogenous and synthetic cannabinoid molecules in regulating P-gp activity in HK-2 cells. This cell line retains many of the phenotypic and functional properties of in vivo proximal tubules (Ryan et al., 1994) and it is an useful in vitro model to study the MDR1-codified protein (Tramonti et al., 2001; Romiti et al., 2002; 2004).
Verapamil (VP) and probenecid were obtained from Sigma Italia (Milan, Italy). AM251, R(+)-methanandamide (R(+)-MET), AEA, 2-AG, PEA and CP55,940 ((-)-cis-3-[2-hydroxy-4-(1,1-dimethylheptyl)phenyl]-trans-4-(3-hydroxypropyl) cyclohexanol) were purchased from Tocris Cookson Ltd (Avonmouth, U.K.). AEA and R(+)-MET were dissolved in ethanol, the other compounds in DMSO (stock solutions) and diluted in distilled water to the required concentrations on the day of experiment.
Cell culture
The immortalized proximal tubule epithelial cell line from normal adult human kidney (HK-2) was purchased from the American Type Cell Collection and cultured as described (Romiti et al., 2002). Briefly, cells were grown in DMEM/F12 medium supplemented with L-glutamine and antibiotics (penicillin and streptomycin), insulin, transferrin, Na-selenite, T3, hydrocortisone, prostaglandin E1 and 5% foetal calf serum. All chemicals, media and cell culture reagents were obtained from Sigma-Aldrich (Milan, Italy).
Medium was changed three times a week and cells were subcultured weekly. All experiments were performed on confluent cells maintained for 48 h in serum-free medium. In experiments testing the effects of cannabinoids on P-gp expression, cells were cultured for 4 days in the presence of a single cannabinoid molecule or its vehicle before protein extraction. Preliminary tests were made to select cannabinoid concentrations that do not decrease cell viability.
Calcein-acetoxymethylester (calcein-AM) test
Evaluation of P-gp activity was performed by the fluorimetric measurement of the intracellular accumulation of calcein produced by ester hydrolysis of the P-gp substrate calcein-AM (Apotech, Geneva, Switzerland), as the transport capacity of P-gp is inversely proportional to the intracellular accumulation of fluorescent calcein (Hauser et al., 1998).
Indeed, calcein is not a P-gp substrate and it cannot leave the cell via the plasma membrane, whereas the nonfluorescent calcein-AM is extruded from the MDR-1-expressing cells (Homolya et al., 1993). Calcein assay was performed in 96-well plates.
HK-2 cells, cultured in microplates, were preincubated for 15 min with cannabinoid molecules. Thereafter, calcein-AM was added at 0.5 μM final concentration and calcein fluorescence measured after 1 h incubation at 37°C at λex=485 nm and λem=538 nm by a fluorescent plate reader (Fluoroskan II, Dasit, Italy). The acknowledged P-gp modulator VP was used as internal standard (positive control).
RNA isolation and reverse transcriptase–polymerase chain reaction (RT–PCR)
Total RNA was isolated from 106 cells using the ‘SV total RNA isolation system kit', which includes a Dnase treatment (Promega, Italy). Reverse transcriptase reaction was performed with ‘Qiagen OmniScript RT kit' (Qiagen, Milan, Italy) on 1.5 μg total RNA. Human CB1 and CB2 receptors were amplified using primers designed to optimize their specificity and thermodynamic properties by using the Oligo-Primer Analysis Software, v. 4.0.
Primers were synthesized by Sigma-Genosys Ltd (Cambridge, U.K.). CB1 forward (5′-CCACTCCCGCAGCCTCCG-3′) and CB1 reverse (5′-ATGAGGCAAAACGCCACCAC-3′) primers yield a 294 bp product (517–810 position in Genbank accession no. X81120); CB2 forward (5′-CGCCGGAAGCCCTCATAC-3′) and CB2 reverse (5′-CCTCATTCGGGCCATTCTTG-3′) primers yield a 523 bp product (318–840 position in Genbank accession X74328).
Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene was used as expression standard and amplified with forward (5′-GTGAAGGTCGGTGTCAACG-3′) and reverse (5′-GGTGAAGACGCCAGTAGACTC-3′) primers yielding a 300 bp product (85–384 position in Genbank accession NM_002046.2). PCR was performed with ‘HotStartTaq Master Mix kit' (Qiagen, Milan, Italy) using 1/10 of cDNA in 12.5 μl final volume. PCR conditions were as follows: 95.0°C 15 min, followed by 1 min at 95.0°C, 1 min at 56.8°C, 1 min at 72.0°C (35 cycles) and then 10 min at 72.0°C for CB1; 95.0°C 15 min, followed by 1 min at 95.0°C, 1 min at 58.0°C, 1 min at 72.0°C (35 cycles) and then 10 min at 72.0°C for CB2; and 95.0°C 15 min, followed by 1 min at 95.0°C, 1 min at 55.0°C, 1 min at 72.0°C (36 cycles) and then 10 min at 72.0°C for GAPDH.
All protocols were carried out in a MyCycler Thermalcycler (Bio-Rad, Milan, Italy) and PCR products were run on ethidium bromide-stained 1% agarose gel.
The DNA extracted from HK-2 cells (by using DNeasy Tissue kit (Qiagen, Milan, Italy)) was used as PCR-positive control, taking advantage from the fact that the amplified sequences belong to single exons.
PCR product identity was confirmed by sequencing (CRIBI Centre, University of Padova, Italy).
SDS–PAGE and Western blotting.
P-gp was immunodetected by semiquantitative Western blot analysis. Crude membranes of HK-2 cells were obtained and extracted as described (Romiti et al., 2002).
Ten micrograms of membrane proteins were separated by SDS–PAGE on 6% acrylamide Laemmli minigels and transferred overnight onto nitrocellulose membranes.
Equal loading conditions in gels were routinely ascertained by staining blots with Ponceau-S. For immunoblotting, mdr Ab-1 polyclonal serum and peroxidase-conjugated secondary antibody (Inalco, Milan, Italy) were used. Blots were developed with the ECL detection system (Amersham Pharmacia Biotech, Italy) and analysed by densitometry (Gel Documentation System Chemi Doc and QuantiOne version 4.3 software, Bio-Rad, Milan, Italy).
Cytotoxicity assay
Cytotoxicity of cannabinoid compounds was tested to rule out the possibility that cytotoxic effects alter both the intracellular fluorescence in the calcein assay and the P-gp expression.
Cells were grown in 96-multiwell plates in the absence or in the presence of cannabinoid molecules for 15 min to study effects of cannabinoids on P-gp activity or for 4 days to study effects of cannabinoids on P-gp expression. Cell viability was then evaluated by the WST-1 (4-[3-(4-Iodophenyl)-2-(4-nitrophenyl)-2H-5-tetrazolio]-1,3-benzene) colorimetric test (Roche, Milan, Italy). WST-1 reagent undergoes intracellular cleavage by mitochondrial dehydrogenases in viable cells forming a dark red formazan product, which is quantified by absorbance at 450 nm.
Briefly, WST-1 was added to each well and after 1 h incubation at 37°C, the absorbance was measured by a microplate reader (Wallac Victor II, Perkin-Elmer, MA, U.S.A.). In preliminary experiments, cytotoxicity of drug vehicles (never greater than 0.1% (vv−1)) was excluded.
Computer-calculated logarithm of partition coefficient (ClogP) and solvent-accessible surface area (SASA)
Correlation between hydrophobicity and biological activity of cannabinoid molecules was tested as well. ClogP value is an accurate predictor of logP(ow) (log octanol/water partition coefficient) (Machatha & Yalkowsky, 2005) and it was calculated by means of the ChemDraw 6.0 program (CambridgeSoft, U.K.) by dissecting the solute under study into chemically meaningful fragments of known hydrophobicity. Molecular models were constructed by using standard geometries (standard bond lengths and angles) with the InsightII molecular modeling program (Biosym/MSI, San Diego, CA, U.S.A.) and a molecular mechanic approach was exploited for energy minimization of the molecules belonging to the entire data set, in order to obtain reasonable 3D arrangements.
The cff91 force field was exploited as well. Energy minimizations were carried out with steepest descent and conjugate gradient minimization algorithms, till 0.001 kcal/mol Å root-mean-square deviation was achieved.
SASA and its polar and nonpolar fractions for each cannabinoids were calculated with the Homology module of InsightII.
Data evaluation and statistical analysis
The percent ratio between calcein-dependent fluorescence emission recorded from cells exposed to the drug and control cells was calculated to assess the P-gp activity. Similarly, in the cytotoxicity assay, the absorbance values are reported as % of control value (no treatment). Results are expressed as mean±s.e.m. from n replicates.
Differences among drug treatments were evaluated by the analysis of variance and the Bonferroni post-test. Immunoblotting results from treated cells were compared to immunoblotted protein in vehicle-treated cells after densitometric acquisition; differences between densitometric values for each treatment vs its control were evaluated by Student's t-test for unpaired data. P0.05 was taken to be significant.

Functional studies
Among endocannabinoids, each tested at 20 μM and compared to VP at the same concentration, only AEA significantly enhanced the intracellular fluorescence after calcein-AM incubation (up to 159. 0±10.8%, n=4; P0.05, analysis of variance followed by Bonferroni post-test). In the presence of 2-AG or PEA, on the contrary, no significant difference with respect to control was evident (fluorescence with respect to control: 123.3±16.9 and 101±5.9%, respectively, n=4 each) (Figure 1).
Figure 1
Figure 1
Effects of cannabinoid molecules and verapamil (standard compound) at 20 μM on calcein accumulation in HK-2 cells. Cells were preincubated for 15 min with the compounds and then calcein-AM 0.5 μM was added. After (more ...)
Among synthetic cannabinoids, all molecules tested (20 μM each) significantly inhibited (P[less-than-or-eq, slant]0.05) calcein-AM efflux from the cell. In the case of CP55,940 a particularly remarkable efflux inhibition appeared (398.7±37.4% of control fluorescence, n=4), which was higher than that of VP (269.2±25.5%, n=4). The inhibitor of multidrug resistance proteins (MRPs) probenecid (1 mM) did not significantly modify the basal calcein efflux in HK-2 cells and response to cannabinoids (Figure 1). This suggests that only P-gp modulation was responsible for the effects observed in calcein assay.
Figure 2 shows the concentration-dependent curves for the three synthetic cannabinoids R(+)-MET, CP55,940 and AM251 in the range 0.6–20 μM.
Figure 2
Figure 2
Concentration-dependent effects of AM251, CP55,940 and R(+)-MET on calcein accumulation in HK-2 cells. Cells were preincubated for 15 min with the tested compounds and then calcein-AM 0.5 μM was added. After 1 h, (more ...)
None of the tested compounds quenched the calcein fluorescence and no cytotoxicity was revealed after a time exposure correspondent to that used in calcein assay (data not shown).
P-gp expression by immunoblotting
Immunodetectable P-gp membrane content was not significantly changed after 4 days incubation with 0.6 and 2 μM R(+)-MET or CP55,940. A modest but significant increase of the protein (by 20%, n=3; P[less-than-or-eq, slant]0.05; Student's t-test for unpaired data) was, on the other hand, observed in the presence of AM251 at 2 μM (Figure 3). At the concentrations 0.6 and 2 μM no cytotoxicity was observed by all the three cannabinoids, whereas a significant (n=3; P[less-than-or-eq, slant]0.05, analysis of variance followed by Bonferroni post-test) viability decrease was observed at 6 μM with CP55,940 and at 20 μM with all drugs (Figure 4).
Figure 3
Figure 3
(a) Western blot analysis (representative photo of three experiments) for P-gp in crude membranes from HK-2 cells cultured for 4 days in basal serum-free medium (lane 1) or in serum-free medium supplemented with 0.6 and 2 μM CP55,940 (lanes (more ...)
Figure 4
Figure 4
Cell viability of HK-2 cells as revealed by WST-1 colorimetric assay after 4 days incubation with the synthetic cannabinoids AM251, CP55,940 and R(+)-MET in the range 0.6–20 μM. Data are mean±s.e.m. from four replicates. (more ...)
CB1 and CB2 receptor expression by RT–PCR
Although DNA products of the expected size (294 bp for CB1 and 523 bp for CB2) and sequence were obtained by performing control PCRs on HK-2 cell DNA, and the control RT–PCR for GAPDH confirmed the quality of the cDNA template, by synthesizing a single 300 bp amplicon, no CB1 and CB2 receptor-specific RT–PCR products were obtained from total RNA of HK-2 cells (Figure 5).
Figure 5
Figure 5
PCR products of CB1 and CB2 cannabinoid receptors from DNA of HK-2 cells and RT–PCR products of GAPDH (housekeeping gene), CB1 and CB2 receptors from total RNA of the same cells, as revealed after 1% agarose gel submarine electrophoresis (more ...)
ClogP and molecular surface areas
Table 1 shows the ClogP values of the tested compounds, ranging from 5.8 for CP55,940 to 7.0 for AM251. Values of SASA with the relative polar and nonpolar fractions are reported as well. No correlation between the P-gp-modulating activity of the cannabinoid compounds and their hydrophobicity or surface area-derived descriptors was found.
Table 1
Table 1
ClogP, SASA and its polar and nonpolar fractions, for each cannabinoid tested as P-gp modulator
Our results give the first evidence that an endocannabinoid molecule and some cannabinergics are modulators of the P-gp activity. Indeed, AEA was demonstrated to decrease calcein efflux via P-gp in cultured HK-2 renal cells. This activity was not observed with 2-AG and PEA. On the other hand, all the synthetic cannabinoids tested, that is, CP55,940, AM251 and R(+)-MET, had a similar effect to AEA, in the following order of P-gp inhibition potency: CP55,940> AM251> R(+)-MET.
The observed cannabinoid activity on P-gp does not seem to be CB receptor-mediated, because of the evidence that no mRNA encoding for CB1 or CB2 receptors can be detected with RT–PCR. On the other hand, the agonist/antagonist profile on CB receptors of the drugs under study already suggested the involvement of these receptors to be unlikely in the P-gp modulation. R(+)-MET and CP55,940 are, in fact, CB agonists, whereas AM251 has been extensively used as a CB1 antagonist and more recently reported as CB2 inverse agonist (New & Wong, 2003).
In addition, the finding that the inhibitory potency of the stable AEA analogue was similar to that of AEA itself suggests that FAAH is also absent in HK-2 cells.
Although a functional response of canine tubular renal cells to the CB agonist CP55,940 has been reported, it was not blocked by CB antagonists, suggesting that the response was dissociated from the activation of CB receptors (Chou et al., 2001). Moreover, the renal expression of CB receptors and the AEA-degrading enzyme, FAAH, was found to be associated with the vasculature (Deutsch et al., 1997).
We found no evidence of cannabinoid cytotoxicity at the concentration and for the drug exposure time we adopted in the calcein assay. Cytotoxicity induced by P-gp inhibition in HK-2 cells has been reported by Zager (2001). In our experiments, a significant decrease in cell viability was observed only after a prolonged exposure (4 days) to cannabinoids and at concentrations higher than 2 μM. In addition, no correlation was observed at 6 μM CP55,940, AM251 or R(+)-MET between the drug inhibitory activity on P-gp and the extent of cytotoxicity 4 days later, although a positive correlation was observed when these cannabinoids were used at a higher concentration of 20 μM. Mechanisms underlying the different cytotoxic profile between drugs such as CP55,940 and AM251, which have at 6 μM similar inhibitory activities on P-gp efflux but different effects on cell survival, remain to be elucidated.
Not even the influence on P-gp expression may give a rational explanation as only a modest increase (by 20%) of P-gp gene transcription by AM251 was observed. It seems therefore reasonable that other factors might be involved, for example, a different drug metabolism and/or cell responses other than those concurring with the P-gp inhibition.
The existence of a specific AEA transporter is actually debated (Glaser et al., 2003; Matthew et al., 2004) and MDR-1 protein has been postulated to be an AEA transmembrane transporter by Goutopoulos & Makriyannis (2002). In this study, we did not investigate the transmembrane transport of cannabinoids, but the P-gp inhibition we observed may be owing to competition of molecules with different hydrophobicity for the same transporter.
Hydrophobic substances are typical substrates for P-gp (Zamora et al., 1988). Nevertheless, a correlation between steroid hydrophobicity and P-gp activity inhibition, reported by Yang et al. (1989), was not confirmed by other authors (Ueda et al., 1992; 1994; Hamilton et al., 2001). In addition, there is evidence that the P-gp hydrophobic substrates gain access to the transporter before leaving the inner side of the plasma membrane for the cytoplasm (Sharom, 2003) and a reduced polar surface area of a drug has been reported to correlate better than drug lipophilicity with an increased membrane permeation rate (Veber et al., 2002). Nevertheless, we did not observe any correlation between both the cannabinoid surface area descriptors, that is, SASA and its polar and nonpolar fractions and the cannabinoid ClogP values, and the inhibition of calcein efflux.
This seems to suggest that the molecular mechanism of calcein efflux inhibition we observed is not based on the simple competition of molecules with different hydrophobicity for the same transporter, such as a cannabinoid and calcein-AM. In this regard, Seelig & Landwojtowicz (2000) observed that although the partitioning into the membrane was the rate-limiting step for P-gp-substrate interaction, dissociation of the P-gp–substrate complex depended on the number and strength of the hydrogen bonds between the substrate and the membrane embedded region of the transporter. Furthermore, specific structural elements that are critical for the recognition by P-gp (Pearce et al., 1989) and different drug binding sites on the protein (Martin et al., 2000) has been suggested.
Irrespective to the molecular mechanism of action, the effect of AEA on P-gp activity lead us to hypothesize a role for the endocannabinoid system in regulating the P-gp activity in kidney and other districts such as intestine, liver and blood-brain barrier where P-gp is expressed and involved in the physiological processes of absorption, distribution, excretion and metabolism (Varma et al., 2003).
Similarly, synthetic cannabinoids involved in therapeutic regimens (Tanigawara, 2000) may have a role in modifying the pharmacokinetic profile and the bioavailability of other drugs. In this perspective, it seems noteworthy to us that a drug structurally similar to AM251, SR141716A (rimonabant®), is currently undergoing clinical trials for obesity, smoking cessation and alcohol abuse (Fernandez & Allison, 2004).
In conclusion, our data show that P-gp activity may be modulated by the endocannabinoid system and by synthetic cannabinoids in HK-2 cells, in a CB1- and CB2-receptor-independent pathway. This opens an interesting scenario in kidney physiology and in the pharmacology and therapeutic potential of cannabinoid drugs.
MDRmultidrug resistance

  • BATETTA B., MULAS M.F., SANNA F., PUTZOLU M., BONATESTA R.R., GASPERI-CAMPANI A., RONCUZZI L., BAIOCCHI D., DESSI S. Role of cholesterol ester pathway in the control of cell cycle in human aortic smooth muscle cells. FASEB J. 2003;17:746–748. [PubMed]
  • BRADSHAW H.B., WALKER J.M. The expanding field of cannabimimetic and related lipid mediators. Br. J. Pharmacol. 2005;144:459–465. [PMC free article] [PubMed]
  • CHAN L.M., LOWES S., HIRST B.H. The ABCs of drug transport in intestine and liver: efflux proteins limiting drug absorption and bioavailability. Eur. J. Pharm. Sci. 2004;21:25–51. [PubMed]
  • CHOU K.J., TSENG L.L., CHENG J.S., WANG J.L., FANG H.C., LEE K.C., SU W., LAW Y.P., JAN C.R. CP55,940 increases intracellular Ca2+ levels in Madin–Darby canine kidney cells. Life Sci. 2001;69:1541–1548. [PubMed]
  • CISTERNINO S., BOURASSET F., ARCHIMBAUD Y., SEMIOND D., SANDERINK G., SCHERRMANN J.M. Nonlinear accumulation in the brain of the new taxoid TXD258 following saturation of P-glycoprotein at the blood–brain barrier in mice and rats. Br. J. Pharmacol. 2003;138:1367–1375. [PMC free article] [PubMed]
  • DAUTREY S., FELICE K., PETIET A., LACOUR B., CARBON C., FARINOTTI R. Active intestinal elimination of ciprofloxacin in rats: modulation by different substrates. Br. J. Pharmacol. 1999;127:1728–1734. [PMC free article] [PubMed]
  • DE PETROCELLIS L., CASCIO M.G., DI MARZO V. The endocannabinoid system: a general view and latest additions. Br. J. Pharmacol. 2004;141:765–774. [PMC free article] [PubMed]
  • DEUTSCH D.G., GOLIGORSKY M.S., SCHMID P.C., KREBSBACH R.J., SCHMID H.H., DAS S.K., DEY S.K., ARREAZA G., THORUP C., STEFANO G., MOORE L.C. Production and physiological actions of anandamide in the vasculature of the rat kidney. J. Clin. Invest. 1997;100:1538–1546. [PMC free article] [PubMed]
  • ELLIOTT J.I., RAGUZ S., HIGGINS C.F. Multidrug transporter activity in lymphocytes. Br. J. Pharmacol. 2004;143:899–907. [PMC free article] [PubMed]
  • FACCI L., DAL TOSO R., ROMANELLO S., BURIANI A., SKAPER S.D., LEON A. Mast cells express a peripheral cannabinoid receptor with differential sensitivity to anandamide and palmitoylethanolamide. Proc. Natl. Acad. Sci. U.S.A. 1995;92:3376–3380. [PMC free article] [PubMed]
  • FERNANDEZ J.R., ALLISON D.B. Rimonabant Sanofi–Synthelabo. Curr. Opin. Invest. Drugs. 2004;5:430–435.
  • FLOREA B.I, VAN DER SANDT I.C., SCHRIER S.M., KOOIMAN K., DERYCKERE K., DE BOER A.G., JUNGINGER H.E., BORCHARD G. Evidence of P-glycoprotein mediated apical to basolateral transport of flunisolide in human broncho-tracheal epithelial cells (Calu-3) Br. J. Pharmacol. 2001;134:1555–1563. [PMC free article] [PubMed]
  • GLASER S.T., ABUMRAD N.A., FATADE F., KACZOCHA M., STUDHOLME K.M., DEUTSCH D.G. Evidence against the presence of an anandamide transporter. Proc. Natl. Acad. Sci. U.S.A. 2003;100:4269–4274. [PMC free article] [PubMed]
  • GORALSKI K.B., HARTMANN G., PIQUETTE-MILLER M.A., RENTON K.W. Downregulation of mdr1a expression in the brain and liver during CNS inflammation alters the in vivo disposition of digoxin. Br. J. Pharmacol. 2003;139:35–48. [PMC free article] [PubMed]
  • GOUTOPOULOS A., MAKRIYANNIS A. From cannabis to cannabinergics: new therapeutic opportunities. Pharmacol. Ther. 2002;95:103–117. [PubMed]
  • GUTMANN H., MILLER D.S., DROULLE A., DREWE J., FAHR A., FRICKER G. P-glycoprotein- and mrp2-mediated octreotide transport in renal proximal tubule. Br. J. Pharmacol. 2000;129:251–256. [PMC free article] [PubMed]
  • HAMILTON K.O., YAZDANIAN M.A., AUDUS K.L. Modulation of P-glycoprotein activity in Calu-3 cells using steroids and beta-ligands. Int. J. Pharmacol. 2001;228:171–179.
  • HAUSER I.A., KOZIOLEK M., HOPFER U., THEVENOD F. Therapeutic concentrations of cyclosporine A, but not FK506, increase P-glycoprotein expression in endothelial and renal tubule cells. Kidney Int. 1998;54:1139–1149. [PubMed]
  • HOMOLYA L., HOLLO Z., GERMANN U.A., PASTAN I., GOTTESMAN M.M., SARKADI B. Fluorescent cellular indicators are extruded by the multidrug resistance protein. J. Biol. Chem. 1993;268:2143–2146.
  • JONKER J.W., WAGENAAR E., VAN DEEMTER L.R., BENDER H.M., DASENBROCK J., SCHINKEL A.H. Role of blood–brain barrier P-glycoprotein in limiting brain accumulation and sedative side-effects of asimadoline, a peripherally acting analgesic drug. Br. J. Pharmacol. 1999;127:43–50. [PMC free article] [PubMed]
  • MACHATHA S.G., YALKOWSKY S.H. Comparison of the octanol/water partition coefficients calculated by ClogP, ACDlogP and KowWin to experimental determined values. Int. J. Pharmacol. 2005;294:185–192.
  • MARTIN C., BERRIDGE G., HIGGINS C.F., MISTRY P., CHARLTON P., CALLAGHAN R. Communication between multiple drug binding sites on P-glycoprotein. Mol. Pharmacol. 2000;58:624–632. [PubMed]
  • MARZOLINI C., PAUS E., BUCLIN T., KIM R.B. Polymorphisms in human MDR1 (P-glycoprotein): recent advances and clinical relevance. Clin. Pharmacol. Ther. 2004;75:13–33. [PubMed]
  • MATTHEW J., MCFARLAND M.J., BARKER E.L. Anandamide transporter. Pharmacol. Ther. 2004;104:117–135. [PubMed]
  • MOLNAR J., SZABO D., PUSZTAI R., MUCSI I., BEREK L., OCSOVSZKI I., KAWATA E., SHOYAMA Y. Membrane associated antitumor effects of crocine-, ginsenoside- and cannabinoid derivates. Anticancer Res. 2000;20:861–867. [PubMed]
  • MONTANO E., SCHMITZ M., BLASER K., SIMON H.U. P-glycoprotein expression in circulating blood leukocytes of patients with steroid-resistant asthma. J. Invest. Allergol. Clin. Immunol. 1996;6:14–21.
  • NEW D.C., WONG YH. BML-190 and AM251 act as inverse agonists at the human cannabinoid CB2 receptor signalling via cAMP and inositol phosphates. FEBS Lett. 2003;536:157–160. [PubMed]
  • PEARCE H.L., SAFA A.R., BACH N.J., WINTER M.A., CIRTAIN M.C., BECK W.T. Essential features of the P-glycoprotein pharmacophore as defined by a series of resurpine analogs that modulate multidrug resistance. Proc. Natl. Acad. Sci. U.S.A. 1989;86:5128–5132. [PMC free article] [PubMed]
  • ROMITI N., TRAMONTI G., CHIELI E. Influence of different chemicals on MDR-1 P-glycoprotein expression and activity in the HK-2 proximal tubular cell line. Toxicol. Appl. Pharmacol. 2002;183:83–91. [PubMed]
  • ROMITI N., TRAMONTI G., DONATI A., CHIELI E. Effects of grapefruit juice on the multidrug transporter P-glycoprotein in the human proximal tubular cell line HK-2. Life Sci. 2004;76:293–302. [PubMed]
  • RYAN M.J., JOHNSON G., KIRK J., FUERSTENBERG S.M., ZAGER R.A., TOROK-STORB B. HK-2: An immortalized proximal tubule epithelial cell line from normal adult human kidney. Kidney Int. 1994;45:48–57. [PubMed]
  • SEELIG A., LANDWOJTOWICZ E. Structure activity relationship of P-glycoprotein substrates and modifiers. Eur. J. Pharm. Sci. 2000;12:31–40. [PubMed]
  • SHAROM F. Probing of conformational changes, catalytic cycle and ABC transporter function ABC Proteins: From Bacteria to Man 2003. New York: Academic Press; 107–133.eds. Cole, S.P.C., Kukler, K. & Higgins, C.F, pp.
  • SMIT J.W., DUIN E., STEEN H., OOSTING R., ROGGEVELD J., MEIJER D.K. Interactions between P-glycoprotein substrates and other cationic drugs at the hepatic excretory level. Br. J. Pharmacol. 1998;123:361–370. [PMC free article] [PubMed]
  • TANIGAWARA Y. Role of P-glycoprotein in drug disposition. Ther. Drug Monit. 2000;22:137–140. [PubMed]
  • TRAMONTI G., ROMITI N., NORPOTH M., CHIELI E. P-glycoprotein in HK-2 proximal tubule cell line. Renal Fail. 2001;23:331–337.
  • UEDA K., OKAMURA N., HIRAI M., TANIGAWARA Y., SAEKI T., KIOKA N., KOMANO T., HORI R. Human P-glycoprotein transports cortisol, aldosterone, and dexamethasone, but not progesterone. J. Biol. Chem. 1992;267:24248–24252. [PubMed]
  • UEDA K., SAEKI T., HIRAI M., TANIGAWARA Y., TANAKA K., OKAMURA N., YASUHARA M., HORI R., INUI K., KOMANO T. Human P-glycoprotein as a multi-drug transporter analyzed by using a transepithelial transport system. J. Physiol. 1994;44:S67–S71.
  • VARMA M.V., ASHOKRAJ Y., DEY C.S., PANCHAGNULA R. P-glycoprotein inhibitors and their screening: a perspective from bioavailability enhancement. Pharmacol. Res. 2003;48:347–359. [PubMed]
  • VEBER D.F., JOHSON S.R., CHENG H.-Y., SMITH B.R., WARD K.W., KOPPLE K.D. Molecular properties that influence the oral bioavailability of drug candidates. J. Med. Chem. 2002;45:2615–2623. [PubMed]
  • YANG C.P., DEPINHO S.G., GREENBERGER L.M., ARCECI R.J., HORWITZ S.B. Progesterone interacts with P-glycoprotein in multi-drug resistant cells and in the endometrium of gravid uterus. J. Biol. Chem. 1989;264:782–788. [PubMed]
  • ZAGER RA. P glycoprotein-mediated cholesterol cycling determines proximal tubular cell viability. Kidney Int. 2001;60:944–956. [PubMed]
  • ZAMORA J.M., PEARCE H.L., BECK W.T. Physical–chemical properties shared by compounds that modulate multidrug resistance in human leukemic cells. Mol. Pharmacol. 1988;33:454–462. [PubMed]
Articles from British Journal of Pharmacology are provided here courtesy of
The British Pharmacological Society

Regulation of Bone Mass, Osteoclast Function, and Ovariectomy-Induced Bone Loss by the Type 2 Cannabinoid Receptor          (full - 2008)

Endocrinology, doi:10.1210/en.2008-0150

Endocrinology Vol. 149, No. 11 5619-5626

Copyright © 2008 by The Endocrine Society

Aymen I. Idris, Antonia Sophocleous, Euphemie Landao-Bassonga, Robert J. van't Hof and Stuart H. Ralston

Rheumatic Diseases Unit, Institute of Genetics and Molecular Medicine, University of Edinburgh, Western General Hospital, Edinburgh EH4 2XU, United Kingdom

Address all correspondence and requests for reprints to: Stuart H. Ralston, Rheumatic Diseases Unit, University of Edinburgh, Western General Hospital, Edinburgh EH4 2XU, United Kingdom. E-mail:



The endocannabinoid system has recently been shown to play a role in the regulation of bone metabolism. The type 2 cannabinoid receptor (CB2) has been reported to regulate bone mass, but conflicting results have been reported with regard to its effects on bone resorption and osteoclast function. Here we investigated the role that CB2 plays in regulating bone mass and osteoclast function using a combination of pharmacological and genetic approaches.

The CB2-selective antagonist/inverse agonist AM630 inhibited osteoclast formation and activity in vitro, whereas the CB2-selective agonists JWH133 and HU308 stimulated osteoclast formation. Osteoclasts generated from CB2 knockout mice (CB2) were resistant to the inhibitory effects of AM630 in vitro, consistent with a CB2-mediated effect.

There was no significant difference in peak bone mass between CB2 mice and wild-type littermates, but after ovariectomy, bone was lost to a greater extent in wild-type compared with CB2–/– mice. Furthermore, AM630 protected against bone loss in wild-type mice, but the effect was blunted in CB2 mice.

We conclude that CB2 regulates osteoclast formation and bone resorption in vitro and that under conditions of increased bone turnover, such as after ovariectomy, CB2 regulates bone loss. These observations indicate that CB2 regulates osteoclast formation and contributes to ovariectomy-induced bone loss and demonstrate that cannabinoid receptor antagonists/inverse agonists may be of value in the treatment of bone diseases characterized by increased osteoclast activity.


 BONE REMODELING IS a complex process that is regulated by an interplay between circulating hormones and locally produced factors that act in a concerted manner to regulate osteoblast and osteoclast activity. Over recent years, there has been increasing interest in the role that the nervous system and neurotransmitters play in the regulation of bone remodeling. Reflecting this fact, the endocannabinoid pathway has recently been implicated as are important regulator of bone turnover and bone mass.

The endocannabinoid system comprises two known receptors [the type 1 (CB1) and type 2 (CB2) cannabinoid receptors], a family of endogenous ligands, and a molecular machinery for ligand synthesis, transport, and inactivation. Cannabinoid receptors belong to the G protein-coupled receptor superfamily and are highly expressed in the brain (CB1), immune system (CB2), and a number of other tissues.

A variety of natural and synthetic ligands have been identified that bind to these receptors. Cannabinoid receptor agonists such as {Delta}9-tetrahydrocannabinol, anandamide, 2 arachinodyl glycerol, CP55,490, and JWH133 bind to the receptors causing inhibition of adenylyl cyclase and activation of ERK kinases and other signaling pathways in target cells. Another class of cannabinoid receptor ligands have been synthesized including AM251, AM630, and SR141716A (rimonabant), which block the effects of cannabinoid receptor agonists.

These compounds were originally referred to as cannabinoid receptor antagonists, but it has now become clear that these compounds cause activation of the receptors in the absence of agonist binding, with opposite effects on downstream signaling cascades as those of the agonists. Accordingly, these compounds are now referred to as cannabinoid receptor antagonists/inverse agonists.

We previously reported that osteoblasts and osteoclasts express cannabinoid receptors and that mice with inactivation of CB1 have increased bone mass and are protected from ovariectomy-induced bone loss. We also found that the CB1-selective antagonist/inverse agonist AM251 inhibited osteoclast differentiation and prevented ovariectomy-induced bone loss in vivo, confirming the importance of CB1 as a regulator of bone resorption. In another study, Tam and colleagues reported that mice with CB1 deficiency on a CD1 background had high bone mass but surprisingly found that mice with CB1 deficiency on a C57BL/6 background had low bone mass.

There is evidence that CB2 also plays a role in regulating bone metabolism. Mice with deficiency of CB2 (CB2–/–) have been reported to develop age-related osteoporosis in association with increased bone turnover.

The CB2-selective agonist HU308 has also been reported to promote osteoblast differentiation, to inhibit osteoclast differentiation, and to protect against ovariectomy-induced bone loss.

On the basis of these observations, it has been suggested that an important function of CB2 is to suppress bone turnover and regulate osteoblast-osteoclast coupling. Contrasting with these findings, however, we and others have reported that pharmacological agonists of cannabinoid receptors stimulate osteoclast formation and bone resorption in vitro in both mouse and human cells. In these studies, however, it was not possible to determine whether the effects were mediated by CB1, CB2, or both receptors.

In view of this, the aim of the present study was to clarify the role that CB2 plays in osteoclast function and ovariectomy-induced bone loss using a combination of pharmacological and genetic approaches.

Materials and Methods

Reagents were obtained from Sigma (Dorset, UK) unless otherwise indicated. The cannabinoid receptor ligands JWH133, AM251, and AM630 were purchased from Tocris Bioscience (Bristol, UK) and HU308 was a kind gift from Dr. R. J. Arends (Organon). The pharmacological properties of the cannabinoid receptor ligands used in the study are summarized in Table 1.

Tissue culture medium was obtained from Invitrogen (Paisley, UK), and primary antibodies were purchased from Cell Signaling Technology (Beverly, MA), unless otherwise stated.

Human IL-1β was obtained from Roche (London, UK); human macrophage colony-stimulating factor (M-CSF) was obtained from R&D Systems (Abingdon, UK); 1,25-(OH)2-vitamin D3 was obtained from Alexis Biochemicals (Nottingham, UK); and mouse receptor activator of nuclear factor-{kappa}B ligand (RANKL) was a gift from Dr. F. P. Ross (Washington University, St. Louis, MO).

 Mouse osteoblast-bone marrow cocultures
Osteoblast-bone marrow cocultures were performed essentially as previously described. Briefly, osteoblasts were isolated from the calvarial bones of 2-d-old mice by sequential collagenase digestion (type I collagenase; Sigma) and cultured in {alpha}-MEM supplemented with 10% fetal calf serum (FCS) and penicillin and streptomycin at 37 C in 5% CO2.

Bone marrow cell populations containing osteoclast precursors were isolated from the long bones of 3- to 5-month-old mice, and erythrocytes were removed by Ficoll Hypaque density gradient centrifugation. The cells were washed with PBS and resuspended in culture medium. Osteoblasts and bone marrow cells were plated together at 10 x 103 cells per well and 2 x 105 cells per well, respectively, on dentine slices in 96-well plates in 150 µl {alpha}-MEM supplemented with 10% FCS, antibiotics, and 10 nM 1,25-(OH)2-vitamin D3 for 7 d. The medium was replaced with the addition of IL-1β (10 U/ml), test compounds were added, and the cultures were terminated on d 8 for actin-ring identification or d 9 for quantification of bone resorption. At the end of the culture period, dentine slices with adherent cells were fixed in 4% paraformaldehyde, washed with PBS, and incubated with phalloidin at 37 C for 30 min to identify actin rings.

The cultures were counterstained for tartrate-resistant acid phosphatase (TRAcP) as previously described, and TRAcP-positive cells with three or more nuclei bearing an actin ring were considered to be active osteoclasts. Cells were then removed from the dentine slices by wiping with tissue paper. Resorption pits were visualized by reflected light microscopy, and the area resorbed was quantified by Image Analysis using custom software developed using Aphelion ActiveX objects (ADCIS, Gif sur Yvette, France) as previously described.

Osteoclast cultures
Osteoclast formation and survival were studied using RANKL- and M-CSF-generated osteoclasts from bone marrow macrophages as described. Briefly, bone marrow cells were flushed from the long bones of 3- to 5-month-old mice, plated onto petri dishes, and incubated for 48 h in the presence of M-CSF (100 ng/ml). Nonadherent erythrocytes were removed, and the adherent cells were washed with PBS and resuspended in culture medium.

The resulting M-CSF-dependent bone marrow macrophages were plated onto either 96-well plates (104 cells per well) or 24-well plates (2 x 105 cells per well) in 150 or 400 µl {alpha}-MEM supplemented with 10% FCS, antibiotics, M-CSF (25 ng/ml), and RANKL (100 ng/ml). Test agents were added on d 5 and the cultures terminated on d 6. The cells were fixed in 4% paraformaldehyde, washed with PBS and stained for TRAcP as described above. TRAcP-positive cells with more than three nuclei were counted as osteoclasts.

Apoptosis and caspase assays
Adherent and nonadherent cells from the osteoclast cultures were collected, fixed with 4% paraformaldehyde, and cytospun onto glass slides. Apoptosis was identified on the basis of characteristic changes in nuclear morphology using 4',6-diamidino-2-phenylindole (DAPI) staining and by terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick end labeling (TUNEL) staining. An average of six microscopic fields per treatment group was analyzed at x200 magnification, and the number of apoptotic cells was quantitated in relation to total cell number.

Levels of inactive (p-30) and activated (p-19) caspase-3 expression were evaluated by Western blotting. Briefly, both adherent and nonadherent cells were washed in PBS and homogenized in lysis buffer (0.1% wt/vol sodium dodecyl sulfate, 0.5% wt/vol sodium deoxycholate in PBS) supplemented with 2% vol/vol protease inhibitor cocktail (Sigma).

The samples were cleared by centrifugation at 12,000 rpm for 10 min at 4 C, and protein concentration was determined using a Bio-Rad (Hercules, CA) protein assay kit. Cell extracts (20 µg/lane) were run on a 10% acrylamide gel and blotted onto a nylon membrane. Activated caspase-3 was detected using a rabbit polyclonal anti-cleaved caspase-3 antibody (1:1000 dilution; Cell Signaling Technology), and inactive caspase-3 was detected using a goat polyclonal anti-caspase-3 antibody (1:100 dilution; Santa Cruz Biotechnology, Inc., Santa Cruz, CA). β-Actin was detected using a mouse monoclonal antibody (1:1000 dilution; Santa Cruz Biotechnology).

Subsequently, the membranes were washed in Tris-buffered saline, incubated with the appropriate secondary antibodies coupled to horseradish peroxidase, and washed in Tris-buffered saline again, and bands were visualized using chemiluminescence (Amersham, Little Chalfont, UK) on a Bio-Rad Fluomax gel imaging system.

Animals and ovariectomy-induced bone loss

Mice with CB2 deficiency were obtained from Dr. Susana Winfield at the National Institutes of Health and were generated as previously described.

These mice have previously been crossed with wild-type C57BL/6 mice for at least 10 generations to create a congenic strain on a C57BL/6 background. To minimize the chance of genetic drift, CB2–/– mice used in this study were generated by mating of heterozygote breeding pairs, and all comparisons between CB2–/– and wild-type mice were performed on littermates.

Animal experiments were approved by the ethical review board of the University of Edinburgh and were conducted in accordance with United Kingdom Home Office regulations. Ovariectomy or sham ovariectomy was performed in 9-wk-old adult female mice. Treatment with test agents was commenced 2 d after ovariectomy or sham ovariectomy by ip administration of the drug in corn oil. Controls received vehicle in corn oil.

The treatment was continued for 19 d and the experiment terminated on d 21. Bone mineral density was measured at the tibial metaphysis by microcomputed tomography (micro-CT), using a Skyscan 1172 scanner. After bone mineral density scanning, the limbs were embedded and processed for bone histomorphometry as previously described.

We also performed some in vitro experiments on osteoclasts generated from mice with CB1 deficiency on an ABH background that were a kind gift from Dr. David Baker (University of London).

Statistical analysis
Comparison between groups was by ANOVA followed by Dunnett’s post test using SPSS for Windows version 11. A P value of ≤0.05 was considered statistically significant.


 Cannabinoid receptor antagonists/inverse agonists inhibit osteoclast formation and bone resorption in osteoblast-bone marrow cocultures
Treatment of mouse osteoblast-bone marrow cocultures for 48 h with the CB2-selective antagonist/inverse agonist AM630 and the CB1-selective antagonist/inverse agonist AM251 inhibited osteoclast formation and bone resorption in a concentration-dependent manner (Fig. 1).

The mean ± SEM concentration of AM630 that half-maximally inhibited osteoclast formation and bone resorption (IC50) was 6.3 ± 0.3 and 7.2 ± 0.1 µM, respectively. Corresponding values for AM251 were 5.9 ± 1.3 and 6.6 ± 0.4 µM (Fig. 1, A–D). Exposure of the cultures to these agents for 24 h also inhibited actin ring formation by about 50% (Fig. 1, E and F). Next, we investigated the effects of AM630, AM251, and the CB2-selective agonists JWH133 and HU308 on RANKL- and M-CSF-induced osteoclast formation in vitro (Fig. 2). We found that AM630 and AM251 both significantly inhibited RANKL-induced osteoclast formation in a concentration-dependent manner with 50% inhibition at values of 330 ± 40 nM for AM630 and 870 ± 100 nM for AM251.

Conversely, JWH133 and HU308 both stimulated osteoclast formation from about 10 nM with maximal stimulation at about 300 nM. At higher concentrations, HU308 inhibited osteoclast formation, whereas JWH133 stimulated osteoclast formation at all concentrations tested (Fig. 2, A and B). We also studied the effect of JWH133 and HU308 on osteoclast fusion by counting the number of osteoclasts with 20 or more nuclei.

As shown in Fig. 2, JWH133 and HU308 significantly increased osteoclast size and nuclearity such that the proportion of cells with more than 20 nuclei rose from about 15% in vehicle-treated cells to 20–25% in JWH133- and HU308-treated cells (Fig. 2B). None of the ligands that we tested significantly affected macrophage growth or viability at concentrations up to 10 µM, excluding a nonspecific toxic effect of the compounds on cell viability (data not shown).

 Figure 1

 FIG. 1. Cannabinoid receptor inverse agonists inhibit osteoclast formation and bone resorption in vitro. A, Osteoclast number in osteoblast bone marrow cocultures exposed to AM630 or AM251 for 48 h; B, representative photomicrographs from the cultures shown in A at low and high power; C, resorbed area in osteoblast-bone marrow cocultures exposed to AM630 or AM251 for 48 h as assessed by reflected light microscopy; D, representative photomicrographs at low and high power from the cultures shown in C; E, osteoclasts with intact actin rings after 24 h exposure to the test substances as visualized by phalloidin staining counted and expressed as a percentage of total osteoclast numbers; F, representative photomicrographs from the cultures are shown in E at low and high power.

The areas of the photomicrographs in the high-power images in B, D, and F are indicated by boxes in the lower-power images. Values in the graphs are means ± SEM and were obtained from three to five independent experiments. *, P < 0.05; **, P < 0.01 vs. vehicle.

 Figure 2

 FIG. 2. Regulation of RANKL-induced osteoclast formation by cannabinoid receptor ligands. A, Mouse bone marrow macrophages were cultured in M-CSF (25 ng/ml) and RANKL (100 ng/ml) for 5 d and then exposed to vehicle, JWH133, AM630, AM251, or HU308 at the concentrations indicated for 24 h. Osteoclast numbers were assessed by counting multinucleated TRAcP-positive cells with three or more nuclei and expressed as a percentage of those in vehicle-treated cultures. B, Number of osteoclasts per well with 20 or more nuclei in the cultures from A. C, Representative photomicrographs from the cultures in A.

Values in the graphs are means ± SEM and were obtained from three to five independent experiments. *, P < 0.01 vs. control cultures; +, P < 0.05 vs. HU308-treated cultures; $, P < 0.05 vs. AM251- and AM630-treated cultures.


 Cannabinoid receptor antagonists/inverse agonists induce osteoclast apoptosis in vitro
To investigate the mechanism of osteoclast inhibition, we studied the effect of AM630 and AM251 on osteoclast apoptosis. These studies showed that both agents induced apoptosis in mature osteoclast at concentrations greater than 1 µM (Fig. 3A) and activated caspase-3 as early as 12 h (Fig. 3C). After 24 h, both agents had induced apoptosis in about 90% of cells, whereas the CB2-selective agonist JWH133 had no significant effect on apoptosis. Representative photomicrographs of osteoclast cultures treated with the cannabinoid receptor ligands JWH133, AM251, and AM630 are shown in Fig. 3B.

 Figure 3

FIG. 3. Cannabinoid receptor inverse agonists induce apoptosis in mouse osteoclast cultures. A, Mouse bone marrow macrophages were cultured in the presence of M-CSF (25 ng/ml) and RANKL (100 ng/ml) for 6 d and then exposed to vehicle, JWH133, AM630, or AM251 for 24 h at the concentrations indicated. Apoptotic cells were visualized by DAPI and TUNEL staining, and cells with fragmented DNA were counted and expressed as a percentage of the total cell number. B, Representative photomicrographs of apoptotic osteoclasts visualized by DAPI and TUNEL staining in the cultures from A. Nuclei from apoptotic cells are stained bright green with the TUNEL assay, and condensed chromatin in the DAPI assay is seen as a bright blue stain (arrows). C, Total cellular protein (50 µg/lane) from the cultures in A was analyzed for expression of β-actin, intact caspase-3 (C3i). and cleaved (active) caspase-3 (C3a) by Western blot analysis. Values in the graphs are means ± SD and were obtained from three independent experiments. **, P < 0.01 vs. control.


Osteoclast inhibition by cannabinoid receptor antagonists/inverse agonists can be mediated through CB1 or CB2
To determine whether the inhibitory effects of the cannabinoid receptor antagonists/inverse agonists were mediated by CB1, CB2, or both receptors, we conducted further studies in M-CSF- and RANKL-generated osteoclasts from CB2 and wild-type littermates that were treated with AM630 and JWH133. In wild-type cultures, we found that JWH133 stimulated osteoclast formation and partially reversed the inhibitory effect of AM630 osteoclast formation at an AM630 concentration of 500 nM.
However, at the higher AM630 concentration of 1000 nM, JWH133 did not significantly rescue the osteoclast-inhibitory effect of AM630 (Fig. 4A). Cultures prepared from CB2–/– mice were resistant to the inhibitory effects of AM630 at a concentrations of 100 nM (Fig. 4B), but at the higher concentration of 300 nM, AM630 inhibited osteoclast formation to a similar extent as in wild-type cultures, although the remaining osteoclasts were larger in CB2–/– cultures (Fig. 4C). In contrast, both concentrations of AM630 tested inhibited osteoclast formation to a similar extent in CB1–/– cultures. Taken together, these data suggest that low concentrations of AM630 inhibit osteoclast differentiation by a CB2-mediated mechanism but that at higher concentrations, selectivity for CB2 is lost and osteoclast inhibition occurs through interaction with CB1 consistent with the results of previous studies.

Figure 4

FIG. 4. Inhibitory effects of AM630 on osteoclast formation are mediated through CB2. A, Mouse bone marrow macrophages from wild-type mice were cultured in the presence of M-CSF (25 ng/ml) and RANKL (100 ng/ml) for 6 d and then exposed to AM630 in the presence of vehicle or JWH133 at a concentration of 1 µM for 24 h at the indicated concentrations. B, Osteoclasts were generated from CB1–/–, CB2–/–, and wild-type littermates as described above and cultured in the presence of AM630 at the concentrations indicated for 24 h. C, Representative photomicrographs from the cultures in B. Values in the graphs are means ± SEM and were obtained from three independent experiments. #, P < 0.05, JWH133 vs. vehicle (A) and wild-type vs. CB2–/– (B); *, P < 0.05 vs. vehicle; **, P < 0.01 vs. vehicle.

 Cannabinoid receptor antagonists/inverse agonists prevents ovariectomy-induced bone loss
To determine whether cannabinoid receptor antagonists/inverse agonists prevented ovariectomy-induced bone loss in vivo, we studied the effects of AM630 on ovariectomy-induced bone loss in wild-type and CB2–/– mice.

Before ovariectomy, there was no significant difference in trabecular bone volume at the tibial metaphysis as assessed by micro-CT in wild-type and CB2–/– mice (Fig. 5, A and B). Analysis of cortical bone and total bone volume similarly showed no difference between the genotypes (data not shown). After ovariectomy, loss of trabecular bone volume (Fig. 5E), trabecular thickness (Fig. 5D), and trabecular number (Fig. 5E) were slightly but significantly less in CB2–/– mice as compared with wild-type littermates.

Treatment of wild-type mice with AM630 at a dose of 0.1 mg/kg·d and 1 mg/kg·d completely prevented bone loss, whereas the CB2–/– mice were resistant to the protective effects of AM630 at 0.1 mg/kg but not at 1.0 mg/kg (Fig. 5F). Bone histomorphometry showed that osteoclast numbers and active resorption surfaces were increased after ovariectomy in vehicle-treated wild-type mice, whereas this was prevented in AM630-treated mice.

Osteoblast numbers increased after ovariectomy as expected but were unaffected by AM630 treatment (Table 2). These data indicate that AM630 inhibits ovariectomy-induced bone loss by interacting with the CB2 receptor at low concentration, but at higher concentrations, selectivity is lost and prevention of bone loss occurs probably by an effect on CB1.

 Figure 5

FIG. 5. Regulation of ovariectomy-induced bone loss by CB2. A, Trabecular bone volume (BV/TV) in wild-type and CB2 mice at age 3 months assessed by micro-CT of the tibia; B, representative micro-CT images from the tibial metaphysis; C, trabecular bone volume (BV/TV) in CB2–/– mice and wild-type littermates subjected to ovariectomy (Ovx) or sham operation at age 9 wk, with changes in BV/TV normalized to those in wild-type sham-operated animals and expressed as percent change; D, trabecular thickness (Tb.Th) from the same experiment, expressed in the same way; E, trabecular number (Tb.N) from the same experiment expressed in the same way; F, changes in BV/TV in CB2–/– and wild-type littermates subjected to ovariectomy or sham ovariectomy and injected daily with AM630 or vehicle at the dose indicated. Values in each panel are means ± SEM from seven to eight mice per group. *, P < 0.05; **, P < 0.01 CB2–/– vs. wild-type; #, P < 0.01, AM630 vs. vehicle.


We have previously reported that CB1 is required for ovariectomy-induced bone loss in mice and that AM251, an antagonist/inverse agonist with selectivity toward CB1, causes osteoclast inhibition and prevents ovariectomy-induced bone loss in mice. We and others also reported that cannabinoid receptor agonists stimulate osteoclast formation and bone resorption in vitro . In contrast, other workers have reported that the CB2-selective agonist HU308 inhibits osteoclast formation in vitro and partially prevents ovariectomy-induced bone loss in vivo. In view of this, the aim of the present study was to further investigate the effects of pharmacological activation and blockade of CB2 on osteoclast differentiation and function in vitro and in vivo in wild-type and CB2–/– mice.

We found that the CB2-selective antagonist/inverse agonist AM630 inhibited osteoclast formation and bone resorption in osteoblast cocultures, with similar potency to the CB1-selective antagonist/inverse agonist AM251. The inhibitory effect in bone resorption may have been due in part to disruption of the actin ring formation, but both agents were also found to inhibit osteoclast differentiation and to cause osteoclast apoptosis in RANKL- and M-CSF-generated osteoclast cultures.

These data therefore suggest that cannabinoid receptor antagonists/inverse agonists inhibit bone resorption mainly by reducing osteoclast formation but indicate that there may also be an inhibitory effect on the ability of mature osteoclasts to form actin rings and adhere to the bone surface.

There was a striking difference in potency in the ability of AM251 and AM630 to cause osteoclast inhibition in RANKL- and M-CSF- generated isolated osteoclast cultures where they were effective in the nanomolar range as opposed to bone marrow osteoblast cocultures where micromolar concentrations were required for osteoclast inhibition. This raises the possibility that cannabinoid receptor antagonists/inverse agonists may influence osteoblast-osteoclast cross talk in a manner that interferes with the direct inhibitory effects of these agents on osteoclasts, but further work will be required to explore the mechanisms responsible.

In keeping with the fact that cannabinoid receptor antagonists/inverse agonists caused osteoclast inhibition, the CB2-selective agonists HU308 and JWH133 significantly increased osteoclast formation at concentrations in the nanomolar range. Although HU308 cause osteoclast inhibition at 10 µM, this is approximately 2000 times greater than the concentration of HU308 required for CB2-mediated adenylyl cyclase inhibition.

This suggests that the inhibitory effects of HU308 at these concentrations may have been nonspecific and mediated by an interaction with pathways other than CB2.

The stimulatory effects of HU308 and JWH133 on osteoclast formation are consistent with previous work that has shown that nonselective cannabinoid receptor agonists including anandamide, 2-arachinodyl glycerol, and CP55,940 stimulate osteoclast formation and bone resorption in vitro at nanomolar concentrations.

The observations reported here differ from those of Ofek and colleagues, who found that HU308 caused osteoclast inhibition at concentrations in the nanomolar range. We cannot readily explain this difference except to note that the observations of Ofek were based in part on studies of RAW 264.7 cells rather than authentic osteoclasts. Although Ofek also studied M-CSF- and RANKL-stimulated bone marrow cultures, the numbers of osteoclasts generated were very low (an average of 15 osteoclasts per culture), and this might also have contributed to the differences observed.

Because the pharmacological ligands currently available are not completely specific and can interact with both CB1 and CB2, further studies were conducted to determine whether the inhibitory effects of AM630 on osteoclast activity were truly mediated by CB2.

Studies in wild-type mice showed that at low concentrations, the osteoclast-inhibitory effect of AM630 was reversed by the CB2-selective agonist JWH133 consistent with a CB2-mediated effect.

At higher concentrations of AM630, however, JWH133 did not reverse the osteoclast inhibition, probably because AM630 is known to act as an inverse agonist of CB1 at high concentrations.

In keeping with this, osteoclasts generated from CB2–/– mice were resistant to the inhibitory effects of AM630 at low concentrations but inhibited osteoclast formation in CB2–/– mice at higher concentrations. In contrast, osteoclasts generated from CB1–/– mice were equally sensitive as wild type to the inhibitory effects of AM630.

To determine whether pharmacological blockade of CB2 can also prevent the bone loss that results from estrogen deficiency, we studied the effects of the CB2-selective antagonist AM630 on ovariectomy-induced bone loss in wild-type and CB2–/– mice.

Administration of AM630 to ovariectomized wild-type mice in a dose of 0.1 and 1 mg/kg·d prevented ovariectomy-induced bone loss, with results virtually identical to those observed with the CB1-selective inverse agonist AM251 as previously reported. Analysis of bone histomorphometry showed that AM630 blocked the increase in osteoclast numbers and active resorption surfaces that followed ovariectomy, demonstrating that prevention of bone loss with AM630 was due to an inhibitory effect on osteoclasts and bone resorption rather than an effect on osteoblasts and bone formation. We found that CB2–/– mice lost slightly less bone than wild-type mice after ovariectomy, indicating that CB2 deficiency may result in a subtle defect in osteoclastic bone resorption.

Treatment with AM630 at a low dose had no significant protective effect on ovariectomy-induced bone loss in CB2–/– mice, but at a higher dose of 1.0 mg/kg·d, AM630 was equally effective at preventing ovariectomy-induced bone loss in CB2–/– mice and wild-type littermates, probably due to an effect on CB1.

In summary, the results reported here indicate that CB2 regulates osteoclast differentiation and function in vitro and ovariectomy-induced bone loss in vivo. We have confirmed that pharmacological inverse agonists of cannabinoid receptors inhibit osteoclast differentiation and bone resorption by a CB2-mediated pathway as well as by interacting with CB1 as previously reported.

Conversely, it appears that the stimulatory effects of cannabinoid receptor agonists on osteoclast formation and bone resorption that we and others have previously observed can also be mediated by an interaction with either or both receptors.

These data suggest that cannabinoid receptor inverse agonists have potential value as antiresorptive drugs that might be of clinical value in the prevention and treatment of diseases such as osteoporosis that are characterized by increased osteoclastic bone resorption.

They also raise the possibility that pharmacological or recreational use of cannabinoid receptor agonists might increase the risk of osteoporosis, by causing increased bone resorption.

This work was supported by grants from the Arthritis Research Campaign (UK) (16304, 15389, 17687, and 17713). A.I.I. is supported by an European Calcified Tissue Society/Amgen fellowship, and A.S. is supported by a Ph.D. Studentship from the University of Edinburgh.

Disclosure Statement: S.H.R., R.V.H., and A.I.I. are inventors on patent applications to protect the use of cannabinoid receptor ligands as treatments for bone disease. A.S. and E.B.-L. have no interests to declare.

First Published Online July 17, 2008

Abbreviations: CB1, Type 1 cannabinoid receptor; CB2, type 2 cannabinoid receptor; DAPI, 4',6-diamidino-2-phenylindole; FCS, fetal calf serum; M-CSF, macrophage colony-stimulating factor; micro-CT, microcomputed tomography; RANKL, receptor activator of nuclear factor-kB ligand; TUNEL, terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick end labeling.


Khosla S 2001 The OPG/RANKL/RANK system. Endocrinology 142:5050–5055

Idris AI, Van't Hof RJ, Greig IR, Ridge SA, Baker D, Ross RA, Ralston SH 2005 Regulation of bone mass, bone loss and osteoclast activity by cannabinoid receptors. Nat Med 11:774–779

Ofek O, Karsak M, Leclerc N, Fogel M, Frenkel B, Wright K, Tam J, Attar-Namdar M, Kram V, Shohami E, Mechoulam R, Zimmer A, Bab I 2006 Peripheral cannabinoid receptor, CB2, regulates bone mass. Proc Natl Acad Sci USA 103:696–7

Tam J, Ofek O, Fride E, Ledent C, Gabet Y, Muller R, Zimmer A, Mackie K, Mechoulam R, Shohami E, Bab I 2006 Involvement of neuronal cannabinoid receptor, CB1, in regulation of bone mass and bone remodeling. Mol Pharmacol 70:786–792

Tam J, Trembovler V, Di M, V, Petrosino S, Leo G, Alexandrovich A, Regev E, Casap N, Shteyer A, Ledent C, Karsak M, Zimmer A, Mechoulam R, Yirmiya R, Shohami E, Bab I 2008 The cannabinoid CB1 receptor regulates bone formation by modulating adrenergic signaling. FASEB J 22:285–294[Abstract/Free Full Text]






The preventive effect of cannabinoids on reperfusion-induced ischemia of mouse kidney (abstact - 2008)

Exp Toxicol Pathol. 2008 Aug;60(4-5):405-10. Epub 2008 Jun 20.


Department of Pathology and Molecular Biology, School of Medicine, Zanjan University of Medical Sciences, Zanjan, Iran.


Artery occlusion of an organ results in ischemia. When the occlusion is opened and blood flow reinstated there will be tissue injuries identified as reperfusion-induced ischemia (RII).

It has been suggested that cannabinoids (CBs) may be involved in the RII. In this study, we assessed the effect of different doses of anandamide analogs and CB receptor agonists: arachidonylcyclopropylamide (ACPA, a CB1 agonist) and JWH133 (a CB2 agonist) in the RII of the mouse kidney. Three doses (0.2, 1 and 5mg/kg, i.p.) of ACPA or JWH133 were used 30min prior initiation of RII.

Kidneys were removed 2 and 24h following RII and checked histologically for the grading of ischemic injury. Appropriate control groups were used as well. RII produced lesion comparable with that of ischemia. Different doses of ACPA or JWH133 prevented RII-induced lesions. It is suggestive of the CB system involvement in the kidney RII in mice.

[PubMed - indexed for MEDLINE]



Ajulemic acid, a synthetic cannabinoid, increases formation of the endogenous proresolving and anti-inflammatory eicosanoid, lipoxin A4 (full - 2009)

  1. Robert B. Zurier
  2. Yee-Ping Sun
  3. Kerri L. George
  4. Judith A. Stebulis
  5. Ronald G. Rossetti
  6. Ann Skulas
  7. Erica Judge 
  8. Charles N. Serhan

+ Author Affiliations

  1. *University of Massachusetts Medical School, Division of Rheumatology, Worcester, Massachusetts, USA; and
  2. Center for Experimental Therapeutics and Reperfusion Injury, Department of Anesthesia Perioperative and Pain Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts, USA
  1. 1 Correspondence: University of Massachusetts Medical School, Division of Rheumatology, 55 Lake Ave, Worcester, MA 01655, USA. E-mail:


Ajulemic acid (AjA), a synthetic nonpsychoactive cannabinoid, and lipoxin A4 (LXA4), an eicosanoid formed from sequential actions of 5- and 15-lipoxygenases (LOX), facilitate resolution of inflammation.

The purpose of this study was to determine whether the ability of AjA to limit the progress of inflammation might relate to an increase in LXA4, a known anti-inflammatory and proresolving mediator.

Addition of AjA (0–30 μM) in vitro to human blood and synovial cells increased production of LXA4 (ELISA) 2- to 5-fold. Administration of AjA to mice with peritonitis resulted in a 25–75% reduction of cells invading the peritoneum, and a 7-fold increase in LXA4 identified by mass spectrometry.

Blockade of 12/15 LOX, which leads to LXA4 synthesis via 15-HETE production, reduced (>90%) the ability of AjA to enhance production of LXA4 in vitro. These results suggest that AjA and other agents that increase endogenous compounds that facilitate resolution of inflammation may be useful for conditions characterized by inflammation and tissue injury.—Zurier, R. B., Sun, Y.-P., George, K. L., Stebulis, J. A., Rossetti, R. G., Skulas, A., Judge, E., Serhan, C. N. Ajulemic acid, a synthetic cannabinoid, increases formation of the endogenous proresolving and anti-inflammatory eicosanoid, lipoxin A4.

The cannabis plant has been a source of medicinal preparations since the earliest written records on pharmacobotany. A major obstacle to acceptance of the drug has been its potent psychoactivity.

This problem has been studied in recent years in attempts to discover synthetic analogs that would retain medicinal properties without the psychotropic effects.

A class of cannabinoid, the carboxyl tetrahydrocannabinols, shows potential as therapy that is free from cannabimimetic central nervous system activity.

These substances, which are metabolites of tetrahydrocannabinol (THC), the psychoactive principle of Cannabis, do not produce behavioral changes in humans at doses several times greater than THC doses given to the same volunteers.

The parent compound in this series, the THC metabolite THC-11-oic acid (Fig. 1 ), is effective in animal models of inflammation and pain at oral doses of 20–40 mg/kg. However, more potent activity is needed for clinical use.

Figure 1.

Figure 1.

Cannabinoid structures. A) Δ9-THC (tetrahydrocannabinol) is the component of Cannabis responsible for psychoactivity. B) THC-11-oic acid is a nonpsychoactive metabolite of Δ9 THC. C) AjA (1′,1′-dimethylheptyl THC-11-oic acid) is a nonpsychoactive synthetic analog of THC-11-oic acid.

It has been known for some time that modifications of the pentyl side chain of THC increase its potency.

In particular, extending the chain length to 7 carbons and introducing branching close to the ring leads to compounds with potencies that are 50–100 times greater than that of THC.

This strategy was employed in designing the structure of 1′,1′-dimethylheptyl-THC-11-oic-acid [trivial name ajulemic acid (AjA)] (5). This synthetic analog of THC-11-oic acid is a potent anti-inflammatory and analgesic agent in several animal models .

We have shown (8)that oral administration of AjA at a dose of 0.1 mg/kg 3×/wk reduces significantly the severity of adjuvant-induced olyarthritis in rats.

Histomorphological evaluation of the joints suggested that although synovial inflammation occurred in AjA-treated animals, it did not progress to cartilage degradation, bone erosion, and distortion of joint architecture as observed in rats given placebo. Thus, AjA treatment appears to facilitate resolution of inflammation in this animal model of joint tissue injury.

Lipoxins are a new class of eicosanoids that arise from the sequential actions of lipoxygenases. Because these compounds are generated through an interaction between lipoxygenase pathways, the term lipoxins (lipoxygenase interaction products) was introduced.

In humans, lipoxins are formed in vivo during multicellular responses such as inflammation. They serve as stop signals in that they prevent leukocyte-mediated tissue injury and stimulate the uptake of apoptotic polymorphonuclear leukocytes (PMNs) at sites of inflammation, thereby facilitating resolution of inflammation. A major problem in joints of patients with rheumatoid arthritis is that inflammation—designed to be protective—often does not resolve. Stable analogs of lipoxin A4 (LXA4) block chemotaxis of PMNs and reduce inflammation in animal models. As noted, AjA treatment of rats with adjuvant arthritis appears to promote resolution of synovial inflammation. Therefore, we designed studies to determine whether AjA—added to cells in vitro or administered to mice in vivo—is associated with an increase in LXA4.

We present results here that indicate that AjA does stimulate production of LXA4, a process associated with counterregulation of an inflammatory response in a murine model of peritoneal inflammation.



AjA was obtained from Organix (Woburn, MA, USA). Its purity was monitored on high-pressure liquid chromatography by comparison with material synthesized previously.

The sample was 97% chemically pure and was >99% chiral pure in the R,R enantiomer. TNFα was from R&D Systems (Minneapolis, MN, USA). Zymosan A, media, and all other reagents were from Sigma Chemical Co. (St. Louis, MO, USA). AjA was dissolved in dimethyl sulfoxide (DMSO), then diluted with minimal essential medium (MEM) and 2% fetal bovine serum (FBS) to achieve appropriate concentrations. The concentration of DMSO was kept constant at 0.3%.


Human peripheral blood (60 ml) obtained by venepuncture (after informed consent approved by the University of Massachusetts Medical School Committee on Protection of Human Subjects), was mixed with 60 ml phosphate-buffered saline (PBS). Aliquots of 1 ml were used for experiments. Samples from separate donors were run in triplicate.

PMNs were obtained from the red blood cell pellet after density-gradient centrifugation of whole blood using histopaque (Sigma). After removing the plasma layer, the peripheral blood mononuclear cell (PBMC) layer, and the histopaque layer, red blood cells were lysed by adding 5 ml cold distilled water. PMN pellets were collected after centrifuging extracts for 10 min at 900 g.

Fibroblast-like synovial cells (FLSs) were from synovial fluid, as we have described.

Synovial fluid was aspirated from joints of patients with rheumatoid arthritis, inflammatory polyarthritis, or osteoarthritis.

Fluid was collected in heparinized syringes, then centrifuged at 300 g for 15 min. The resulting pellets were suspended in 7 ml of MEM with 15% heat-inactivated FBS, 1% nonessential amino acids, and 1% penicillin/streptomycin solution, and were plated in 25-ml tissue culture flasks. Cultures were incubated at 37°C with 5% CO2 for 24–48 h, after which medium was aspirated and cultures were washed with PBS to remove nonadherent cells. Growth medium was replaced every 3 to 4 days. After 10 to 14 days, adherent cells were removed from flasks by trypsinization, washed, and transferred to 6-well tissue culture plates in fresh medium. FLSs were passaged (split 1:3) when they reached confluence, generally at 11 to 14 days. Passages 2 through 6 were used for experiments.

Zymosan A particles were suspended in PBS at 10 mg/ml. Serum (2%) was added, and the zymosan suspension was incubated for 1 h at 37°C, then washed once with PBS. Zymosan was then resuspended in serum-free RPMI at the desired concentration.

Cell viability

The integrity of cell membranes was assessed by Trypan blue exclusion. Viability was >95% in all experiments.

Murine peritonitis

Peritonitis was induced in FVB male mice, 6 to 8 wk of age (Charles River Laboratories, Durham, NC, USA), that were fed laboratory Rodent Diet 5001 (Purina Mills, St. Louis, MO, USA). After anesthesia with isoflurane, AjA (1.5 mg/kg), or vehicle was administered in 100 μl PBS intravenously through the tail vein. In other experiments, AjA in safflower oil, or safflower oil alone, was given by mouth for 3 days to male CD-1 mice (6–8 wk old) before Zymosan A (1 mg/ml PBS) was injected into the peritoneal space.

In accordance with the Harvard Medical Area Standing Committee of Animals protocol No. 02570 and the University of Massachusetts Medical School Animal Research Review Group, mice were sacrificed after 2–3 h, and peritoneal lavages were collected in Dulbecco’s PBS (minus magnesium and calcium).

Aliquots of lavage were stained with Trypan blue, and cells were counted using light microscopy.

For differential leukocyte counts, 300 μl of the lavage was added to 300 μl of 15% bovine serum albumin and centrifuged onto microscope slides at 2200 rpm for 4 min using a Cytofuge (Statspin, Norwood, MA). Slides were allowed to air dry, and cells were visualized using a modified Wright-Giemsa stain (Sigma).

Lipoxin A4

Samples were diluted 1:2 with methanol, then acidified to pH 3.5 with 1 N hydrochloric acid, then centrifuged (8000 g for 15 min). Supernatants were then applied to Supelco C18 minicolumns (Supelco, Bellfonte, PA, USA); columns were washed with water, then with 1 ml hexane, and lipoxin was eluted with methyl formate.

The column was filled with 99% methyl formate, capped, placed in a 15-ml centrifuge tube, and centrifuged 30 min at 1500 g. Extracted samples were dried under nitrogen, resuspended in new extraction buffer, and stored at −80°C until assay. LXA4 was monitored by ELISA (Neogen Corporation, Lansing, MI, USA). The sensitivity of the assay is ∼10 pg/ml.

Liquid chromatography tandem mass spectrometry (LC/MS/MS) identification of LXA4

Peritoneal lavages were collected, and 2 vol of methanol was added. Samples were then extracted with solid-phase C18 cartridges.

The resulting methyl formate eluants were taken to dryness with a stream of nitrogen and resuspended in methanol (100 μl) in preparation for LC/MS/MS analysis. Samples were analyzed using liquid chromatography-photodiode array detector (PDA)-tandem mass spectrometry (LC_PDA_MS/MS) (ThermoFinnigan, San Jose, CA, USA). Reverse-phase LC was conducted with a LUNA C18–2 (100×2×5 mm) column.

The column was kept in a column heater (30°C). The LC system used a P-4000 quaternary LC pump (ThermoFinnigan). The column was eluted at a flow rate of 0.2 ml/min with methanol:water:acetic acid (65:35:0.01, v/v/v) from 0 to 10 min.

Statistical analysis

Changes in LXA4 production and in cell counts were analyzed for significance by a one-tailed Student’s t test for groups with equal variance. For experiments in which it was necessary to control for variability among groups, significance was assessed by an analysis of covariance (ANCOVA). In both cases, a value of P < 0.05 was considered significant.


AjA increased LXA4 production by TNFα stimulated PMNs in a dose-dependent manner (Fig. 2A ). In a series of 4 experiments, 30 μM AjA increased LXA4 release from TNFα-stimulated PMNs 2.60 ± 0.35-fold (mean±sd); P = 0.04 vs. untreated control cells.

Figure 2.

Figure 2.

AjA stimulates LXA4 production by human cells. A) PMNs (3×106) incubated with vehicle alone (0 AjA) or treated 1 h with AjA and stimulated 4 h with 10 ng/ml rhTNFα. Supernatant LXA4 determined by ELISA. Values are means of 6 replicates; error bars represent sd. *P < 0.04 vs. 0 AjA; Student’s t test (n=6). B) Peripheral blood incubated without or with AjA (3, 10, 30 μM) for 5 h. Supernatant LXA4 concentration measured by ELISA. Values are means of 5 experiments with blood from 5 separate donors; error bars are sd. *P < 0.05 vs. 0 AjA; ANCOVA (3 and 10 μM, n=4; baseline and 30 μM, n=5).

Optimal generation of LXA4 requires transcellular biosynthesis. The addition of AjA to whole blood in vitro also enhanced production of LXA4 in a dose-dependent manner (Fig. 2B ). In cells in which LXA4 was increased maximally by zymosan, LXA4 was not increased substantially more by AjA (not shown). The observations indicate that AjA is by itself an agonist for LXA4. Although the ratio of AjA-induced LXA4 to baseline LXA4 was similar across experiments, the absolute amounts of LXA4 induced by stimulation of cells by zymosan and/or by AjA varied widely from experiment to experiment. Therefore, results from 3 experiments presented in Fig. 2B were assessed by ANCOVA.

Although joint tissue injury in patients with rheumatoid arthritis is likely due to a multicellular assault on cartilage and bone, studies in animals and humans suggest that joint damage can proceed with participation of synovial cells alone.

Therefore, we examined the influence of AjA on LXA4 production by 4 different cultures of human FLSs. We have not observed differences in responses of FLSs to AjA whether cells were derived from patients with inflammatory polyarthritis (rheumatoid, psoriatic) or osteoarthritis. Addition of AjA to FLSs increased LXA4 release from unstimulated and TNFα (1 ng/ml)-stimulated cells in a concentration-dependent manner (Fig. 3 ). In these experiments, TNFα did not stimulate LXA4 maximally (not shown). Thus, AjA did stimulate further LXA4 production in TNFα-stimulated cells.

Figure 3.

Figure 3.

AjA stimulates LXA4 production by human FLSs. FLSs were incubated with vehicle (0 AjA) or AjA (10, 20, 30 μM) for 1 h, then stimulated with 1 ng/ml TNFα for 18 h. Supernatant LXA4 determined by ELISA. Values are means of 4 experiments; error bars are sd. *P < 0.05 vs. 0 AjA; ANCOVA (n=4).

We next assessed the actions of AjA in an in vivo murine model of acute inflammation. Results from these experiments in which 6 CD-1 mice were administered 10 mg/kg/day AjA by mouth for 3 days (total daily dose ∼0.3 mg) before intraperitoneal injection of zymosan are shown in Fig. 4 . AjA treatment reduced the total number of cells invading the peritoneum by 69%.

Figure 4.

Figure 4.

Treatment of murine peritonitis with oral administration of AjA. Male CD-1 mice were treated by mouth with 10 mg/kg/day AjA or with safflower oil (vehicle control) for 3 days before i.p. injection of 1 mg zymosan A.

Peritoneal fluid obtained 3 h after zymosan installation. Values are means of total cell counts determined by light microscopy; error bars are sd. *P = 0.03 vs. untreated controls; ANCOVA, two separate experiments, 3 mice/group (n=6). NS, no stimulation.

Having shown an induction of LXA4 by AjA in vitro, we next examined whether LXA4 generation was also increased in vivo. To this end, we analyzed LXA4 in peritoneal lavages from 4 AjA-treated animals by LC/MS/MS and compared these with lavages from 5 mice that received vehicle. Complete MS/MS analysis of the fraction at 7.2 min resulted in a fragmentation pattern consistent with that for LXA (Fig. 5A ).

The prominent MS/MS product ions that are diagnostic for LXA4 were noted at m/z 351 ([M-H]) (Fig. 5A, a ), m/z 333 ([M-H]–H2O), m/z 315 ([M-H]–2H2O), m/z 307 ([M-H]–CO2 (Fig. 5A, b ), m/z 289 ([M-H]–H2O–CO2), m/z 271 ([M-H]–2H2O–CO2), m/z 251 ([M-H]–CHO(CH2)4CH3) (Fig. 5A, d ), m/z 233 ([M-H]–CHO(CH2)4CH3–H2O), m/z 207 ([M-H]–CHO(CH2)4CH3–CO2), m/z 189 ([M-H]–CHO(CH2)4CH3–H2O–CO2), m/z 235 ([M-H]–CHO(CH2)3COOH) (Fig. 5A, c ), and m/z 115 (CHO(CH2)3COOH) (Fig. 5A, c′ ).

Having confirmed the identity of the material eluted at 7.2 min as LXA4 via MS/MS identification of diagnostic ions, it was then determined that this material represented ∼0.5 ng of LXA4 (Fig. 5A ). When compared to baseline levels of 0.06 ng/mouse obtained in the mice who received vehicle only, this represented a 7.33-fold increase in LXA4 generation (Fig. 5B ).

AjA treatment reduced total cell counts in peritoneal lavages but did not reduce numbers of mononuclear cells significantly (not shown). The reduced total cell count was due to a reduction in the PMN counts in peritoneal lavages from AjA-treated mice (Fig. 5C ). Taken together, these results demonstrate that LXA4 generation is associated with reduction of PMN infiltration by AjA in vivo.

Figure 5.

Figure 5.

AjA stimulation of endogenous LXA4 generation and reduction of peritoneal infiltration of PMNs in zymosan-A-induced peritonitis. AjA (1.5 mg/kg/100 μl sterile saline) was injected into tail vein of 6- to 8-wk-old male FVB mice, followed by i.p. injection of 1 mg zymosan A.

Peritoneal lavages were collected, volumes were measured, and 2 vol of methanol was added. Following solid-phase extraction, samples were then analyzed using LC/MS/MS. Aliquots of lavages were assessed for PMN concentration as described in Materials and Methods. A) MS/MS spectrum of AjA-treated samples.

Chromatogram spectra are representative of n = 4; a–d and c′ denote prominent diagnostic ions used for identification (see inset). a, ([M-H]) = m/z 351; b, ([M-H]–CO2) = m/z 307; c, ([M-H]–CHO(CH2)3COOH) = m/z 235; c′, (CHO(CH2)3COOH) = m/z 115; d, ([M-H]—CHO(CH2)4CH3) = m/z 215. B) LXA4 generation quantified in AjA-treated samples vs. control samples. Values represent LXA4 generation per mouse. Samples were combined before quantitation to allow for adequate detection. C) PMN concentrations quantified in peritoneal lavages from mice treated with AjA vs. untreated controls. Values are means; error bars are sd. *P < 0.05 vs. untreated controls; Student’s t test (n=4).

In an effort to determine which biosynthetic pathway might be influenced by AjA, we treated human whole blood with inhibitors of cytoplasmic phospholipase A2 (cPLA2), and 12/15 lipoxygenase (12/15 LOX). Results suggest that most of the increase in LXA4 initiated by AjA resulted from the conversion of endogenous arachidonic acid to LXA4 by the 15 LOX-initiated route. The addition to cells of the 12/15 LOX inhibitor baicalein blocked the AjA-induced increase in LXA4 substantially (Fig. 6 ). It is likely that AjA stimulated the release of arachidonic acid from phospholipid precursors in both murine exudate cells and human whole blood. However, the addition to mouse peritoneal exudate cells or human whole blood of the cPLA2 inhibitor methylarachidonylfluorophosphate did not alter the AjA-induced increase in LXA4 (data not shown), thus precluding our assessment of specific phospholipases in the process.

Figure 6.

Figure 6.

Blockade of AjA enhancement of LXA4. Human blood was incubated with AjA for 1 h, and LXA4 was assessed as described in Materials and Methods. AjA (25 μM)-induced production of LXA4 was reduced significantly when the 12/15 LOX inhibitor baicalein was added to cells before incubation with AjA. Values are means; error bars represent sd. *P < 0.05 vs. AjA alone; Student’s t test (n=3).


Results of experiments presented in this report indicate that addition of AjA in vitro to human PMNs, whole blood, or FLSs increases production of LXA4 by these cells. AjA itself is an agonist for LXA4 generation and can stimulate further LXA4 production in activated cells. In addition, administration of AjA in vivo increases LXA4 in and reduces total cellular and PMN infiltration into peritoneal fluid in murine peritonitis, a time-honored model for acute inflammation. The anti-inflammatory actions of LXA4 are well documented. It blocks neutrophil activation and antagonizes peptido-leukotrienes. In addition, in a murine model of inflammatory ear edema, LXA4 analogs are more potent anti-inflammatory agents than equimolar concentrations of dexamethasone.

Thus, maintaining plasma or tissue levels of AjA should provide a way to limit tissue injury during states of mild chronic inflammation.

For example, in a model of chronic inflammation (adjuvant-induced arthritis in rats), oral administration of 0.1 mg/kg AjA 3×/wk for 1 mo prevented joint tissue injury.

In the short-term studies of acute inflammation presented in this article, mice were treated with an oral dose of 10 mg/kg AjA for 3 days or 1.5 mg/kg i.v. once.

Use of higher doses of AjA used in these experiments does not preclude the potential clinical use of AjA. Although dose translation from rodent to human is tenuous at best, using rodent body surface area as described by Reagan-Shaw et al. for converting the 10 mg/kg/day dose of AjA in mice, yields an equivalent dose of 56 mg/day for a 70-kg human.

In the only published clinical trial of AjA , administration of 40 and 80 mg/day AjA for 7 days reduced neuropathic pain without induction of cannabimimietic effects.

Although AjA does increase release of arachidonic acid from cell membranes, the cPLA2 inhibitor methylarachidonylfluorophosphate did not block AjA-induced increases in LXA4. Thus, either the inhibitor did not block cPLA2 adequately, or sufficient arachidonate was available by virtue of cell activation and/or the action of other phospholipases.

For example, THC-induced release of arachidonic acid from mouse peritoneal cells is mediated by phospholipase D activity. Blockade of 12/15 LOX, which converts endogenous arachidonic acid to LXA4 by the 15 LOX route, does interfere with the ability of AjA to enhance production of LXA4. We have also observedthat AjA increases expression of cyclooxygenase 2, leading to an increase in prostaglandin J2 (PGJ2), presumably by an increase in PGD synthase. Both PGD2 and PGJ2, like LXA4, facilitate resolution of inflammation.

Efforts to limit tissue injury by facilitating resolution of inflammation in, for example, a disease like rheumatoid arthritis, must include induction of naturally occurring proresolving agonists such as LXA4 by stromal cells, such as fibroblasts, at the site of inflammation. We show here that AjA increases LXA4 secretion from human FLSs derived from patients with inflammatory arthritis.

The amounts of LXA4 generated by the FLSs in culture are small in these experiments. However, human synovial fibroblasts exhibit functional LXA4 receptors, and nanomolar concentrations of LXA4 block IL-1β-induced production by these cells of the inflammatory cytokines IL-6 and IL-8, and of matrix metalloproteinases.

Therefore, it is of interest that AjA also reduces production of IL-1β from human peripheral blood and synovial fluid monocytes, of IL-6 from human monocyte-derived macrophages, and of matrix metalloproteinases from human synovial cells .

A major defect in patients with diseases characterized by chronic inflammation is a lack of physiological resolution of inflammation. Given that inflammation is a primitive protective response and given the hard-earned knowledge that suppression of inflammation can be freighted with adverse events, it makes sense to facilitate the next step in this physiological process: resolutio. As noted, both AjA and LXA4 facilitate resolution of inflammation. Our observation (Fig. 6) that inhibition of 12/15 LOX blocks the AjA-induced increase in LXA4 suggests that most of the increase results from the conversion of endogenous arachidonic acid to LXA4 by the 15-lipoxygenase initiated route. Whatever the precise mechanisms whereby AjA increases LXA4, it is clear that reduction by AjA of acute inflammation in a murine model of peritonitis is associated with a concomitant increase in endogenous LXA4.

It has long been known that not all products of the cyclooxygenase pathway are mediators of inflammation. Indeed, several of the eicosanoids modulate inflammation, and prostaglandin E1 (PGE1) and prostaglandin J2 (PGJ2), for example, function as anti-inflammatory agents. Similarly, by virtue of limiting PMN infiltration to the inflamed site, and enhancing clearance of apoptotic PMNs by macrophages, LXA4 hastens resolution of inflammation. LXA4, generated in rat kidneys during experimental immune complex-mediated glomerulonephritis, antagonizes the tissue-injuring actions of leukotrienes in this animal model. In addition, an LXA4 analog reduces expression of interferon gamma-induced genes associated with nephritis.

Thus, development of analogs of LXA4 and of compounds that increase endogenous LXA4 production may prove useful as therapeutic agents for diseases characterized by chronic inflammation and related tissue injury.


This study was supported by U.S. National Institutes of Health (NIH) grants DA 3691 and AI 1056362 (R.B.Z.) and GM38765 and P50-DE016191 (C.N.S.). E.J. was supported by NIH 5T35HL00771 (Short-Term Training for Minority Students). We thank Dr. Sumner Burstein for helpful discussion of the manuscript.

  • Received July 24, 2008.
  • Accepted November 26, 2008.


  1. Sebulis, J., Johnson, D. R., Rossetti, R. G., Burstein, S. H., Zurier, R. B. (2008) Ajulemic acid, a synthetic cannabinoid acid, induces an antiinflammatory profile of eicosanoids in human synovial cells. Life Sci. 83,666-670



Cream with endocannabinoids effective in the treatment of pruritus due to kidney disease (news 2005)

Researchers of the University of Wroclaw, Poland, investigated the effects of an ointment with structured physiological lipids and endocannabinoids in 21 patients with pruritus due to end-stage failure of kidney function. So-called uremic pruritus is still a common symptom in patients with end-stage renal failure.

However, there is no effective treatment for this condition. All patients applied the tested cream twice daily for a period of three weeks. Global pruritus and dry skin were examined before the trial, on study visits at weekly intervals, and two weeks after completion of the study.

After 3-week therapy pruritus was completely eliminated in 8 patients. Dry skin was significantly improved. Researchers noted that "it is very probable that the observed decrease of pruritus with the test product therapy was not only the result of dry skin improvement but that the addition of endocannabinoids may have also played a role."

(Source: Szepietowski JC, Szepietowski T, Reich A. Efficacy and tolerance of the cream containing structured physiological lipids with endocannabinoids in the treatment of uremic pruritus: a preliminary study. Acta Dermatovenerol Croat 2005;13(2):97-103.)