Introduction

Two types of G-protein coupled cannabinoid receptors, CB1 and CB2, have been well characterized (Pertwee 2006), and although the healthy nervous system is richly endowed with CB1 receptors and appears to be relatively devoid of CB2 receptors, CB2 receptors predominate in the immune system (Cabral and Dove Pettit 1998; Klein et al. 2003). The immunosuppressive effects on T lymphocytes and macrophages by Δ9-tetrahydrocannabinol, the main psychotropic constituent of cannabis (marijuana), appear to be largely related to its activity at CB2 receptors (Cabral and Dove Pettit 1998; Klein et al. 2003). The consequences of these immunological impairments have been shown in animal models to promote the pathogenesis of infections caused by intracellular microorganisms (Cabral and Dove Pettit 1998; Klein et al. 2003).

Our laboratory has been interested in the effects of substances of abuse on the expression of HIV-1 in microglia, which are the major cell type productively infected by this virus in the central nervous system. In a recent study, we found that in contrast to morphine, which up-regulated HIV-1 expression, the synthetic nonselective CB1/CB2 receptor agonist WIN55,212-2 inhibited viral replication (Peterson et al. 2004).

The purpose of the present study was to characterize the cannabinoid receptor types present in human fetal brain-derived microglia and to determine which cannabinoid receptor type is involved in the anti-HIV-1 activity of WIN55,212-2. Based on literature indicating the primacy of CB2 receptors in cannabinoid-mediated immunomodulation and data that WIN55,212-2 has some degree of selectivity for CB2 receptors (Pertwee 1997), we hypothesized that CB2 receptors would be the main target of WIN55,212-2. Although support for this hypothesis was found, the results of this study also demonstrated that the CB1-selective antagonists SR141716A and AM-251 behave as agonists in microglia.

Materials and methods

Reagents

The following reagents were purchased from the source as indicated: Dulbecco’s modified Eagle’s medium (DMEM), Hanks’ balanced salt solutions (HBSS), trypsin, penicillin, streptomycin, and WIN55,212-2 [(R)-(+)-[2,3-dihydro-5-methyl-3-(4-morpholinylmethyl)] pyrolo-[1,2,3-de]-1,4-benzoxazin-6-yl]-1-naphthalenylmethanone mesylate] (Sigma, St. Louis, MO, USA); fetal bovine serum (FBS) (Hyclone, Logan, UT, USA); CP55,940 [(−)-cis-3R-[2-Hydroxy-4-(1,1-demethylheptyl)phenyl]-trans 4R-3(3-hydroxypropyl)-1R-cyclohexanol], were kindly provided by Raj K. Razdan (Organix, Inc.); WIN55,212-3 [S(−)-[2,3-dihydro-5-methyl-3-(4-morpholinylmethyl)pyrolo[1,2,3-de]-1,4-benzoxazin-6-yl]-1-naphthalenyl-methanone mesylate], SR141716A [N-(piperidin-1-yl)-5-(4-chlorophenyl)-1-(2,4-dichlorphenyl)-4-methyl-1H-pyrozole-3-carboxamidehydrochloride], and SR144528 [N-(1S)-endo-1,3,3-trimethylbicyclo[2.2.1]heptan-2-yl]5-(4-chloro-3-methylphenyl)-1-(4-methylbenzyl)-pyrazole-3-carboxamide)] were kindly provided by S. Thayer, University of Minnesota; AM-251 [N-(piperdin-1-yl)-5-(4-iodophenyl)-1-(2,4-dichlorophenyl)-4-methyl-1H-pyrazole-3-carboxamide] (Tocris Bioscience, Ellisville, MO, USA); JWH 0-15 [(2-methyl-1-propyl-1H-indol-3-yl)-1-naphthalenylmethanone] was kindly provided by Billy Martin, Virginia Commonwealth University.

Microglial cell cultures

Highly enriched cultures of human microglia were isolated from fetal brain tissue specimens by using a protocol approved by the Human Subject Research Committee at our institution as previously described (Peterson et al. 2004). Briefly, brain tissues were dissociated after trypsinization (0.25%) and plated into 75-cm2 Falcon culture flasks in DMEM containing 10% FBS, penicillin (10 U/ml), and streptomycin (100 U/ml). The medium was replenished 4 days later. Microglia were harvested after 10–14 days with purity of >99% (stained by anti-CD68 antibody).

Drug treatment and viral infection

Microglial cultures (1 × 105 cells/well in a 48-well plate) were pretreated without (control) or with CB1/CB2 receptor agonists (WIN55,212-2 and CP55,940), CB2-selective agonist JWH 0-15, CB1-selective antagonists SR141716A and AM-251, CB2-selective antagonist SR144528, or the inactive enantiomer WIN55,212-3 (Savinainen et al. 2005) for 3 h prior to HIV-1 infection. In one experiment SR144528 (a CB2-selective antagonist) (Howlett et al. 2002) was added 30 min prior to treatment with CP55,940 or WIN55,212-2. All CB receptor ligands were constituted in DMSO with the exception of CP55,940, which was constituted in ethanol. Untreated (control) cultures contained equivalent concentrations of these diluents.

HIV-1SF162, a monocytotropic (R5) variant, was obtained from the NIH AIDS Research and Reference Reagent Program (National Institute of Allergy and Infectious Diseases, Bethesda, MD) and was added to microglia at a multiplicity of infection (MOI) of 0.02. After 24 h of absorption with HIV-1 at 37°C, cannabinoid-treated and untreated control cells were washed three times and resuspended in culture medium (DMEM, 10% FBS) containing the indicated concentrations of drugs or drug combinations.

Assessment of HIV-1 expression

Supernatants from microglial cell cultures were collected at 7 days postinfection and assayed for HIV-1 p24 Ag production by enzyme-linked immunosorbent assay (Beckman Coulter, Fullerton, CA, USA) as previously described (Peterson et al. 2004).

Cell viability

To assess the effect of cannabinoid receptor ligands on cell viability, two assays were used: microscopic evaluation of trypan blue dye exclusion and MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide mitochondrial dehydrogenase) (Sigma) assays. MTT was added to cell cultures at a final concentration of 1 mg/ml, and after 4 h of incubation, the assay was stopped by adding lysis buffer [20% SDS (w/v) in 50% N,N-dimethyl formamide, pH 4.7] followed by overnight incubation. The absorbance (OD) measured at 600 nm reflects mitochondrial integrity.

RT-PCR assay for cannabinoid receptor expression

Total RNA isolated with Qiagen RNeasy Mini kit (Qiagen, Valencia, CA) from microglia was treated with DNase (Ambion, Austin, TX) followed by reverse transcription (RT) for cDNA synthesis. Total RNA was reverse-transcribed to cDNA with SuperScript™ II reverse transcriptase, 5× cDNA synthesis buffer, 0.1 M dithiothreitol (DTT) (Invitrogen, Carlsbad, CA, USA), dNTP, oligo (dT)12–15 and random primer (Amersham Biosciences, Piscataway, NJ, USA) to a final volume of 20 μl. The same amounts of RNA were used in no RT reaction as controls to monitor DNA contamination. The cDNA was amplified by RT-polymerase chain reaction (RT-PCR) (FastStart DNA Master Plus SYBR Green detection with Roche LightCycler 2.0; Roche, Indianapolis, IN, USA, USA) with primers specific for CB1 and CB2 receptors. The amplified PCR products were electrophoresed for size verification. The primer sequences were: 5′-GCGTACTGCTTCTGTTCATCGTGTAT-3′ (sense) and 5′-GCAGGTCCTTACTCCTCAGAGCATAG-3′(antisense) forCB1; 5′-CCTCTCAGCACTAGTCTCCTACCTGC-3′ (sense) and 5′-TCCACTCCGTAGAGCATAGATGACAG-3′ (antisense) for CB2. The amplified PCR products were 363 and 436 bp, respectively. The amplification protocols were 95°C: 10 s, 67°C: 10 s and 72°C: 18 s for CB1 and CB2 receptors.

Assessment of CCR5

The effects of cannabinoid ligands on microglial cell expression of the chemokine receptor CCR5 was examined by flow cytometry. Microglia (3 × 105/0.5 ml medium) were incubated for 48 h in Teflon-coated vials in the absence (control) or presence of WIN55,212-2 or WIN55,212-3. Cells were then stained with fluorescein isothiocyanate-labeled anti-CCR5 antibody followed by flow cytometric analysis. Data are expressed as the percentage of microglia positive for CCR5 by using the FACSDiva software provided by the manufacturer (BD Biosciences, San Jose, CA, USA).

Statistical analysis

For analysis of the effect of cannabinoid receptor ligands on viral expression, p24 Ag levels in supernatants of cells treated with cannabinoid agonists and antagonists were expressed as % inhibition relative to p24 Ag levels in supernatants of untreated (control) cells. A mixed-effects, repeated-measures model was used that accounts for within-person correlations and intrinsic differences among individuals. The Tukey method was applied to adjust for multiple comparisons, and the Mixed Procedure in SAS version 8.2 was used to perform this analysis. Analysis of the effect of WIN55,212-2 on CCR5 expression involved the Student t-test as a measure of percent inhibition of CCR5 relative to control (untreated) microglia.

Results

Effect of CP55,940 on HIV-1 expression in microglia

Of the cannabinoid agonists that we had studied previously, only WIN55,212-2 potently inhibited HIV-1 expression (Peterson et al. 2004). As WIN55,212-2 binds with somewhat higher affinity to CB2 than CB1 receptors, we evaluated the nonselective CB1/CB2 receptor agonist CP55,940, a synthetic cannabinoid that binds with high affinity to both receptors. Suppression of viral expression by CP55,940 was concentration-dependent (Fig. 1) and similar to that previously observed with WIN55,212-2 (Peterson et al. 2004). Maximal inhibition (approximately 40% at 10−6 M), was comparable to WIN55,212-2 (Fig. 1). Microglial cell viability was assessed at 7 days of incubation in medium alone (control), CP55,940 (10−6 M), or WIN55,212-2 (10−6 M), and >98% of cells were viable in all treatment groups by trypan blue dye exclusion and MTT assays.

Fig. 1
figure 1

Concentration response of CP55,940 on HIV-1 expression. Microglia were incubated in culture medium alone (control) or were treated with indicated concentrations of CP55,940 or WIN55,212-2 for a total of 7 days. Vehicle (DMSO) had no effect on HIV-1 expression compared to medium alone (data not shown). Data are expressed as percent of control (untreated cells) p24 Ag values (6.3  ±  1.3 ng/ml, mean ± SEM; N  =  3 experiments using microglia from different brain tissue specimens). *P  <  0.05, **P  <  0.01 versus control.

Full size image

Cannabinoid receptors in microglia

Microglia isolated from five different brain tissue specimens were evaluated by RT-PCR for CB1 and CB2 receptor mRNA. As shown in Fig. 2a, microglia isolated from each of the five brain specimens contained mRNA for both CB1 and CB2 receptors.

Fig. 2
figure 2

CB1 and CB2 receptors and the effects ofcannabinoid receptor ligands on HIV-1 expression in microglia. (a) Total RNA from microglia obtained from five different brain specimens (lanes 1–5) were reverse-transcribed with RT to cDNA followed by RT-PCR amplification. Amplified PCR products were electrophoresed for CB1 (upper gel) and CB2 receptors (lower gel). RNA from each specimen was also reverse transcribed without RT as controls to monitor for DNA contamination. (b) Effects of CB1 and CB2 receptor agonists and antagonists on HIV-1 expression in microglia. Microglia were incubated in culture medium alone (control) or treated with indicated cannabinoid receptor ligands (10−6 M) for a total of 7 days. Data are expressed as percent of control p24 Ag values (6.3  ±  1.3 ng/ml, mean ± SEM; N  =  3 experiments using microglial cells from different brain tissue specimens). *P  <  0.05, **P  <  0.01 versus control. (c) Concentration response of AM-251 and SR141716A on HIV-1 expression. Microglia were incubated in culture medium alone (control) or were treated with indicated concentrations of AM-251 or SR141716A for a total of 7 days. Data are expressed as percent of control (untreated cells) p24 Ag values (0.23  ±  0.01 ng/ml, mean ± SEM; N  =  2 experiments using microglia from different brain tissue specimens). *P  <  0.05, **P  <  0.01 versus control. (d) Effect of CB2 receptor antagonist SR144528 on CP55,940 and WIN55,212-2-mediated inhibition of HIV-1 expression in microglia. Microglia were incubated in culture medium alone (control) or were treated with 10−6 M CP55,940, WIN55,212-2, or SR144528 alone, or were pretreated with SR144528 prior to CP55,940 or WIN55,212-2 and assessed after 7 days. Data are expressed as percent of control p24 Ag values (6.7  ±  3 ng/ml, mean ± SEM; N  =  four experiments using microglial cells from different brain tissue specimens). **P  <  0.01, versus control; ++P  <  0.01 versus CP55,940 or WIN 55,212-2 alone.

Full size image

Effects of CB1 and CB2 receptor agonists/antagonists on HIV-1 expression

Next, the effects of CB1 and CB2 receptor antagonists were compared to WIN55,212-2 and CP55,940. Microglia were treated with CB1/CB2 receptor agonists (CP55,940, WIN55,212-2), a CB2-selective agonist (JWH 0-15), and CB1-selective (SR141716A) and CB2-selective (SR144528) receptor antagonists at concentrations ranging between 10−9 and 10−6 M (Howlett et al. 2002). The inactive, neutral enantiomer of WIN55,212 (WIN55,212-3) was used as a control (Savinainen et al. 2005). The results are shown for the most potent concentration (10−6 M) for each of the agonists that had anti-HIV-1 activity (Fig. 2b). WIN55, 212-2 significantly inhibited HIV-1 expression as compared to its inactive enantiomer WIN55,212-3, which suggests that the inhibition of viral expression is receptor-mediated. The CB2-selective agonist JWH 0-15 had a similar degree of viral expression inhibition as CP55,940. While the CB2 receptor antagonist SR144528 had no effect on HIV-1 expression, surprisingly, the CB1 receptor antagonist SR141716A inhibited viral expression to a degree similar to that seen with CP55,940. This effect was further characterized by evaluating another CB1-selective antagonist AM-251, which had a similar inhibitory effect on HIV-1 expression to SR141716A (Fig. 2c). Additionally, the CB2-selective antagonist SR144528 did not abrogate the inhibitory effect on HIV-1 expression of SR141716A, nor did the CB1-selective antagonist AM-251 abrogate the inhibitory effect on HIV-1 expression of WIN55,212-2 (data not shown).

Because SR144528 by itself did not inhibit viral expression, microglia were pretreated with SR144528 prior to adding WIN55,212-2 or CP55,940. SR144528 significantly blocked the antiviral effect of both nonselective agonists (Fig. 2d), supporting the hypothesis that the mechanism of inhibition of viral expression in microglia involves CB2 receptors.

Effect of the enantiomers of WIN55,212 on expression of CCR5 in microglia

Based on previous studies on other psychotropic agents showing modulation of CCR5 in their effects on viral replication in macrophages (Guo et al. 2002; Szabo et al. 2003), we evaluated the effect of WIN55,212-2 on this coreceptor for HIV-1 entry into microglia. Microglia were treated with WIN55,212-2 or its inactive, neutral enantiomer WIN55,212-3 (Savinainen et al. 2005), and the percentage of microglia expressing CCR5 was compared to control cells as determined by flow cytometry (Fig. 3). WIN55,212-2 inhibited CCR5 (39.3  ±  8.3%, mean ± SEM), whereas treatment with the WIN55, 212-3 had little effect on CCR5 expression (5.1  ±  14.4% inhibition). The stereoselectivity of this response is a strong indication of a receptor-mediated process being involved. The interference with viral entry by down-regulation of CCR5 is one potential mechanism by which WIN55,212-2 inhibits HIV-1 expression in microglia.

Fig. 3
figure 3

Effect of the enantiomers of WIN55,212 on expression of CCR5 in microglial. Microglia were incubated for 48 h in the absence or presence of WIN55,212-2 (10−6 M) or the inactive enantiomer WIN55,212-3 (10−6 M). Cells were then stained with fluorescein isothiocyanate-labeled anti-CCR5 antibody followed by flow cytometry analysis. Data are expressed as percent inhibition of CCR5 relative to control (untreated) microglia (23  ±  8.5%, mean ± SEM; N  =  5 experiments using microglial cells from different brain tissue specimens). *P  <  0.05 versus control.

Full size image

Discussion

The purpose of this study was to characterize the cannabinoid receptor types present in human fetal brain-derived microglia and to determine which cannabinoid receptor type is involved in the anti-HIV-1 activity of WIN55,212-2. CB1 and CB2 receptors have been identified in macrophages/microglia isolated from rodent brain (Sinha et al. 1998; Carlisle et al. 2002; Facchinetti et al. 2003; Cabral and Marciano-Cabral 2005), whereas CB2 receptor mRNA was exclusively present in microglia obtained from adult human brain surgical specimens (Klegeris et al. 2003). The finding of CB1 and CB2 receptors in fetal brain-derived microglia could thus represent an age-related phenomenon or simply that CB1 and CB2 receptors are up-regulated during the preparation of the microglial cell cultures.

Although we hypothesized that WIN55,212-2 was acting through CB2 receptors, the presence of both CB1 and CB2 receptor mRNA expression in microglia raised the question of whether CB1 or CB2 was involved. In the course of studying this question, four observations were made: (1) the inhibitory effect of WIN55,212-2 on HIV-1 expression is not unique, in that CP55,940 also suppressed viral expression; (2) CB2 receptors are involved in WIN55,212-2 inhibitory effect, as witnessed by significant blockade with the CB2 receptor-selective antagonist SR144528 and the similar inhibitory effect of JWH 0-15, which has been shown to inhibit microglia activation (Ehrhart et al. 2005), suggesting that suppressed microglial activation and decreased HIV-1 replication may be related; (3) down-regulation of CCR5 receptors appeared to be one mechanism whereby WIN55,212-2 inhibits HIV expression, and (4) the involvement of CB1 in this process cannot be adequately assessed because of the surprising inhibitory effect of the CB1 receptor-selective antagonists SR141716A and AM-251. This paradoxical agonist activity of SR141716A has been previously identified in other macrophage/microglia models (Gross et al. 2000; Puffenbarger et al. 2000), including a report in which SR141716A’s inhibitory effect on multiplication of intracellular Brucella suis in human monocytes prompted the authors to suggest that this cannabinoid be considered for treatment of infections caused by intracellular gram-negative bacteria (Gross et al. 2000). The observation in the present study that agents such as WIN55,212-2, CP55,940, and SR141716A (which has also been shown to be effective in the treatment of obesity (Despres et al. 2005) and is about to be marketed under the trade name Rimonabant®), have the capacity to inhibit HIV-1 expression should foster further exploration of cannabinoids as potential antiviral agents. On the other hand, results from animal models raise concern regarding possible effects of the natural cannabinoid Δ-THC in promoting HIV-1 infection (Baldwin and Roth 2005).