Activation of Cannabinoid Type Two Receptors (CB₂) Diminish Inflammatory Responses in Macrophages and Brain Endothelium

Materials and Methods

Cells and Reagents

Primary human BMVEC were isolated from the temporal cortex of brain tissue obtained during surgical removal of eleptogenic foci in adult patients (Persidsky et al. 1997). BMVEC were cultured as described (Persidsky et al. 2006; Ramirez et al. 2008). Peripheral blood mononuclear cells were obtained from HIV-1, −2, and hepatitis B seronegative donors by leukopheresis and then purified by counter current centrifugation to generate pure (>98 %) populations of monocytes (Gendelman et al. 1988). Monocytes were obtained from the Human Immunology Core of the University of Pennsylvania (Philadelphia, PA). Monocytes were incubated for 7 days in media containing macrophage colony stimulating factor (M-CSF) and allowed to differentiate into monocyte-derived macrophages (MDM).

PCR Arrays (Inflammatory Response and Autoimmunity)

PCR-based microarrays for evaluating the expression of genes commonly involved in inflammatory response were performed using the Taqman Array Human Immune Response (Invitrogen, Carlsbad, CA). The arrays used to assess the anti-inflammatory effects of CB₂ agonists on BMVEC and MDM inflammatory responses were configured in a 96-well plate consisting of a focused panel of 92 targets and 4 endogenous control genes. qPCR was performed on a StepOnePlus Real Time PCR system (Invitrogen). BMVEC were exposed to TNFα (20 ng/ml, for 24 h, purchased from R&D Systems, Minneapolis, MN) in the presence of CB₂ agonist (O-1966 at 10 μM acquired from Organix, Woburn, MA or JWH-133 at 10 μM purchased from Tocris Bioscience, Ellisville, MO). Primary human monocytes were differentiated into MDM for 7 days in the presence of GM-CSF and then exposed to either LPS (50 ng/ml) alone or in combination with 10 μM of the selective CB₂ agonist, JWH133. After 24 h, the cells were lysed and the total RNA from each sample was isolated. Using reverse transcription for first-strand cDNA synthesis, the converted cDNA was then mixed with Taqman universal PCR master mix (ABI) and added to array plates containing probes to immune response-associated genes (Taqman Array Human Immune Response, ABI). The raw data were analyzed with DataAssist software (ABI) using the delta-delta Ct method (Relative Quantification). The data were expressed as relative gene expression (fold) compared to the untreated control.

Transendothelial Migration Assay

Transendothelial migration assays were performed as described (Rom et al. 2013, 2014). MCP-1/CCL2 (50 ng/mL, purchased from R&D Systems) was used as a relevant chemokine. BMVEC were pretreated with CB₂ agonist (O-1966 at 10 μM). Treatments were removed prior to monocyte introduction. Chemotaxis was allowed for 2 h. The data are shown as fold difference in migration (mean±SEM) from triplicate determinations, calculated from the number of migrated monocytes for each experimental condition divided by the number of migrated monocytes in the untreated, no chemoattractant control.

Transendothelial Electrical Resistance (TEER)

BMVEC were plated on collagen type I coated 96W20idf electrode arrays (ABI) and were treated with recombinant human soluble (s) CD40L (purchased from ProSpec, East Brunswick, NJ) in the presence of CB₂ agonist (JHW-133, O1966) or lysophosphatidic acid (LPA, 10 μmol, purchased from Santa Cruz Biotechnology, Santa Cruz, CA). TEER measurements were performed using the 1600R ECIS System (Applied Biophysics) as described (Ramirez et al. 2010a, 2012). The results are presented as the average percent change from baseline TEER (expressed as average±SEM) from at least three independent experiments consisting of four to six replicates each.

ELISA

Genes identified from the PCR array to be affected by CB₂ activation during inflammatory insult were evaluated for protein expression using ELISA. After MDM were treated as indicated in the figure, conditioned medium was collected to measure IFNγ-inducible protein of 10 kDa (IP-10) or CXCL10, and TNFα using conventional double sandwich ELISA from R&D Systems. Assays were performed according to the manufacturer's instructions.

Statistical Analysis

Data are expressed as the mean±SEM of experiments conducted multiple times. Multiple group comparisons were performed by one-way analysis of variance with Dunnet's posthoc tests (migration, ELISA). Statistical analyses were performed utilizing Prism v6.0c software (GraphPad Software, San Diego, CA). Differences were considered significant at p<0.05.

Results

HIV infection is known to be associated with chronic inflammation and activation of endothelium in different organs (Graham et al. 2013). Such endothelial cell responses can lead to end-organ injury, and interventions diminishing the state of activation (like CB₂ activation) should prevent tissue injury occurring during infection (Ramirez et al. 2012). To evaluate whether the anti-inflammatory effect of CB₂ activation applies to brain endothelial cells, we profiled the expression of genes commonly involved in the regulation of inflammatory response and autoimmunity. Using a commercial PCR-based focused array, 92 genes relevant to inflammatory responses were analyzed in untreated BMVEC (control), BMVEC activated with TNFα (20 ng/ml, 24 h) and TNFα-activated BMVEC in the presence of CB₂ agonists (O-1966 or JWH-133). The array analyses revealed that 33 gene targets were upregulated more than 2-fold in the TNFα-treated BMVEC when compared to untreated control cells. The addition of either O-1966 or JWH-133 resulted in the suppression of 31 and 32 genes, respectively, out of the 33 upregulated by TNFα. Fifteen genes belong to cytokine/chemokine group, 8 are adhesion molecule and receptors; the rest are from a group of cellular response and transcription factors. Table 1 shows the pro-inflammatory mediators that were most affected by CB₂ activation. When compared to BMVEC that were treated with TNFα only, cytokine-stimulated BMVEC exposed to O-1966 showed a 97 % attenuation in iNOS, a 88 % reduction in CXCL10 and CXCL11, and a 40–50 % decrease in ICAM-1, IL-8, CCL2 and CD40 expression among others. Similarly, JWH-133 reduced pro-inflammatory gene expression in TNFα-stimulated BMVEC: 98 % for iNOS, 97 % for CXCL10 and CXCL11, 88 % for ICAM-1 and IL-15, and 75 % for CCL2, MCSF-1 and IL-18 and CD40.

Changes in CD40 expression by CB₂ agonists are of interest because the functional relevance of CD40 expression in BMVEC was previously established in the context of HIV-1 infection (Ramirez et al. 2010b). We showed that exposure of primary human BMVEC to soluble CD40L (sCD40L) upregulated the expression of adhesion molecules ICAM-1 and VCAM-1, which caused a four-fold increase in monocyte adhesion to BMVEC and stimulated migration across an in vitro BBB model. It has been suggested that increased levels of sCD40L are associated with cognitive impairment in HIV infection (Sui et al. 2007). Now, we have assessed the effect of sCD40L on barrier tightness and found that average TEER measurements showed a rapid 5–20 % drop in resistance in a dose-dependent manner (Fig. 1a). The positive control, LPA (a known activator of RhoA) (Yamamoto et al. 2008), caused significant decrease of TEER indicating an appropriate path-ophysiologic response of BBB models. To test the protective effects of CB₂ agonists, O-1966 and JWH-133 (at10 μM), prevented a TEER drop caused by sCD40L (Fig. 1b) indicating that CB₂ activation can protect the BBB under inflammatory conditions.

Fig. 1

Barrier function is compromised in sCD40L-treated BMVEC. a TEER, an indicator of barrier integrity, was measured (by ECIS) in BMVEC monolayers untreated or treated with 1, 10 and 100 ng/ml sCD40L or LPA (10 μmol). The resistance was measured at 4000 Hz in 10 min intervals for the duration of the time shown. Treatments were initiated (arrow) after stable resistance was reached. b TEER was measured after sCD40L application (50 ng/ml) with or without CB₂ agonists (O-1966 and JWH-133 at 10 μM). BMVEC monolayers were treated with CB₂ agonists for 2 h prior to the addition of sCD40L. Both agonists prevented TEER decrease. Each data point is represented as the percent of the mean value±SEM (n=3). Open (white) arrowhead indicates time of CB₂ agonist application; closed (black) arrowhead indicates time of sCD40L addition

Next using migration assays in an in vitro BBB model, we tested whether CB₂ activation in endothelial cells could prevent monocyte passage across BMVEC monolayers. We used CCL2 as a relevant cytokine that is upregulated in the CNS under neuroinflammatory conditions. Application of CCL2 to the lower chamber of BBB constructs increased monocyte migration 6-fold, as compared to models without chemokine addition. Pre-treatment of BMVEC with O-1966 attenuated monocyte migration across endothelial monolayers by 50 % (Fig. 2).

Fig. 2

CB₂ stimulation blocked monocyte migration across BMVEC monolayers. BMVEC were pretreated with O-1966 (10 μM) that was removed prior to monocyte introduction. The migration assay was performed using 2.5×10 ^{BMVEC/insert with calcein-AM labeled monocytes added to BMVEC seeded on collagen-coated FluoroBlok BD inserts. CCL2 (MCP-1, 30 ng/ml) was used as a relevant chemokine. Chemotaxis was allowed for 2 h. Data are shown as mean±SEM. *} p<0.01

Activation of macrophages/microglia in the CNS and secretion of pro-inflammatory factors is believed to cause neuronal dysfunction and HAND (Kaul and Lipton 2006). Selective CB₂ activation was shown to diminish HIV replication in macrophages (Ramirez et al. 2013) and prevent monocyte migration across the BBB (Rom et al. 2013). To evaluate whether the anti-inflammatory effect of CB₂ activation applies to monocyte-derived macrophages (MDM), we profiled the expression of genes commonly involved in the regulation of inflammatory responses (Table 2). Using a commercial PCR-based focused array, 92 genes (including cytokines and cytokine receptors; chemokines and chemokine receptors) relevant to inflammatory responses were analyzed in untreated MDM (control), MDM activated with LPS (50 ng/ml, 24 h) and LPS-activated MDM in the presence of the CB₂ agonist, JWH133. The array analyses revealed that 50 genes were up-regulated more than 2-fold in LPS-treated MDM when compared to untreated control cells. The addition of JWH133 resulted in the suppression of 35 genes out of the 50 genes that were up-regulated by LPS (13 from cytokine/chemokine group, 10 belonging to adhesion molecule/receptors, and 8 to cell response/transcription factors). These included CXCL10, CXCL11, CCL19, CCL5, CCR7 (99–95 % attenuation), TNFα, ICAM-1, and CCL3 (64 % reduction of expression) among others.

Next, we measured the inflammatory molecules corresponding to genes affected by CB₂ stimulation in MDM. ELISA assays for TNFα and CXCL10/IP-10 were performed to validate the gene expression results acquired from the gene array immune response panel. LPS stimulation resulted in a 90-fold increase of TNFα production and CB₂ activation by JWH133 resulted in a 98 % suppression of TNFα secretion (Fig. 3a). Similarly, LPS activation of MDM led to a 242-fold increase in IP-10 production, and JWH-133 treatment reduced IP-10 secretion by 97.5 % (Fig. 3b). These results indicate that CB₂ agonist effectively suppresses potent inflammatory mediators secreted by activated MDM.

Fig. 3

The selective CB₂ agonist, JWH133, attenuated secretion of inflammatory mediators from LPS-activated MDM. Conditioned medium from MDM was collected after 24 h of treatment as indicated in panels a and b. TNFα and CXCL10/IP-10 levels were evaluated using ELISA (R&amp;D Systems). a LPS exposure increased TNFα secretion by 90-fold when compared to the untreated control, whereas LPS with CB₂ agonist down-regulated the secretion by 98 % when compared to LPS-treated MDM. b CXCL10/IP-10, when compared to untreated control, showed a 242-fold induction by LPS that was decreased by 97.5 % by JWH133. The results are expressed as mean values±SD with asterisks indicating statistical significance (p<0.05)

Discussion

The rapid move towards legalization of marijuana (now accepted for medicinal or recreational use in 20 states in the US) poses a significant challenge for biomedical research, namely identifying what compounds in the cannabis plant may possess beneficial effects in diseases (2014). Identification of CB₁ and CB₂ receptors with clearly distinct cellular distribution patterns and functions offers this opportunity. High levels of CB₂ expression on immune cells and endothelial cells and its upregulation by inflammatory mediators suggest its participation in immune responses and resolution of inflammation (Buch 2013).

Such effects are very pertinent for HIV-1 infection in and outside of the CNS. HAND (Letendre et al. 2011; Spudich 2014) continues to be highly prevalent and the progression of HAND is associated with biomarkers of chronic immune activation driven by low levels of virus infection in monocytes and macrophages (Cross et al. 2013; Marcotte et al. 2013). Additional interventions further dampening HIV replication and diminishing chronic neuroinflammation are critically needed. Data presented in this report indicate that a number of pro-inflammatory molecules (playing an important role in HAND progression) have been decreased by CB₂ agonists in brain endothelium and human macrophages. Diminished gene expression has been accompained by attenuated protein secretion (by MDM) and funtional assays mimicking BBB injury (monocyte migration, TEER). These molecules (such as CCL2, TNFα, CXCL10, CCL5) are among those shown to be increased in HAND and are considered as neurotoxins driving neuronal dysfunction. Similarly, a number of such molecules (ICAM-1, CXCL10, CXCL11, CD40, VEGF) were downregulated in BMVEC and such changes were accompanied by decreased monocyte migration across BBB models and preservation of BBB integrity after application of sCD40L. These observations further confirm previously published data in BMVEC (attenuation of adhesion molecule expression, prevention of leukocyte/BMVEC interactions in vitro and in vivo, prevention of BBB leakiness, etc.) (Ramirez et al. 2012) and monocytes/macrophages (decresed migration/adhesion, attenuation of integrin expression, decreased lamellipodia formation, HIV replication) (Ramirez et al. 2013; Rom et al. 2013).

Overall, our data parallel previous work indicating the therapeutic potential of CB₂ activation. In immune cells, CB₂ stimulation decreased production of pro-inflammatory factors (Puffenbarger et al. 2000; Facchinetti et al. 2003) that are implicated in neuronal injury during HIV-1 CNS infection (Kraft-Terry et al. 2010). Neuroprotective effects of CB₂ agonists are associated with suppression of microglia activation (Klegeris et al. 2003; Eljaschewitsch et al. 2006) via inhibition of the release of neurotoxic factors. In vitro and in vivo studies have shown that cannabinoids can act on glia, enhancing the release of the anti-inflammatory cytokines, IL-4 and IL-10 (Molina-Holgado et al. 1998). CB₂ signaling interfered with the enhanced expression of iNOS and CCR2 induced by IFNγ in mouse microglial cells (Racz et al. 2008). Selective CB₂ agonist diminished neuroinflammation in a rodent model of HIVE (Gorantla et al. 2010). Chronic neuroinflammation documented in HIV infection (Suh et al. 2014) requires adjunctive therapies in addition to ART. Highly selective CB₂ agonists with better pharmacologic properties could be one such therapy.

Acknowledgments