Critical Role of Mast Cells and Peroxisome Proliferator-Activated Receptor γ in the Induction of Myeloid-Derived Suppressor Cells by Marijuana Cannabidiol In Vivo

Abstract

Cannabidiol (CBD) is a natural nonpsychotropic cannabinoid from marijuana (Cannabis sativa) with anti-epileptic and anti-inflammatory properties. Effect of CBD on naive immune system is not precisely understood. In this study, we observed that administering CBD into naive mice triggers robust induction of CD11b(+)Gr-1(+) myeloid-derived suppressor cells (MDSC) in the peritoneum, which expressed functional arginase 1, and potently suppressed T cell proliferation ex vivo. Furthermore, CBD-MDSC suppressed LPS-induced acute inflammatory response upon adoptive transfer in vivo. CBD-induced suppressor cells were comprised of CD11b(+)Ly6-G(+)Ly6-C(+) granulocytic and CD11b(+)Ly6-G(-)Ly6-C(+) monocytic subtypes, with monocytic MDSC exhibiting higher T cell-suppressive function. Induction of MDSC by CBD was markedly attenuated in Kit-mutant (Kit(W/W-v)) mast cell-deficient mice. MDSC response was reconstituted upon transfer of wild-type bone marrow-derived mast cells in Kit(W/W-v) mice, suggesting the key role of cKit (CD117) as well as mast cells. Moreover, mast cell activator compound 48/80 induced significant levels of MDSC in vivo. CBD administration in mice induced G-CSF, CXCL1, and M-CSF, but not GM-CSF. G-CSF was found to play a key role in MDSC mobilization inasmuch as neutralizing G-CSF caused a significant decrease in MDSC. Lastly, CBD enhanced the transcriptional activity of peroxisome proliferator-activated receptor γ in luciferase reporter assay, and PPAR-γ selective antagonist completely inhibited MDSC induction in vivo, suggesting its critical role. Together, the results suggest that CBD may induce activation of PPAR-γ in mast cells leading to secretion of G-CSF and consequent MDSC mobilization. CBD being a major component of Cannabis, our study indicates that marijuana may modulate or dysregulate the immune system by mobilizing MDSC.

Figure 1. Induction of CD11b⁺Gr-1⁺ cells in response to CBD in vivo

A) Time-course of CD11b⁺Gr-1⁺ cell accumulation in WT peritoneum following CBD (20 mg/kg, i.p.) administration. Representative dot plots from FACS analysis are shown for the time points as indicated. B) Absolute number of MDSC calculated from frequency of CD11b⁺Gr-1⁺ cells and total viable cells in each peritoneum, mean ± SD (n=4). C) Representative dot plots from FACS analysis of cells harvested from peritonea of WT mice (n=4) 16 h after injecting with various doses of CBD as indicated showing dose-dependent induction of CD11b⁺Gr-1⁺ double positive cells. D) Mean±SD of absolute MDSC numbers from n=4 mice per group. E, F) Induction of CD11b⁺Gr-1⁺ cells by CBD is independent of TLR4. TLR4-mutant C3H/HeJ mice (n=4) injected with vehicle or 20 mg/kg CBD (i.p.). After 12 h peritoneal exudate cells were analyzed by FACS. Error bars indicate SD. Student’s t-test: *P<0.05, **P<0.01 compared vs control.

Figure 2. Comparison of in vivo cellular response to CBD and thioglycollate (TG)

WT mice (n=3) were injected (i.p.) with CBD (10 mg/kg) or 0.5 mL of 3% TG. Peritoneal exudate cells were harvested at 4 h or 12 h post-injection and analyzed side-by-side by FACS for Gr-1 (Ly6-G) expression with same voltage settings. Representative histograms are shown (A). B) Mean fluorescence intensities (MFI) for Gr-1 expression. C) Absolute cell numbers per PM for CBD-MDSC (Gr-1^int) and TG-neutrophils (Gr1^high). D) Syngenic lymph node-derived T cells were stimulated with ConA and co-cultured without or with enriched Gr-1+ cells. Data represent mean ± SD from 3 mice (B, C) or from triplicate determinations (D).

Figure 3. Functional characterization of CD11b⁺Gr-1⁺ cells induced by CBD

Expression of functional Arginase: A) Western blot analysis of lysates of cells harvested from the peritonea, 16 h following the administration of vehicle or 20 mg/kg CBD (n=4); B) Spectrophotometric assay for arginase activity. Cell lysates were analyzed for arginase activity using L-arginine as the substrate, and detecting the L-ornithine formed. Data represent mean ± SD. Student’s t-test, **P<0.01. C) T cell suppression assay in vitro. CD11b⁺Gr-1⁺ cells harvested from the peritoneum of WT mice were purified to >90% purity, irradiated, plated at 1:10 and 1:2 (MDSC:T cell) ratios with syngenic T cells (2×10⁵ T cells/ well) and stimulated with ConA. T cells cultured without MDSC but stimulated with ConA served as control (TC). Arginase inhibitor (nor-NOHA) was added to some wells at 10 or 100 μM as indicated. Proliferation was assessed at 72 h by [3H]thymidine assay. Data is mean ± SD of quadruplicate determinations and representative of two experiments. Student’s t-test: **P<0.01, ***P<0.001 compared to TC control; ^†P<0.05 compared vs corresponding MDSC:TC without nor-NOHA. D) Suppression of acute inflammatory response to LPS in vivo. Purified CBD-induced CD11b⁺Gr-1⁺ cells (5×10⁶/ mouse) were adoptively transferred into naïve WT mice two hours before injecting with LPS. One hour after LPS challenge, TNF-α levels in sera were analyzed by ELISA. Data represent mean ± SD (n=3 mice); Student’s t-test: ***P<0.001, **P<0.01.

Figure 4

A) CBD induces mobilization of CD11b⁺Gr-1⁺ MDSCs from BM. Representative dot plots of peritoneal and bone marrow cells harvested 0 or 12 h after CBD (20 mg/kg) administration showing frequency (%) CD11b⁺Gr-1⁺ cells (gated). B) Peritoneal cells from CBD injected mice harvested after 12 h were analyzed for CD31 and Ki-67 expression on MDSCs by triple staining along with CD11b and Gr-1. Histograms show expression of CD31 and Ki-67 on gated CD11b⁺Gr-1⁺ MDSC population (open histograms). Filled histograms represent isotype Ab staining controls. C) Phenotypic analysis for the expression of other markers in peritoneal cells induced by CBD. Peritoneal cells harvested after 12 h from vehicle or CBD (20 mg/kg) injected mice were stained for indicated markers and analyzed by FACS. The frequency (%) of positive population is indicated. D, E) Spleen cells harvested after 12 h following vehicle or CBD administration were analyzed for CD11b⁺Gr-1⁺ MDSC by FACS. Representative dot plots showing frequency and mean absolute MDSC numbers (n=3 mice) shown. Error bars indicate SD. ***P<0.001, **P<0.01, * P<0.05 based on Student’s t test (CBD vs Veh control).

Figure 5. Analysis of CBD-induced MDSC subtypes

WT mice (n=4) were injected with vehicle or 20 mg/kg CBD and after 16 h peritoneal exudate cells were triple-stained for CD11b, Ly6-G and Ly6-C and analyzed by FACS. Representative dot plots with gated CD11b⁺Ly6-G⁺Ly6-C+^(int) granulocytic (Gr) and CD11b⁺Ly6-G–^(neg)Ly6-C⁺ monocytic (Mo) MDSC subtypes are shown with frequencies indicated (A). Absolute numbers of each sub type is calculated based on frequency and total cell numbers and represented as mean ± SD from 4 mice (B). Student’s t test, **P<0.01 compared to vehicle control. C) MDSC sub types were purified by FACS sorting and used in T cell suppression assay at indicated ratios with syngenic T cells stimulated with ConA. T cell proliferation was assessed at 72 h by [3H]thymidine incorporation. Mean ± SD of quadruplicate determinations and representative of two separate experiments shown. *P<0.05, Student’s t test.

Figure 6. Attenuated induction of MDSC in mast cell-deficient mice

FACS analysis of CD11b⁺Gr-1⁺ MDSC from peritoneum 12 h after injecting with vehicle or CBD in WT or mast cell-deficient Kit^W/W-v mice (n=3). Representative dot plots are shown (A). Absolute MDSC cell number is represented as mean ± SD from 3 mice (B). **P<0.01, Student’s t test. C) Cells were analyzed for MDSC subtypes by flow cytometry and frequency of each subset, namely CD11b⁺Ly6-G⁺Ly6-C+^(int) granulocytic (Gr) and CD11b⁺Ly6-G–^(neg)Ly6-C⁺ monocytic (Mo) MDSC are shown.

Figure 7. Analysis of chemokines

CBD (20 mg/kg) was injected i.p. into groups of WT mice (n=4) for each time point. G-CSF, GM-CSF, CXCL1 and M-CSF levels in the peritoneal exudates were analyzed by ELISA (A-C). Blocking experiment with anti-G-CSF in vivo: WT mice (n=3) were injected with isotype control IgG or anti-G-CSF Ab (10μg/mouse) 1 h before injecting with CBD (20mg/kg). Peritoneal exudate cells were harvested after 12 h, and analyzed by FACS for MDSC. Representative dot plots are shown for each treatment (D); Absolute number of MDSC per peritoneum (n=3 mice) (E). F) G-CSF levels determined by ELISA in the peritoneal exudates of WT and mast cell deficient Kit^W/W-v mice 16 h after injection with 20 mg/kg CBD. Error bars indicate SD. Student’s t-tests: **P<0.01, *P<0.05.

Figure 8. Adoptive transfer of WT mast cells restores CBD response in mast-cell deficient mice

Bone marrow-derived mast cells (BMMC) were generated by culturing WT bone marrow cells in the presence of recombinant mouse IL-3 and stem cell factor as described in methods. Purity of BMMC after 5 weeks of culture as determined by FACS analysis for c-Kit expression (A); Left histogram isotype control IgG, right histogram c-Kit Ab. BMMC (6×10⁶) were adoptively transferred i.v. into mast cell-deficient Kit^W/W-v mice. Littermates (+/+) were used as WT controls. Mast cells were allowed to engraft and six weeks after transfer, mice with or without adoptively transferred mast cells were injected with vehicle or CBD (i.p.). After 16 h, peritoneal exudate cells were analyzed by flow cytometry (B, C). Peritoneal lavage fluids were analyzed for G-CSF by ELISA (D). Data represent mean ± SD (n=3 mice per group). Student’s t- test: *P<0.05. Successful engraftment of adoptively transferred mast cells was determined by Giemsa staining (Table I).

Figure 9. Accumulation of MDSC in response to compound-48/80

WT mice (n=4) were injected with vehicle or different doses of c-48/80 i.p. and peritoneal exudate cells were analyzed by flow cytometry for MDSC. Representative dot plots with frequency of gated CD11b⁺Gr-1⁺ MDSC are shown (A). Absolute MDSC numbers from 4 mice are represented as mean ± SD (B). C, D) Flow cytometric analysis for MDSC subtypes as described before. Student’s t test, **P<0.01 compared to vehicle control.

Figure 10. PPARγ plays a critical role in the induction of MDSC by CBD in vivo

Flow cytometric analysis of MDSC from peritoneum 12 h after injecting with vehicle or CBD in WT or Trpv1^−/− (VR1-KO) mice. Frequencies (A) and absolute numbers of MDSC (B) are depicted. Data represent mean ± SD (n=3 mice). Flow cytometric analysis of MDSC from peritoneum 12 h after injecting with vehicle or CBD in WT mice with or without pretreatment of mice using specific inhibitor (BADGE) to block PPARγ receptors (C, D). Data represent Mean ± SD (n=4 mice). Student’s t-test: ***P<0.001. E) Murine mast cells secrete G-CSF in response to CBD in a PPARγ-dependent manner. Normal murine cloned mast cells (MC/9) were treated in culture (10⁶ cells/well) with 1 or 10 μM CBD. In some wells PPARγ-inhibitor BADGE was added at 1 or 10 μM as indicated. Culture supernatants were harvested after 24 h and analyzed for G-CSF by ELISA. Data represent mean ± SD of triplicate determinations and representative of two experiments. Student’s t-test: **P<0.01, *P<0.05.

Figure 11. CBD enhances PPARγ transcriptional activity

A) Expression of PPARγ on mast cells. PPARγ message was analyzed on murine cloned mast cells (MC/9) and P815 mast cell line by RT-PCR. B) PPARγ-luciferase reporter and control vector used in the experiment. Consensus binding site sequence in the PPAR transcriptional response element (TRE) repeats is shown. C) Murine P815 cells were transfected with inducible PPAR-Luciferase (Firefly) reporter along with constitutive Renilla-luciferase construct followed by treatment with vehicle (control), CBD (0.1 μM and 1 μM), or known PPARγ agonist Troglitazone (10 μM). Transcriptional activity of PPARγ was measured by luciferase assay by luminometry. Relative luciferase units were normalized using Renilla construct and fold change was calculated. Mean ± SD of triplicate determinations from a representative experiment are shown. Student’s t-test, *P<0.05 compared to control.

Figure 12. Proposed mechanism induction of MDSC in vivo by Cannabidiol

Induction of functional MDSC from BM precursors by CBD and their accumulation in the periphery appears to be primarily mediated by chemokines, predominantly G-CSF with likely involvement of mast cells (MC) and PPARγ in the process.