Development of an in Vitro Peptide-Stimulated T Cell Assay to Evaluate Immune Effects by Cannabinoid Compounds

Abstract

Previous studies demonstrated that cannabinoids exhibit immunosuppressive effects in experimental autoimmune encephalomyelitis (EAE), the animal model of multiple sclerosis (MS). To ask questions about treatment timing and investigate mechanisms for immune suppression by the plant-derived cannabinoids, cannabidiol (CBD) and Δ⁹-tetrahydrocannabinol (THC), an in vitro peptide stimulation of naive splenocytes (SPLC) was developed to mimic T cell activation in EAE. The peptide was derived from the myelin oligodendrocyte glycoprotein (MOG) protein, which is one component of the myelin sheath. MOG peptide is typically used with an immune adjuvant to trigger MOG-reactive T cells that attack MOG-containing tissues, causing demyelination and clinical disease in EAE. To develop the in vitro model, naïve SPLC were stimulated with MOG peptide on day 0 and restimulated on day 4. Cytokine analyses revealed that CBD and THC suppressed MOG peptide-stimulated cytokine production. Flow cytometric analysis showed that intracellular cytokines could be detected in CD4+ and CD8+ T cells. To determine if intracellular calcium was altered in the cultures, cells were stimulated for 4 days to assess the state of the cells at the time of MOG peptide restimulation. Both cannabinoid-treated cultures had a smaller population of the calcium-positive population as compared to vehicle-treated cells. These results demonstrate the establishment of an in vitro model that can be used to mimic MOG-reactive T cell stimulation in vivo.

Introduction

Cannabinoids are well-established to modulate immune responses in vivo and in vitro. Indeed, we and others have shown that the plant-derived cannabinoids, cannabidiol (CBD) and Δ⁹-tetrahydrocannabinol (THC) suppress the autoimmune disease experimental autoimmune encephalomyelitis (EAE) in mice (reviewed in [1]). One mechanism to induce EAE in mice uses active immunization with a peptide derived from myelin oligodendrocyte glycoprotein (MOG) protein, which is one component of the myelin sheath. The immune response raised against MOG results in activation of MOG-reactive T cells that have the potential to destroy MOG-containing tissues, contributing to demyelination [2]. Specifically, our study demonstrated that oral CBD given for 5 days after EAE disease initiation protected mice from disease and reduced the percentage of splenic T cells producing IFN-γ [3]. As EAE is a model to study multiple sclerosis (MS), these results suggest that cannabinoids might provide therapeutic benefit for MS patients. The above referenced review also included human studies of MS in which cannabinoids were used and they concluded that cannabinoids as an add-on treatment provided modest improvement of spasticity, pain, bladder control, and quality of life [1].

While the above results show that the EAE model in mice might help predict therapeutic usefulness of cannabinoids for human MS, there is continued interest in developing in vitro approaches, which would reduce animal use and allow in-depth investigation of mechanisms by which cannabinoids alter immune responses. The 2D2 MOG-specific T cell receptor (TCR) mice serve as one source of T cells that can be used in vitro to study MOG-specific responses [4]. Alternatively, primary MOG-specific T cell lines can be established from the lymph nodes of EAE mice [5]. Primary MOG-specific T cell lines have been used extensively to examine effects and mechanisms by which CBD exhibits immunosuppression in vitro [6–9]. Some of the major effects observed after treating MOG-specific T cells with 5 μM CBD include suppression of proliferation [9], suppression of IL-17 signaling [6, 8], and induction of anergy [7].

One challenge with using T cells from 2D2 MOG TCR mice or primary MOG T cell lines is that they are not naïve cells responding to MOG for the first time and therefore, more closely mimic secondary responses. We have previous data, at least for THC, that demonstrates that primary T cell-dependent responses are more sensitive to suppression than secondary responses. First, THC robustly suppressed the in vivo T cell-dependent antibody response to sheep red blood cells when THC was administered at the time of sensitization as compared to after sensitization [10]. Second, THC suppressed the in vitro cytotoxic T cell response to P815 target cells only when THC was added during the elicitation, but not the effector, phase [11].

Thus, th purpose of these studies was to establish an in vitro MOG-stimulated T cell model using initially naïve splenocytes and assess the effects of CBD and THC. We hypothesized that MOG peptide would stimulate naïve mouse splenocytes to produce cytokines from T cells and that cannabinoids would inhibit them. We also suspected that there would be more robust cytokine production if the cells were restimulated with the peptide; therefore, we could study the effects of cannabinoids on both the primary response (i.e., initial MOG stimulation) and secondary response (i.e., MOG restimulation) in vitro. Another advantage of this model is that SPLC were used, a mixed population of naive mature immune cells that allows the T cell response to develop in the context of other cells, for instance, antigen presenting cells (APCs) that likely present the MOG peptide to T cells to initiate the immune response. Finally, these studies allowed us to investigate the effects of cannabinoids on intracellular calcium in MOG-stimulated SPLC, which is one assessment that would be difficult to conduct in EAE in vivo.

Materials and Methods

Cannabinoids.

CBD and THC were provided by the National Institute on Drug Abuse (NIDA) through the NIDA Drug Supply Program.

Cell culture.

Mouse splenocytes (SPLC) were seeded at 5×10⁶ cells/ml in 1 ml of complete media in a 48-well plate. Complete media was 1X RPMI containing 2% bovine calf serum, 1% penicillin-streptomycin, and 50 μM 2-mercaptoethanol. Cells were treated on day 0 with cannabinoids and peptides (either MOG_35–55 denoted as “MOG” or OVA_257–264 denoted as “OVA”; 100 μg/ml) in that order allowing 1 hr in between treatments (Figure 1). Appropriate vehicle (VH) controls for cannabinoids were used and are noted in the figure legends. The MOG peptide stimulation was incubated for 4 days then 0.6 ml of media was removed and replaced. For cultures in which peptide stimulation was continued, cannabinoids, and MOG peptide (100 μg/ml) were again added as dictated by the experimental design, in that order, allowing 1 hr in between treatments. Peptide restimulated cultures were incubated for an additional 1, 2, or 3 days. For cultures in which the initial MOG peptide stimulation was followed by PMA/Io, 0.6 ml of media was removed and replaced with media containing 40 nM PMA/0.5 μM Io for various times, depending on the experimental endpoint.

Figure 1.

In vitro peptide stimulation approach. SPLC were seeded at 5×10⁶ cell/well in 1 ml in 48-well plates in each of two plates. SPLC were treated with vehicle (VH; 0.1% ethanol) or CBD or THC (10 μM each) on day 0. After 1 hr, cells received peptide (either MOG or OVA; 100 μg/ml). Plates 1 and 2 were used to collect supernatants and cells for cells stimulated for days 4 and 7, respectively. For those cultures to be harvested at day 7 (i.e., in plate 2), ~600 μl of media was removed and replenished, and cells were treated with VH, CBD or THC followed by peptide after 1 hr on day 4. For some experiments, day 4 cells were stimulated overnight with PMA/Io and/or used to assess intracellular calcium levels. For experiments with antagonists, antagonist or VH was added for 1 hr prior to cannabinoid treatment on day 1.

ELISAs.

Supernatants were obtained from cultures at various times over the culture period and stored at −80°C until ELISA analysis. ELISA assays were conducted with purified and biotinylated antibody matched pairs and recombinant proteins for standards (IL-2, IFN-γ; all reagents obtained from BioLegend) or kits were utilized (GM-CSF kit from BioLegend; TNF-α kit from ThermoFisher). One day prior to the ELISA, Immunolon IV strip wells were coated with a primary antibody for the cytokine of interest in coating buffer (1 M sodium bicarbonate for matched pairs or respective coating buffer provided in kit). Wells were washed with ELISA wash buffer (0.5% tween 20 in phosphate buffered saline; PBST) three times followed by three rinses in deionized water. Wells were then blocked with 3% bovine serum albumin in phosphate buffered saline (3% BSA-PBS) for matched pairs, or with respective blocking/diluent buffer provided in kit for at least 1 hr at room temperature (RT). After another wash cycle, standard curves were prepared using recombinant proteins, then standards and samples were loaded into the wells, using various dilutions for samples to ensure absorbance values fell below the saturated part of the standard curve. Standards and samples were incubated for at least 1 hr at RT. After washing, biotinylated secondary antibodies were prepared in the appropriate diluent buffer and allowed to incubate in the wells for 1 hr at RT. Cytokines were detected following incubation with horseradish peroxidase-streptavidin for at least 30 min at RT, another wash cycle, incubation with the tetramethylbenzidine (TMB) substrate for no more than 15 min at RT, and termination of the reactions with 2N H₂SO₄. The yellow color absorbance was detected using an endpoint assay at 450 nm. Cytokine concentrations were normalized to cell numbers enumerated at the time of supernatant collection using the cell counting capability on the ACEA Novocyte (ACEA/Agilent) flow cytometer using forward and side scatter characteristics typical of leukocytes.

Extracellular staining for flow cytometry.

Cells were placed in U-bottom 96-well plates and centrifuged at 500 x g for 5 min at room temperature (RT). After a wash in 1X PBS, cells were incubated with near IR fixable viability dye (NIR-FVD) for 20 min in the dark at RT. After another wash in 1X PBS, cells were incubated with mouse Fcblock (anti-CD16/32) in flow cytometry buffer (FCM; 1X Hank’s Balanced Salt Solution with 1% BSA) for 10 min in the dark at RT. Antibodies directed against extracellular proteins were prepared in 50 μl FCM and incubated with the cells for at least 30 min in the dark at RT. Various extracellular antibodies were used, including CD4-FITC, APC-CD8, PECy7-F4/80, BV785-CD19, and PE-tetramer. All antibodies for flow cytometry were obtained from BioLegend (San Diego, CA) except MOG T cell-specific tetramer and its negative control were obtained through the NIH Tetramer Core Facility. Cells were then fixed with Cytofix (BD Biosciences, San Jose, CA) and either analyzed on the ACEA Novocyte or subjected to intracellular staining.

Intracellular staining for flow cytometry.

For cells in which intracellular cytokines were quantified, cultures were incubated for the last 4 hr with Brefeldin-A and phorbol ester plus calcium ionophore (PMA/Io). After 4 hours, cells were stained for extracellular proteins and fixed as indicated above for extracellular staining. On the day of intracellular staining (usually within 72 hr of the extracellular staining), cells were washed with permeabilization buffer. Intracellular antibodies were prepared in 50 μl permeabilization buffer and incubated with the cells for at least 30 min in the dark at RT. Various intracellular antibodies were used, including APC-TNF-α and PE-IFN-γ (BioLegend). Cells were then immediately analyzed on the ACEA Novocyte.

Flow cytometry data analysis.

Antibody beads and viability beads were used to create single fluorescence controls to set compensation. Cellular controls included fluorescence minus one (FMO) controls in which all antibodies and NIR FVD are included with the exception of one. FMO controls and unstimulated samples are used to guide negative gate setting. Cells were analyzed on the ACEA Novocyte then gate setting, percent population, and mean fluorescence intensity (MFI) were calculated using NovoExpress software.

Calcium assay.

Cells that were treated with cannabinoids and MOG for 4 days were assessed for intracellular calcium levels using the Fluo-4 calcium assay kit (Abcam). Briefly, cells from 3 wells of cells treated with VH, CBD or THC (all plus MOG stimulation) were harvested into 15-ml conical tubes and centrifuged at 500×g for 5 min. After a PBS wash, unloaded cells (i.e., not stained with calcium dye) were incubated in 1 ml 1X Hank’s Buffered Saline Solution (HBSS) and loaded cells were incubated in 1 ml staining cocktail according to the manufacturer’s instructions (0.9 ml 1X HBSS, 0.1 ml F127 staining reagent, and 0.5 μl Fluo-4) for 30–60 min at 37°C. Cells were then centrifuged again as above then incubated in 10 ml Ca2+-KREB buffer (129 mM NaCl, 5 mM KCl, 1.2 mM KH₂PO₄, 1.2 mM MgSO₄, 1mM CaCl₂, 5 mM NaHCO₃, 10 mM HEPES, 2.8 mM glucose, and 0.2% bovine serum albumin) for 15 min at 37°C. After another centrifugation, cells were suspended in 1 ml Ca2+-KREB buffer and analyzed on the ACEA Novocyte. A 100-μl aliquot of cells was mixed 1:1 with either Ca2+-KREB buffer or P/I (prepared using 4 μl of 40 μM PMA/0.5 mM ionomycin stock into 2 ml Ca2+-KREB buffer). Events (100,000) were captured using the FITC channel to detect changes in Fluo-4 dye for intracellular calcium changes.

Data analysis.

ELISA data are presented as mean ± SD from a minimum of triplicate cultures. Separate cultures were conducted at least twice for all data. Percent data were transformed using arcsin(Y/100) before using one-way or two-way ANOVA as dictated by the experimental design. Post hoc tests as described in the figure legends were conducted to detect differences between groups. Calcium data are presented as mean ± SD from a minimum of three separate experiments.

Results

CBD and THC suppressed MOG peptide-stimulated cytokine production.

In order to induce cytokine production using MOG peptide from naïve mice, SPLC from female mice were seeded at a relatively high density and stimulated with the peptide on days 0 and 4. Cells were also treated with vehicle (0.1% ethanol) or CBD or THC (10 μM each) on day 0 only, on day 4 only, or on both days. As seen in figure 2, cells could be induced to produce IFN-γ, TNF-α, and GM-CSF by day 7 after two treatments with MOG peptide, while IL-2 as only induced following the initial stimulation. For IFN-γ, TNF-α, and GM-CSF, the general pattern was that cells receiving drug treatments on both days produced the most robust inhibition, while cells receiving CBD or THC only on day 0 also inhibited cytokine production. For those cells receiving CBD or THC only after the initial 4-day stimulation with MOG peptide, the inhibition was not as robust. Together these results suggest that the degree to which CBD and THC suppress MOG peptide-stimulated cytokine production is time-dependent, with those cells treated before the initial activation producing more robust inhibition than those cells treated 4 days after peptide.. The experiments were repeated using SPLC from male mice with specific focus on IFN-γ and TNF-α since these are sensitive targets of cannabinoids [12]. The effects of cannabinoids on TNF-α were more variable although the patterns of IFN-γ and TNF-αlooked similar to females, especially at day 7 (Figure 3). In order to verify that the in vitro peptide stimulation was not limited to MOG, cells from female SPLC were stimulated with an ovalbumin (OVA) peptide again focusing on IFN-γ and TNF-α (Figure 4). Again, with OVA peptide stimulation there was more robust inhibition of cytokine production (at least for TNF-α) when the cannabinoid was added at day 0 as compared to only on day 4. In order to understand the magnitude of stimulation by MOG or OVA, untreated cultures were incubated overnight and assessed for cytokine production; no untreated cultures exceeded 20% of the peptide stimulated VH (Supplemental Figure 1).

Figure 2.

Cannabinoids inhibited MOG peptide-stimulated T cell responses in SPLC from female mice. SPLC were seeded at 5×10⁶ cell/well in 1 ml in 48-well plates and treated with cannabinoids and MOG peptide as indicated in Figure 1. Supernatants were used in ELISA analyses and cells were used to conduct a total cell count using the ACEA Novocyte. Cytokine concentrations were determined using a standard curve in the ELISA then values were normalized to total cell counts within each experiment. Data were then normalized within experiments to the average VH/VH on day 7 (except IL-2, which was normalized to day 4 since levels dropped over time) so that data from separate experiments could be combined. Each experiment was run with ≥3 separate wells per treatment group. * p< 0.05 as compared to the respective VH/VH within each day using two-way ANOVA followed by Tukey’s multiple comparison test. Data are mean ± SD. A., TNF-α, 2 experiments; B., IFN-γ, 3 experiments; C., GM-CSF, 2 experiments; D., IL-2, 2 experiments.

Figure 3.

Cannabinoids inhibited MOG peptide-stimulated T cell responses in SPLC from male mice. SPLC were seeded at 5×10⁶ cell/well in 1 ml in 48-well plates and treated with cannabinoids and MOG peptide as indicated in Figure 1. Supernatants were used in ELISA analyses and cells were used to conduct a total cell count using the ACEA Novocyte. Cytokine concentrations were determined using a standard curve in the ELISA then values were normalized to total cell counts within each experiment. Data were then normalized within experiments to the average VH/VH on day 7 so that data from separate experiments could be combined. Each experiment was run with ≥3 separate wells per treatment group. * p< 0.05 as compared to the respective VH/VH within each day using two-way ANOVA followed by Tukey’s multiple comparison test. Data are mean ± SD. A., TNF-α, 3 experiments; B., IFN-γ, 3 experiments.

Figure 4.

Cannabinoids inhibited OVA peptide-stimulated T cell responses in SPLC from female mice. SPLC were seeded at 5×10⁶ cell/well in 1 ml in 48-well plates and treated with cannabinoids and OVA peptide as indicated in Figure 1. Supernatants were used in ELISA analyses and cells were used to conduct a total cell count using the ACEA Novocyte. Cytokine concentrations were determined using a standard curve in the ELISA then values were normalized to total cell counts within each experiment. Data were then normalized within experiments to the average VH/VH on day 7 so that data from separate experiments could be combined. Each experiment was run with ≥3 separate wells per treatment group. * p< 0.05 as compared to the respective VH/VH within each day using two-way ANOVA followed by Tukey’s multiple comparison test. Data are mean ± SD. A., TNF-α, 3 experiments; B., IFN-γ, 3 experiments.

MOG peptide stimulation increased the T cell population in SPLC, which was suppressed by CBD and THC.

While the cytokine stimulation by the peptides suggested that T cells were producing at least some of the cytokines, we next used flow cytometry to quantify T cells in the female SPLC cultures. First, we assessed whether there was an enrichment of T cells over the 7-day culture period (Figure 5). As compared to typical naïve SPLC cell percentages (listed along the bottom), there was an increase in CD4+ T cells to ~23%, CD8+ T cells to ~20% and F4/80+ cells (macrophages) to ~5%. CD19+ B cells dropped to ~29%. We also used a tetramer stain that is specific for the MOG peptide, and that staining increased to over 1%. The tetramer was specific as it stained CD4+ T cells derived from an EAE mouse spleen (Supplemental Figure 2). We next examined the effects of CBD and THC on CD4+ and CD8+ T cell percentages and found that there was no effect on T cells by CBD or THC treated only on day 4, but those cells that received CBD or THC only on day 0 (or on both day 0 and 4) exhibited lower percentages of CD8+ T cells and higher percentages of CD4+ T cells (Figure 6). Finally, in order to show that T cells could produce cytokines in response to MOG peptide, we performed intracellular staining for TNF-α in female SPLC. This was purposefully conducted on day 6, the day before we measured cytokine in the culture supernatants on day 7. CBD and THC both suppressed intracellular cytokine production in CD4+ and CD8+ T cells and again, those cells treated only on day 4 were relatively resistant to suppression, while those treated at the time of initial MOG peptide stimulation exhibited more suppressive effects (Figure 7).

Figure 5.

T cells and macrophages were enriched in the MOG peptide-stimulated cultures from female mice. SPLC were seeded at 5×10⁶ cell/well in 1 ml in 48-well plates and treated with MOG peptide. On day 4, cells were restimulated with MOG peptide. On day 7, cells were harvested and stained for various immune cell markers, CD4 and CD8 (T cells), CD19 (B cells) and F4/80 (macrophages). Cells were also stained with a tetramer that recognized MOG-specific T cells. For comparison, typical percentages of naïve SPLC are provided along the bottom. Top, cells stained with the negative tetramer control; Bottom, cells stained with tetramer. Results are representative of two separate experiments.

Figure 6.

Cannabinoids inhibited MOG peptide-stimulated CD8+ T cell percentages in SPLC from female mice. SPLC were seeded at 5×10⁶ cell/well in 1 ml in 48-well plates and treated with cannabinoids and MOG peptide as indicated in Figure 1. At day 7, cells were harvested and stained for CD4 and CD8. Data were then normalized within experiments to the average VH/VH on day 7 so that data from separate experiments could be combined. Each experiment was run with ≥3 separate wells per treatment group. * p< 0.05 as compared to the respective VH/VH using Tukey’s test. Data are mean ± SD. A., CD4, 3 experiments; B., CD8, 3 experiments.

Figure 7.

Cannabinoids inhibited MOG peptide-stimulated T cell responses in SPLC from female mice. SPLC were seeded at 5×10⁶ cell/well in 1 ml in 48-well plates and treated with cannabinoids and MOG peptide until day 6. Cells were treated for the last 4 hr of the culture with Brefeldin-A + P/I. Cells were then stained for CD4 or CD8 then intracellularly for TNF-α. Data were then normalized within experiments to the average VH/VH on day 6 so that data from separate experiments could be combined. Each experiment was run with ≥3 separate wells per treatment group. * p< 0.05 as compared to the respective VH/VH using Tukey’s test. Data are mean ± SD. A., CD4, 3 experiments; B., CD8, 3 experiments.

CBD and THC suppressed intracellular calcium.

Many cytokines, including TNF-α and IFN-γ, are dependent on increased intracellular calcium [13, 14]. To examine whether the cannabinoid-mediated suppression of cytokine production was dependent on alterations in intracellular calcium, cannabinoid plus MOG-treated cells for 4 days were restimulated with phorbol ester plus calcium ionophore (P/I) instead of MOG peptide. This allowed us to evaluate the state of the cells at the time before they would receive the MOG peptide restimulation on day 4. P/I was used for the calcium studies because we anticipated we would be able to measure an induced calcium signal in real time on the instrument with P/I but not with MOG peptide. Because this restimulation approach was different (i.e., using P/I instead of MOG peptide), we first established that cells initially stimulated with MOG for 4 days followed by overnight P/I stimulation would also result in cytokine suppression by cannabinoids. As seen in Figure 8A and B, those cells that were treated with CBD and THC before initial MOG peptide stimulation on day 0 suppressed cytokine production on day 5 after receiving P/I restimulation on day 4. Thus, we proceeded with the assessment of cannabinoids on intracellular calcium in the MOG-stimulated cells on day 4. An example of the intracellular calcium assessment is shown in Figure 8C. Cells that had been treated with VH, CBD, or THC plus MOG on day 0 were either not loaded or loaded with calcium dye on day 4. Unloaded cells were used as negative controls and guided the placement of the positive calcium gate. Loaded cells produced a positive peak in the FITC channel and an additional peak was noted when the cells were treated with P/I in real time (designated as “hi Ca” for high calcium). We noted that both CBD and THC treatment of MOG-stimulated cells on day 0 reduced the high calcium peak induced by P/I in real time (Figure 8C). Furthermore, both CBD and THC suppressed the total and high calcium populations as shown by percent and reduced the mean fluorescence intensity (MFI) of the calcium signal (Figure 9). Interestingly, THC did not alter the MFI in the high calcium signal which might be due to the fact that it reduced, but did not eliminate, the high calcium peak (refer to Figure 8C)..These results suggest that cells initially stimulated with MOG peptide on day 0 and restimulated with P/I on day 4 were able to produce a robust calcium signal that was reduced if the cells also received CBD or THC on day 0.

Figure 8.

Cannabinoids inhibited cytokine production and intracellular calcium in SPLC from female mice. SPLC were seeded at 5×10⁶ cell/well in 1 ml in 48-well plates and treated with cannabinoids and MOG peptide until day 4. Cells were either stimulated overnight with P/I for cytokine production assessment on day 5 (A., B.) or used for calcium determinations on day 4 (C). Each experiment was run with ≥3 separate wells per treatment group for ELISAs. For calcium, cells were loaded with Fluo-4 dye then analyzed on the FITC channel by flow cytometry. Unloaded cells were used to set the positive calcium gate. Loaded cells were either untreated or treated with P/I (40 nM/0.5 μM) immediately before the analysis to assess effects on intracellular calcium using 100,000 events per sample (data in right column under the + sign are those that received P/I). Data were normalized within experiments to the average VH control within each experiment so that data from separate experiments could be combined. * p< 0.05 as compared to the respective VH using Dunnett’s test. Data are mean ± SD. A., TNF-α, 3 experiments; B., IFN-γ, 3 experiments; C., representative calcium data from 5 separate experiments.

Figure 9.

Cannabinoids inhibited intracellular calcium in SPLC from female mice. SPLC were seeded at 5×10⁶ cell/well in 1 ml in 48-well plates and treated with cannabinoids and MOG peptide until day 4. Cells were then used for calcium determinations on day 4 as described in Figure 8 using only the P/I-restimulated cells. Data were normalized within experiments to the average VH control within each experiment so that data from separate experiments could be combined (n = 5 for CBD; n = 3 for THC). * p< 0.05 as compared to the respective VH using Dunnett’s test. Data are mean ± SD. A., Total calcium population % gated; B., total calcium population MFI; C., high calcium population % gated; D., high calcium population MFI.

Discussion

In this work, we established that naïve SPLC would produce cytokines in response to MOG or OVA peptides in vitro, and that both CBD and THC could inhibit peptide-stimulated cytokine production. CBD and THC also produced a long-term inhibition (i.e., at least 4 days) of intracellular calcium, which likely contributed to the inability of the cells to produce cytokines [13, 14]. Importantly, we also showed that at least some of the cytokines produced in the culture could be produced in T cells. Thus, these studies established an antigen-specific model of T cell stimulation.

It was interesting that both MOG and OVA stimulated cytokines in the naïve SPLC; to the mouse, MOG is a self-peptide and OVA is foreign. These results show that there exists a precursor population of T cells in the mouse reactive to MOG or OVA. In fact, we purposefully seeded the cells at a relatively high density since we expected that the precursor frequency of MOG- or OVA-reactive T cells in a mouse spleen would be low. Moreover, the MOG and OVA stimulations were done in the absence of adjuvant (i.e., a toll like receptor agonist). The lack of adjuvant with MOG peptide in vitro was different than how MOG peptide is used to induce disease in vivo (MOG peptide is emulsified with Complete Freund’s Adjuvant [CFA]; [15]). Despite this difference, even without adjuvant in vitro there was a small but detectable MOG-specific T cell population that could be observed with tetramer staining, which was the same population of T cells that could be stained in an EAE mouse after MOG + CFA in vivo.

Both CBD and THC were more efficacious when treated at the beginning of the culture period before peptides were initially added, and the inhibition persisted for 7 days, even if cannabinoids were not added again at the time of peptide restimulation. This is consistent with previous data from in vivo studies in which THC suppressed immune function when delivered at or near the time of initial immune stimulation [10, 11]. These results confirm that cannabinoids act by suppressing an early T cell activation event. Work in human peripheral blood mononuclear cells (PBMCs) showed that THC suppressed CD40L expression and intracellular calcium but had no effect on glycogen synthase kinase-3β (GSK3β) or phospholipase Cγ (PLCγ) [16]. Here we also showed that the intracellular calcium population was reduced by both THC and CBD. Unlike the work in human PBMCs, however, in which the intracellular calcium was assessed within the first few hours after activation with anti-CD3/28 [16], here the calcium population was shown to be suppressed 4 days after the cells were treated with CBD and THC plus MOG peptide. Thus, the current studies confirm that intracellular calcium is one early signal that is affected by cannabinoids, but the studies also demonstrate that the calcium perturbation persists.

One of the challenges with a disease like MS/EAE is that it involves multiple immune cell types that come from various tissues so no one cell type will completely mimic the immune response. Thus, in vitro approaches allow for examination of endpoints that cannot be conducted in a whole animal; the intracellular calcium measurement is one example of an assessment that would have been difficult to assess in vivo. It also would have required too many animals to evaluate T cell responses early and after a restimulation.

In these studies, TNF-α and IFN-γ were two of the most sensitive endpoints of suppression of both CBD and THC. An extensive examination in human PBMCs with CBD and toll like receptor stimuli showed that IL-6 and IL-1β were the most sensitive targets [17]. Although the two studies differ with species and means of stimulation, these data suggest that CBD has the potential to suppress various immune functions in multiple species. It is possible that this in vitro peptide stimulation could be done in human PBMCs; the difficulty lies in identifying peptides that are haplotype matched for the particular donor. For example, in some MS patients bearing DRB alleles of the major histocompatibility complex (MHC), myelin basic protein peptide (MBP; amino acids 85–99)-reactive T cells can be found [18]. Thus, in DRB-expressing PBMCs, MBP peptides might stimulate naïve T cell cultures, even in healthy individuals.

We utilized a single concentration of CBD or THC in these studies (10 μM; 3.1 μg/ml) since we evaluated several different timing parameters. This is on the higher end of what might be expected in vivo, especially if comparing blood concentrations of parent compounds since THC and CBD are extensively metabolized in the liver [19, 20]. However, it has been reported that concentrations of CBD and THC in the lymph can exceed plasma by 100–250 times [21]. As one example following a 22-day administration of 20 mg/kg CBD in humans as part of a possible treatment for epilepsy, CBD plasma levels were measured at almost 0.4 μg/ml [19], suggesting the potential for CBD in the lymph to be as high as 40 μg/ml.

Overall,, these studies demonstrate the ability to stimulate MOG-reactive T cell responses in vitro. The advantages of this model system over other in vitro models are that it is conducted in mature, naïve splenocytes, the stimulation is physiologically relevant requiring presentation by APCs, and effects of drugs and chemicals in T cells can be assessed at various times relative to stimulation. In our work, this model has provided us the means to examine effects of cannabinoids after initial peptide stimulation and restimulation, and we confirmed that CBD and THC affect an early signal that was persistent. Thus, the results confirm other studies in which the mechanism by which THC affected T cell function involved suppression of an early signal [10, 11], and demonstrated that disruption of an early signal also happened with CBD. Moreover, we used the initially treated and stimulated cells on day 0 to evaluate their ability to exhibit an intracellular calcium signal at day 4 and found that they could not produce as robust an intracellular calcium signal once they had been treated with CBD or THC. Together with other studies demonstrating cannabinoid suppression of MOG-stimulated T cell responses in vitro [6–9] and suppression of EAE in vivo [3, 9, 22–25], these results suggest that cannabinoids might provide therapeutic benefit for individuals with MS, perhaps through alleviation of (neuro)inflammation or inflammatory pain. In summary, MOG peptide-stimulated SPLC provide an antigen-specific model that has relevance to the EAE model but with much reduced animal use.

Supplementary Material

1

Supplemental Figure 1. Cytokine production from untreated cells. SPLC were seeded at 5×106 cell/well in 1 ml in 48-well plates and allowed to incubate overnight. Supernatants were used in ELISA analyses and cells were used to conduct a total cell count using the ACEA Novocyte. Cytokine concentrations were determined using a standard curve in the ELISA then values were normalized to total cell counts within each experiment. Data were then normalized within experiments to the average VH/VH on day 7 (except IL-2, which was normalized to day 4 since levels dropped over time) so that data from separate experiments could be combined. Each experiment was run with ≥3 separate wells per treatment group.

2

Supplemental Figure 2. MOG tetramer stained T cells from EAE mice. Spleens were obtained from mice that developed clinical signs from EAE. Cells were stained with the MOG-specific tetramer and the negative control. Only the cells derived from mice undergoing EAE and stained with the tetramer were positive.

Highlights.

Cytokine production from T cells can be stimulated with MOG peptide.

CBD and THC suppress cytokine production in MOG peptide-stimulated cells in vitro.

CBD and THC suppress intracellular calcium in MOG peptide-stimulated cells in vitro.