Targeted deletion of cannabinoid receptors CB1 ... - Semantic Scholar

Targeted deletion of cannabinoid receptors CB1 and CB2 produced enhanced inflammatory responses to influenza A/PR/8/34 in the absence and presence of ⌬9-tetrahydrocannabinol John P. Buchweitz,*,† Peer W. F. Karmaus,† Kurt J. Williams,‡ Jack R. Harkema,†,‡ and Norbert E. Kaminski*,†,1 Departments of *Pharmacology and Toxicology and ‡Pathobiology and Diagnostic Investigation and †Center for Integrative Toxicology, Michigan State University, East Lansing, Michigan, USA

Abstract: We have previously reported that ⌬-9tetrahydrocannabinol (⌬9-THC)-treated mice challenged with influenza virus A/PR/8/34 (PR8) developed increased viral hemagglutinin 1 (H1) mRNA levels and decreased monocyte and lymphocyte recruitment to the pulmonary airways when compared with mice challenged with PR8 alone. The objective of the present study was to examine the role of cannabinoid (CB1/CB2) receptors in mediating the effects of ⌬9-THC on immune and epithelial cell responses to PR8. In the current study, ⌬9-THC-treated CB1/CB2 receptor null (CB1ⴚ/ⴚ/ CB2ⴚ/ⴚ) and wild-type mice infected with PR8 had marked increases in viral H1 mRNA when compared with CB1ⴚ/ⴚ/CB2ⴚ/ⴚ and wild-type mice challenged with PR8 alone. However, the magnitude of the H1 mRNA levels was greatly reduced in CB1ⴚ/ⴚ/CB2ⴚ/ⴚ mice as compared with wild-type mice. In addition, ⌬9-THC-treated CB1ⴚ/ⴚ/CB2ⴚ/ⴚ mice infected with PR8 had increased CD4ⴙ T cells and IFN-␥ in bronchoalveolar lavage fluid with greater pulmonary inflammation when compared with ⌬9-THC-treated wild-type mice infected with PR8. ⌬9-THC treatment of CB1ⴚ/ⴚ/CB2ⴚ/ⴚ mice in the presence or absence of PR8 challenge also developed greater amounts of mucous cell metaplasia in the affected bronchiolar epithelium. Collectively, the immune and airway epithelial cell responses to PR8 challenge in ⌬9-THC-treated CB1ⴚ/ⴚ/CB2ⴚ/ⴚ and wild-type mice indicated the involvement of CB1/CB2 receptor-dependent and -independent mechanisms. J. Leukoc. Biol. 83: 785–796; 2008.

and pyrolysis products of smoked marijuana or via oral consumption, which can include a synthetically derived therapeutic for the treatment of AIDS-induced wasting or chemotherapy-induced emesis (e.g., dronabinol). Despite its highly debated therapeutic use, it is well established that the plantderived CB ⌬9-THC has immunosuppressive properties. CBs, such as ⌬9-THC, have been found to increase host susceptibility to pathogens in a number of in vivo experimental models [1], including HSV [2, 3], Listeria monocytogenes [2], Legionella pneumophila [4], and more recently, influenza virus A/PR/8/34 (PR8) [5]. The mechanism for increased susceptibility to pathogens appears to be correlated with the suppression of cell-mediated immune responses by ⌬9-THC, which suppresses immune cell function by decreasing cytokine secretion and expression of costimulatory molecules by APC [6], suppressing the migration of monocytes and T cell subsets (CD4⫹ and CD8⫹ T cells) into the infected tissue [5], altering the cytoplasmic polarization of cytolytic T cells toward pathogen-infected cells [3], and suppressing the release of cytokines, such as IL-12 and IFN-␥, which are essential for competent, cell-mediated immunity regulated by TH1 cells [7]. Additionally, in a mouse model of allergic airway disease, ⌬9-THC attenuated the TH2 cytokine response to an inhaled allergen, decreasing mRNA expression of the cytokines IL-2, IL-4, IL-5, and IL-13 elicited by OVA challenge [8]. Production of IL-2 is especially crucial to T cell-dependent responses via its role in T cell clonal expansion. Accordingly, ⌬9-THC and other CB compounds have been shown to suppress lymphocyte proliferation in vitro [9] as well as markedly suppress IL-2 expression, in part, through inhibition of the NFAT [10, 11]. The precise mechanism by which ⌬9-THC modulates APC and T cell function as well as the role of CB1 and CB2 have remained elusive.

Key Words: mice 䡠 lung 䡠 influenza 䡠 inflammation

INTRODUCTION In humans, the primary routes of exposure to ⌬-9-tetrahydrocannabinol (⌬9-THC) occur through inhalation as a complex mixture of chemicals, including cannabinoid (CB) compounds 0741-5400/08/0083-785 © Society for Leukocyte Biology

1 Correspondence: Department of Pharmacology and Toxicology, 315 National Food Safety and Toxicology Building, Michigan State University, East Lansing, MI 48824-1317, USA. E-mail: [email protected] Received September 11, 2007; revised November 2, 2007; accepted November 6, 2007. doi: 10.1189/jlb.0907618

Journal of Leukocyte Biology Volume 83, March 2008 785

We have recently reported [5] that ⌬9-THC increased host susceptibility to influenza virus PR8, a hemagluttinin 1, neuraminadase 1 (H1N1) influenza strain, in the C57BL/6 mouse. Influenza virus is a common respiratory pathogen that infects epithelial cells lining the respiratory tract, as well as mononuclear cells [12]. During the course of infection, inflammatory cells comprised primarily of lymphocytes and neutrophils infiltrate the pulmonary airways [13]. Viral clearance is achieved through immunocompetent, virus-specific T cell responses. We have recently reported that treatment of mice with increasing doses of ⌬9-THC led to a dose-dependent decrease in recruitment of monocytes and lymphocytes (CD4⫹ and CD8⫹ T cells) with a concomitant increase in viral load at 7 days postinfection (dpi) [5]. These results suggest that ⌬9-THC disrupted cell-mediated immune responses that are critically involved in viral clearance and recovery from influenza infection. Two subtypes of CB receptors have been characterized: CB1 (brain receptor) and CB2 (peripheral receptor). The CB2 receptor is predominantly expressed in cells of the immune system, including neutrophils, monocytes, NK cells, T cells, and B cells [14, 15], and has been thought to mediate many of the immunomodulatory effects produced by CBs. In particular, the CB-induced inhibition of antigen processing and presentation in macrophage [16, 17] and antitumor activity in a murine model of lung cancer [18] has been shown to be mediated by the CB2 receptor. There is, however, emerging evidence for CB-mediated effects on cytokine production pertinent to Tdependent humoral and cell-mediated immune responses that occur independently of CB1/CB2 receptors. Specifically, the suppression of IL-2 production by a variety of immunosuppressive CBs is not attenuated by the CB receptor antagonists SR141716A and SR144528 [19]. Likewise, the endogenous CB (endoCB), 2-arachidonyl glycerol, suppressed PMA/ionomycin-elicited IFN-␥ expression similarly in splenocytes obtained from CB receptor null mice (CB1⫺/⫺/CB2⫺/⫺) as compared with splenocytes from wild-type mice [20]. Additional evidence for CB receptor-independent effects of CBs is illustrated by the plant-derived CB, cannabidiol, which possesses weak affinity for the CB receptors CB1 and CB2, yet exhibits potent antiinflammatory activity including the reduced production of PGE2, NO and other oxygen-derived free radicals in response to carrageenan-induced inflammation [21], inhibition of the release of IL-1, TNF-␣, and IFN-␥ by PBMC [22], and suppressed MIP-1␣, MIP-1␤, and IL-8 in the human T cell lymphotropic virus 1 B cell line [23]. Collectively, these studies illustrate different aspects of host immunity that are CB receptor (CB1 and CB2)-dependent and -independent. The objective of the present study was to evaluate the role of CB1 and CB2 receptors in the suppression of cell-mediated immunity in mice infected with PR8 by ⌬9-THC. Accordingly, CB1⫺/⫺/CB2⫺/⫺ mice and background strain C57BL/6 mice were compared by using a mouse model of host resistance to infection with murine-adapted PR8. Measurements of viral load and immune cell recruitment and secretion of cytokines in the airways were determined. In addition, previously characterized, morphologic changes in airway epithelium, including apoptosis and mucous cell metaplasia (MCM), were quantified as a measure of viral and immune cell-mediated, pathologic 786

Journal of Leukocyte Biology Volume 83, March 2008

outcomes [5]. Results from this study suggest that ⌬9-THC is modulating immunological and pulmonary airway responses to influenza virus through CB receptor (CB1 and CB2)-dependent and -independent mechanisms.

MATERIALS AND METHODS Animals Two independent experiments were conducted as a part of this study. In the first experiment, 32 female C57BL/6 mice (8 –10 weeks old, from Charles River, Portage, MI, USA) and 32 female CB1⫺/⫺/CB2⫺/⫺ mice (8 –16 weeks old), bred at Michigan State University (MSU; East Lansing, MI, USA) from breeding pairs provided by Dr. Andreas Zimmer [24], were randomly assigned to one of four experimental groups (n⫽8 per group) by genotype. The mice were transferred to plastic cages containing sawdust bedding (four mice per cage) and quarantined for 1 week in an animal holding room that was maintained at 21–24°C and 40 – 60% relative humidity with a 12-h light/dark cycle. The mice were provided food (Purina Certified Laboratory Chow) and water ad libitum and not used for experimentation until their body weight was 17–20 g. In the second experiment, 30 female C57BL/6 mice (8 –10 weeks old) and 34 female CB1⫺/⫺/CB2⫺/⫺ mice (8 –16 weeks old) were randomly assigned to one of four experimental groups (n⫽5–14 per group) by genotype. CB1⫺/⫺/ CB2⫺/⫺ mice were subjected to an extensive battery of serological screens prior to breeding, as well as designated offspring sentinels after breeding. Tests for all pathogens assayed were negative. All mice used in this study were kept free of pathogens and respiratory disease and used in accordance with guidelines set forth by the Institutional Animal Care and Use Committee at MSU.

⌬9-THC ⌬9-THC was provided by the National Institute on Drug Abuse (Bethesda, MD, USA) as a resin, which was solubilized in corn oil at concentrations that yielded a dose of 75 mg/kg for administration to mice by oral gavage in a volume of 100 ␮l.

Influenza PR8 instillation The laboratory of Dr. Alan Harmsen (Montana State University, Bozeman, MT, USA) generously supplied influenza PR8. Mice were anesthetized with 4% isoflurane in oxygen, and 50 ␮l PR8 in pyrogen-free saline was instilled as 25 ␮l per nare at a total dose of 50 pfu.

Experimental design In both experiments, mice were orally administered ⌬9-THC for 5 consecutive days at a dose of 75 mg/kg. On the 3rd day of dosing, mice were instilled intranasally with influenza virus (50 pfu) 4 h prior to ⌬9-THC administration. All mice were killed at 7 dpi. In the first experiment, the right lung lobes were ligated and removed for RNA analysis, and the left lung lobe was lavaged with saline for cellular and biochemical measurements and then fixed in 10% neutral-buffered formalin for immunohistochemical analysis as described below. In the second experiment, the entire lung was immersed in TRI reagent (Sigma Chemical Co, St. Louis, MO, USA) for RNA isolation and analysis. These experiments required the breeding of a sufficient number of female mice to support a large n value per treatment group. As breeding was a limiting factor in the design of these experiments, it was deemed necessary in the initial experiment to divide the right and left lung lobes to perform mRNA, bronchoalveolar lavage fluid (BALF), and immunohistochemical measurements. The assumption was made that the virus and associated immune response would be equally distributed between the right and left lung lobes and that there would be uniform and most importantly, discernible differences between treatment with ⌬9-THC or vehicle. The trends in the results were similar to those observed previously [5]; however, a second experiment was performed to provide clarity in our measurement of viral hemagglutinin 1 (H1) mRNA for whole lung homogenates.

http://www.jleukbio.org

Necropsy, lavage collection, and tissue preparation Mice were first anesthetized by an i.p. injection of 0.1 ml 12% pentobarbital solution, and then a midline laparotomy was performed, and mice were exsanguinated by cutting the abdominal aorta. Immediately after death, the trachea was exposed and cannulated, and the heart and lung were excised en bloc. The right lung lobes were ligated, removed, and immersed in TRI reagent and stored at – 80°C until RNA was isolated. The left lung lobe was lavaged with 1 ml sterile saline, instilled through the tracheal cannula, and withdrawn to recover BALF. The left lung lobe was then inflated under constant pressure (30 cm H2O) with 10% neutral-buffered formalin (Sigma Chemical Co.) for 1 h. The tracheal airway was ligated, and the inflated lobes were stored in a large volume of the same fixative for at least 24 h until further processing. For the second study, the entire lung was immersed in TRI reagent and stored at – 80°C until RNA was isolated.

RNA isolation Total RNA was isolated from the lung using TRI reagent (Sigma Chemical Co.). The evaluation of the relative levels of H1 mRNA were determined using the TaqMan real-time multiplex RT-PCR with custom-designed TaqMan primers and probe to the target gene and the manufacturer’s predeveloped primers and probe to 18S (Applied Biosystems, Foster City, CA, USA). The primers and probe to the target gene and endogenous reference gene were designed specifically to exclude detection of genomic DNA. Aliquots of isolated tissue RNA (1 ␮g total RNA) were converted to cDNA using random hexamers (Invitrogen, Carlsbad, CA, USA). The resultant cDNA (2 ␮l) was added to a reaction mixture that consisted of the target gene primers and probe, endogenous reference primers and probe (18S rRNA), and Taqman Universal Master Mix to a final volume of 30 ␮l. Following PCR, amplification plots (change in dye fluorescence vs. cycle number) were examined, and a dye fluorescence threshold within the exponential phase of the reaction was set separately for the target gene and the endogenous reference (18S). The cycle number at which each amplified product crosses the set threshold represents the comparative threshold cycle (CT) value. The amount of target gene normalized to its endogenous reference was calculated by subtracting the endogenous reference CT from the target gene CT (⌬CT). Relative mRNA expression was calculated by subtracting the mean ⌬CT of the control samples from the ⌬CT of the treated samples (⌬⌬CT). The amount of target mRNA, normalized to the endogenous reference and relative to the calibrator (i.e., RNA from control), is calculated by using the formula: 2–⌬⌬CT.

BALF cellularity The total number of leukocytes in BALF was counted with a hemacytometer. In brief, 10 ␮l BALF supernatant was added to a hemacytometer, and the number of leukocytes in the four-etched corner quadrants was counted and multiplied by 2500 to yield the number of leukocytes per ml of sample. The percent of total leukocytes consisting of eosinophils, lymphocytes, monocytes, and neutrophils was determined from counts of 200 cells in a cytospin sample stained with Diff-Quick (Dade Behring, Newark, DE, USA). The percentage of eosinophils, lymphocytes, monocytes, and neutrophils was multiplied by the total number of leukocytes determined from the hemacytometer to yield the respective number of each cell type per ml of sample.

Total protein Total protein was quantified in BALF using the bicinchoninic acid (BCA) method as provided by the manufacturer (Pierce, Rockford, IL, USA). In brief, BALF samples were centrifuged at 300 g for 5 min to pellet cellular debris. Supernatants were removed and stored at –20°C until testing was performed. BALF supernatant (25 ␮l) or BSA protein standard (25–2000 ␮g/ml) was incubated in a 96-well microplate with 200 ␮l of a 50:1 mixture of reagents A (sodium carbonate, sodium bicarbonate, BCA, and sodium tartrate in 0.1 M sodium hydroxide) and B (4% cupric sulfate) for 30 min at 37ºC. Samples were read on a Biotek microplate reader at 562 nm.

Inflammatory and TH1/TH2 cytokines BALF samples were centrifuged at 300 g for 5 min to pellet cellular debris. BALF supernatants were stored at –20°C until testing was performed. The

cytokines IL-6, IL-10, MCP-1, IFN-␥, TNF-␣, and IL-12p70 were detected simultaneously by using the cytometric bead array (CBA) mouse inflammation kit (BD PharMingen, San Diego, CA, USA), and the TH1/TH2 cytokines IL-2, IL-4, IL-5, IFN-␥, and TNF-␣ were detected simultaneously by using the mouse TH1/TH2 kit. In brief, 50 ␮l BALF supernatant from each sample was incubated individually with a mixture of capture beads and 50 ␮l PE detection reagent consisting of PE-conjugated anti-mouse IL-6, IL-10, MCP-1, IFN-␥, TNF-␣, and IL-12p70 or 50 ␮l PE detection reagent consisting of PEconjugated anti-mouse IL-2, IL-4, IL-5, IFN-␥, and TNF-␣. The samples were incubated at room temperature for 2 h in the dark. After incubation, samples were washed once and resuspended in 300 ␮l wash buffer before analysis on a BD FACSCalibur flow cytometer. Data were analyzed using CBA software (BD PharMingen). Standard curves were generated for each cytokine using the mixed cytokine standard provided with the respective kits. The concentration of each cytokine was determined by interpolation from the corresponding standard curve that spanned concentrations ranging from 20 to 5000 pg/ml for each cytokine measured.

T-lymphocyte flow cytometry After counting the number of cells retrieved in BALF, cells were pelleted by centrifugation at 300 g for 5 min, supernatants were removed, and cells were reconstituted in 150 ␮l flow cytometer (FCM) buffer [PBS supplemented with 2% (w/v) BSA and 0.09% (w/v) sodium azide] with purified anti-mouse CD16/CD32 (Fc␥RIII/II) antibody (BD Biosciences, San Jose, CA, USA) to block for 30 min. The samples were then divided into two groups of equal volume, washed, and reconstituted in FCM buffer. One group received antibodies directed against CD3 (APC anti-mouse CD3ε), CD4 [PE anti-mouse CD4 (L3T4)], and CD8 [FITC anti-mouse CD8a (Ly-2)], and the other group received the isotype control antibodies. Samples were allowed to incubate 1 h at 4°C, washed twice, and then fixed for 10 min with Cytofix (BD Biosciences). Samples were then washed again and reconstituted to a volume of 300 ␮l for analysis on the BD FACSCalibur flow cytometer. The total number of events taken per each sample was 3000.

Microdissection The intrapulmonary airways of the fixed left lung lobe from each mouse were microdissected according to a modified version of the technique of Plopper et al. [25]. Beginning at the lobar bronchus, airways were split down the long axis of the main axial airway through the 12th airway generation. Two transverse tissue blocks were excised at the level of the 5th (proximal) and 11th (distal) airway generation. Tissue blocks were embedded in paraffin, sectioned at a thickness of 5 ␮m, and then stained with H&E for light microscopic examination. Other paraffin sections were stained with alcian blue (pH⫽2.5)/periodic acid Schiff (AB/PAS) to detect neutral and acidic mucosubstances in the respiratory epithelium lining the intrapulmonary-conducting airways (large and small diameter bronchioles).

Immunocytochemistry Hydrated paraffin sections (5– 6 ␮m-thick) from formalin-fixed lung tissues were treated with 0.05% proteinase K for 2 min and washed with 1 N HCI for 1 h. Sections were then treated with 3% H2O2 (in methanol) to block endogenous peroxidase and then incubated with a polyclonal antibody to caspase-3 (CAS-3; Abcam, Cambridge, MA, USA) at 20 ␮g/ml for 1 h. Immunoreactive CAS-3 was visualized with the Vectastain Elite ABC kit (Vectastain Laboratories Inc., Burlingame, CA, USA) using 3⬘,3⬘-diaminobenzidine tetrahydrochloride (Sigma Chemical Co.) as a chromagen.

Histopathology scores for inflammation A histopathology score was established based on a previously established scoring system [5]. In brief, the scores assigned were: 0 ⫽ No inflammation; 1 ⫽ mild, inflammatory cell infiltrate of the perivascular/peribronchiolar compartment; 2 ⫽ moderate, inflammatory cell infiltrate of the perivascular/ peribronchiolar space with modest extension into the alveolar parenchyma; and 3 ⫽ severe, inflammatory cell infiltrate of the perivascular/peribronchiolar space with large inflammatory foci found in the alveolar parenchyma. A board-certified pathologist scored the two left lung sections (generations 5 and 11) independently and without prior knowledge of treatment groups. A mean score with the SEM was calculated for each treatment group.

Immune responses of CB1/CB2 receptor-null mice to PR8

787

CAS-3 and mucosubstance numeric cell densities Slides of lung sections stained immunohistochemically for CAS-3 or stained for mucosubstances (AB/PAS) were examined. Numeric cell densities were determined for epithelial cells immunohistochemically reactive to CAS-3 via light microscopy by counting the number of nuclear profiles of these immunoreactive epithelial cells lining the bronchiolar epithelium at generation 5 and dividing by the length of the underlying basal lamina. Numeric cell densities for CAS-3 were expressed as the number of CAS-3-positive epithelial cells per mm basal lamina. In a similar manner, numeric cell densities were determined for epithelial cells staining with AB/PAS. The numeric cell density of epithelial cells staining for mucosubstances was expressed as the number of AB/ PAS-positive epithelial cells per mm basal lamina.

Statistical analysis Data are expressed as mean ⫾ SEM. Outliers were identified and excluded from sample sets by using Grubb’s test. The differences in outcomes between treatment groups within each genotype were analyzed by two-way ANOVA with multiple comparisons made by the Student Newman Keuls post hoc test using SigmaStat software, Version 2.03, from Jandel Scientific (San Rafael, CA, USA). Differences between treatment groups and genotypes were made using a Student’s t-test. The criterion for significance was taken to be P ⬍ 0.05.

RESULTS Qualitative health assessment of CB1⫺/⫺/CB2⫺/⫺ and wild-type mice challenged with PR8 CB1⫺/⫺/CB2⫺/⫺ and wild-type mice treated with corn oil or ⌬9-THC and intranasally instilled with saline were normal in appearance and activity level. Infection with PR8 in the presence or absence of ⌬9-THC treatment led to marked differences in the gross appearance of CB1⫺/⫺/CB2⫺/⫺ and wildtype mice. Specifically, CB1⫺/⫺/CB2⫺/⫺ mice were notably more gaunt in comparison with wild-type mice, suggesting that the mice were dehydrated. In addition, the CB1⫺/⫺/CB2⫺/⫺ mice were lethargic and displayed unkempt fur. In contrast, wild-type mice infected with PR8 in the presence or absence of ⌬9-THC treatment maintained normal grooming habits and exhibited similar activity levels as the saline-instilled controls. Upon gross examination of the lungs, CB1⫺/⫺/CB2⫺/⫺ mice had extensive hemorrhage throughout all lung lobes. Wild-type mice also displayed evidence of hemorrhage, albeit to a much lesser extent than CB1⫺/⫺/CB2⫺/⫺ mice.

The viral load of PR8 in the pulmonary airways of CB1⫺/⫺/CB2⫺/⫺ and wild-type mice on Day 7 after challenge To estimate the viral load of PR8 infection, the levels of mRNA for the viral surface protein, H1, were assessed by quantitative real-time PCR. For CB1⫺/⫺/CB2⫺/⫺ and wild-type mice, viral H1 mRNA was elevated markedly above detection levels set by saline controls at 7 dpi in the lungs of all mice challenged with PR8 (Fig. 1). However, vehicle-treated wild-type mice exhibited markedly greater viral H1 mRNA levels than vehicletreated CB1⫺/⫺/CB2⫺/⫺ mice. Infection of ⌬9-THC-treated CB1⫺/⫺/CB2⫺/⫺ mice with PR8 resulted in an increase (P⫽0.056) in H1 mRNA levels when compared with mice instilled with PR8 alone. In ⌬9-THC-treated wild-type mice infected with PR8, there was a significant increase in viral H1 788


Fig. 1. Pulmonary expression levels of H1 mRNA. The effects of ⌬9-THC on H1 mRNA levels in lungs from CB1⫺/⫺/CB2⫺/⫺ and wild-type mice challenged with PR8. The mRNA levels for H1 were determined by real-time PCR with 18S used as an internal loading control. Data represent the mean ⫾ SEM, presented as the fold change in H1 mRNA levels normalized to those of mice instilled with saline. *, A significant difference between corn oil-treated mice infected with PR8 and respective controls instilled with saline; #, a significant effect elicited by ⌬9-THC-treated mice infected with PR8 when compared with corn oil-treated mice infected with PR8 by ANOVA; ⫹, a significant difference between CB1⫺/⫺/CB2⫺/⫺ and wild-type mice within a respective treatment group.

mRNA levels when compared with wild-type mice instilled with PR8 alone. The overall level of expression of H1 mRNA was markedly less in CB1⫺/⫺/CB2⫺/⫺ mice when compared with wild-type mice in the absence or presence of ⌬9-THC treatment.

⌬9-THC enhanced vascular permeability induced by PR8 infection of the airways in CB1⫺/⫺/CB2⫺/⫺ and wild-type mice The detection of increased levels of total BALF protein is indicative of increased vascular permeability at the alveolar/ capillary interface during the inflammatory response to PR8. The total BALF protein levels in PR8-infected CB1⫺/⫺/CB2⫺/⫺ and wild-type mice, in the absence of ⌬9-THC treatment, were two-fold greater than saline controls (Fig. 2). ⌬9-THC-treated CB1⫺/⫺/CB2⫺/⫺ mice infected with PR8 exhibited a 33% increase in total BALF protein when compared with CB1⫺/⫺/ CB2⫺/⫺ mice challenged with PR8 alone. Conversely, ⌬9THC-treated wild-type mice infected with PR8 had a 50% decrease in total BALF protein when compared with wild-type mice challenged with PR8 alone. When comparing CB1⫺/⫺/ CB2⫺/⫺ and wild-type mice, the amount of total protein detected in mice treated with ⌬9-THC and infected with PR8 was two-fold greater in CB1⫺/⫺/CB2⫺/⫺ mice than amounts of total protein observed in wild-type mice receiving the same treatment.

CB1⫺/⫺/CB2⫺/⫺ mice treated with ⌬9-THC exhibit a distinctly different composition of leukocytes recruited to the airways in response to PR8 infection when compared with wild-type mice Primary influenza infection elicits an inflammatory response consisting of a mixed population of leukocytes that infiltrates the pulmonary airways. In particular, there is a marked influx of http://www.jleukbio.org

Fig. 2. Pulmonary vascular leakage assessed by total protein in BALF. The effects of ⌬9-THC on total protein detected in BALF supernatants from CB1⫺/⫺/CB2⫺/⫺ and wild-type mice challenged with PR8. Data represent the mean ⫾ SEM, presented as the amount of protein in ␮g/ml. *, A significant difference between corn oil-treated mice infected with PR8 and respective controls instilled with saline; ⫹, a significant difference between CB1⫺/⫺/CB2⫺/⫺ and wild-type mice within a respective treatment group.

neutrophils and lymphocytes into the airways with monocytes and eosinophils, representing a smaller portion of the total population of BALF leukocytes. In the current study, the total number of leukocytes in BALF was three-fold greater in PR8-infected wildtype mice and four-fold greater in CB1⫺/⫺/CB2⫺/⫺ mice than in their respective noninfected controls (Fig. 3A). ⌬9-THC-treated mice infected with PR8 exhibited a two-fold increase above saline-instilled controls in CB1⫺/⫺/CB2⫺/⫺ and wild-type mice. However, the difference in total leukocytes retrieved in the ⌬9-

THC-treated CB1⫺/⫺/CB2⫺/⫺ mice infected with PR8 was significantly less than CB1⫺/⫺/CB2⫺/⫺ mice challenged with PR8 alone. To further assess differences in individual leukocyte populations retrieved by lavage, differential cell counts were performed. Neutrophils (Fig. 3B) were increased markedly in PR8infected CB1⫺/⫺/CB2⫺/⫺ and wild-type mice. Specifically, ⌬9THC-treated wild-type mice infected with PR8 exhibited two-fold increases in neutrophils when compared with wild-type mice challenged with PR8 alone. Conversely, there were marked decreases in the number of BALF neutrophils for PR8-infected CB1⫺/⫺/CB2⫺/⫺ mice treated with ⌬9-THC when compared with CB1⫺/⫺/CB2⫺/⫺ mice challenged with PR8 alone. BALF lymphocytes were increased markedly (Fig. 3C) with PR8 infection in CB1⫺/⫺/CB2⫺/⫺ and wild-type mice. ⌬9-THC-treated CB1⫺/⫺/ CB2⫺/⫺ and wild-type mice infected with PR8 exhibited attenuation in the number of lymphocytes retrieved. There was no change in the number of monocytes retrieved in BALF from PR8-challenged, wild-type mice as compared with saline controls. A two-fold increase was observed, however, in the number of BALF monocytes from PR8-infected CB1⫺/⫺/CB2⫺/⫺ mice that was attenuated in ⌬9-THC-treated CB1⫺/⫺/CB2⫺/⫺ mice infected with PR8 (Fig. 3D). Lastly, PR8 infection of CB1⫺/⫺/CB2⫺/⫺ and wild-type mice resulted in an increase in the number of eosinophils retrieved in BALF (Fig. 3E). ⌬9-THC treatment did not affect the number of eosinophils in BALF induced by PR8.

BALF CD4⫹ and CD8⫹ T cell levels following PR8 challenge in CB1⫺/⫺/CB2⫺/⫺ and wild-type mice To evaluate the contribution of CD4⫹ and CD8⫹ T cells within the pool of BALF lymphocytes responding to PR8 challenge, the number of CD4⫹ T cells (Fig. 4A) and CD8⫹

Fig. 3. Recruitment of leukocytes to the pulmonary airways of mice challenged with PR8. The effects of ⌬9-THC on leukocyte recruitment to the pulmonary airways of CB1⫺/⫺/ CB2⫺/⫺ and wild-type mice challenged with PR8 were enumerated. Total leukocyte (A) counts and differential cell counts of neutrophils (B), lymphocytes (C), monocytes (D), and eosinophils (E) were assessed. Data represent the mean ⫾ SEM, presented as the number of cells per ml BALF. *, A significant difference between corn oil-treated mice infected with PR8 and respective controls instilled with saline; #, a significant effect elicited by ⌬9-THC-treated mice infected with PR8 when compared with corn oil-treated mice infected with PR8; ⫹, a significant difference between CB1⫺/⫺/CB2⫺/⫺ and wild-type mice within a respective treatment group.


789

CB receptor-deficient mice exhibit unique differences in epithelial and leukocytic chemokine and cytokine secretion in the pulmonary airways

Fig. 4. Enumeration of CD4⫹ and CD8⫹ T cells in BALF. The effects of ⌬9-THC on CD4⫹ T cell (A) and CD8⫹ T cell counts (B) recruited to the airways of CB1⫺/⫺/CB2⫺/⫺ and wild-type mice challenged with influenza. Data represent the mean ⫾ SEM presented as the number of CD4⫹ and CD8⫹ T cells. *, A significant difference between corn oil-treated mice infected with PR8 and respective controls instilled with saline; #, a significant effect elicited by ⌬9-THC-treated mice infected with PR8 when compared with corn oiltreated mice infected with PR8; ⴙ, a significant difference between CB1⫺/⫺/ CB2⫺/⫺ and wild-type mice within a respective treatment group. N.D., Not detected.

T cells (Fig. 4B) was enumerated by flow cytometry. Background levels of CD4⫹ T cells were greater in corn oil and ⌬9-THC-treated CB1⫺/⫺/CB2⫺/⫺ mice than in wild-type mice with the same treatments. There were no observed differences in the number of CD4⫹ T cells retrieved in BALF between saline-instilled controls and CB1⫺/⫺/CB2⫺/⫺ mice infected with PR8 alone. Alternatively, ⌬9-THCtreated CB1⫺/⫺/CB2⫺/⫺ mice infected with PR8 had significantly increased numbers of CD4⫹ T cells above background. The increase in BALF-associated CD4⫹ T cells in CB1⫺/⫺/CB2⫺/⫺ mice treated with ⌬9-THC and infected with PR8 was comparatively greater than the respective treatment group in wild-type mice. CD8⫹ T cells were not detected in BALF from saline-instilled controls in CB1⫺/⫺/ CB2⫺/⫺ and wild-type mice. However, there were marked increases in the number of BALF-associated CD8⫹ T cells for CB1⫺/⫺/CB2⫺/⫺ and wild-type mice challenged with PR8. In addition, there was a trend toward decreased numbers of BALF-associated CD8⫹ T cells in ⌬9-THC-treated wild-type mice infected with PR8. 790


One mode of cellular communication within a mixed population of leukocytes and between infected epithelium and leukocytes is through secretion of cytokines and chemokines. PR8infected wild-type and CB1⫺/⫺/CB2⫺/⫺ mice had significant increases in levels of the chemokine MCP-1 (Fig. 5A) and proinflammatory cytokines TNF-␣ (Fig. 5B), IL-6 (Fig. 5C), and IFN-␥ (Fig. 5D) measured in BALF. In addition, there was a modest enhancement of IL-10 concentrations (Fig. 5E) following PR8 challenge in CB1⫺/⫺/CB2⫺/⫺ and wild-type mice. There were no changes in IL-12p70 in BALF in the presence or absence of ⌬9-THC treatment or PR8 challenge (Fig. 5F) in CB1⫺/⫺/CB2⫺/⫺ or wild-type mice. ⌬9-THC-treated CB1⫺/⫺/ CB2⫺/⫺ mice infected with PR8 had an attenuation of IFN-␥ and IL-10 concentrations in BALF as compared with corn oil-treated mice infected with PR8. Conversely, the concentrations of chemokines and cytokines detected in BALF in PR8infected wild-type mice were unaffected by ⌬9-THC treatment. There were, however, comparative differences between detectable levels of TNF-␣ and IFN-␥ (P⫽0.061) in BALF of PR8infected wild-type and CB1⫺/⫺/CB2⫺/⫺ mice. The TH2 cytokines IL-2 (Fig. 6A), IL-4 (Fig. 6B), and IL-5 (Fig. 6C) were also quantified in BALF by CBA. Of particular interest, PR8- infected CB1⫺/⫺/CB2⫺/⫺ mice had marked increases in concentrations of IL-5, which were comparatively two-fold less than concentrations observed in the respective PR8-infected wild-type mice. The concentrations of IL-2 and IL-4 in BALF were unaffected by PR8 challenge in CB1⫺/⫺/ CB2⫺/⫺ mice. ⌬9-THC-treated CB1⫺/⫺/CB2⫺/⫺ mice infected with PR8 had modestly attenuated concentrations of IL-2 in BALF but had no effect on the concentrations of IL-4 or IL-5. PR8-infected wild-type mice had marked increases in IL-5 concentrations in BALF, mild increases in the concentrations of IL-2, and no detectable differences for IL-4 when compared with saline-instilled mice on 7 dpi. ⌬9-THC-treated wild-type mice infected with PR8 had a modest attenuation of IL-2 concentrations in BALF with no observed effect on IL-4 or IL-5 concentrations.

The absence of CB receptors CB1 and CB2 enhances the observed pulmonary histopathology As reported previously by our laboratory, infection of mice with PR8 induced significant pulmonary inflammation 7 dpi in wild-type mice (Fig. 7, A–D) [5]. Infection of the CB1⫺/⫺/ CB2⫺/⫺ mice with PR8 resulted in a similar inflammatory reaction (Fig. 7E). As we have reported, the inflammation consisted of primarily lymphocytes and neutrophils with fewer monocytes and plasma cells centered upon the bronchoalveolar junction, which extended out into the surrounding parenchyma. Treatment of PR8-infected wild-type mice with ⌬9-THC resulted in a significant decrease in inflammation 7 dpi compared with corn oil-treated mice infected with PR8 (Fig. 7D). These mice often had no to few inflammatory cells present; the remaining inflammation was comprised of primarily lymphohttp://www.jleukbio.org

Fig. 5. Secretion of inflammatory chemokines and cytokines into the pulmonary airways. The effects of ⌬9-THC on the release of soluble chemical mediators from leukocytes and epithelium in response to PR8 challenge of the pulmonary airways of CB1⫺/⫺/CB2⫺/⫺ and wild-type mice. BALF concentrations of the inflammatory chemokine MCP-1 (A) and cytokines TNF-␣ (B), IL-6 (C), IFN-␥ (D), IL-10 (E), and IL-12p70 (F) were determined by CBA analysis. Data represent the mean ⫾ SEM, presented as the concentration of chemokine or cytokine in pg/ml. *, A significant difference between corn oil-treated mice infected with PR8 and respective controls instilled with saline; #, a significant effect elicited by ⌬9-THC-treated mice infected with PR8 when compared with corn oil-treated mice infected with PR8; ⫹, a significant difference between CB1⫺/⫺/CB2⫺/⫺ and wild-type mice within a respective treatment group.

cytes and alveolar monocytes. In contrast, treatment of PR8infected CB1⫺/⫺/CB2⫺/⫺ mice with ⌬9-THC resulted in a vigorous, inflammatory and cellular reaction in the mice (Fig. 7F), where there were large numbers of mature lymphocytes around the conducting airways and extended into the surrounding alveolar parenchyma, where they were admixed with fewer monocytes and neutrophils. The bronchiolar epithelium was moderately hypertrophied. Treatment of wild-type and CB1⫺/⫺/ CB2⫺/⫺ mice with corn oil vehicle or ⌬9-THC alone did not

result in significant histologic changes in any of the lung sections examined.

⌬9-THC affects the magnitude of the inflammatory response to PR8 in wild-type and CB1- and CB2-deficient mice The magnitude and severity of inflammation observed in tissue sections isolated from the left lung lobe were scored independently (0 –3, where 0⫽no inflammation; 1⫽mild, inflammatory

Fig. 6. TH2 cytokines secreted into the pulmonary airways. The effects of ⌬9-THC on the concentrations of the TH2 cytokines IL-2 (A), IL-4 (B), and IL-5 (C) in BALF from CB1⫺/⫺/CB2⫺/⫺ and wild-type mice challenged with PR8. Data represent the mean ⫾ SEM, presented as the concentration of TH2 cytokine in pg/ml. *, A significant difference between corn oil-treated mice infected with PR8 and respective controls instilled with saline; #, a significant effect elicited by ⌬9-THC-treated mice infected with PR8 when compared with corn oil-treated mice infected with PR8; ⴙ, a significant difference between CB1⫺/⫺/CB2⫺/⫺ and wild-type mice within a respective treatment group.


791

Fig. 7. Inflammatory response to PR8 in the proximal section of the left lung lobe from wild-type and CB1⫺/ ⫺/⫺ ⫺/CB2 mice. Light photomicrographs of the hilar region of the left lung lobe of wild-type and CB1⫺/⫺/ CB2⫺/⫺ mice sectioned at generation 5 along the main axial airway (maa). (A) Corn oil-treated wild-type mouse instilled with saline (SAL) with no evidence of an inflammatory cell infiltrate in the peribronchiolar (maa)/ perivascular (a) submucosa (sm) and no evidence of inflammation in the alveolar parenchyma (p). (B) ⌬9THC-treated wild-type mouse instilled with saline with marked inflammatory cell infiltrate of the submucosa and diffuse inflammation within the alveolar airspace. (C) Corn oil-treated wild-type mouse instilled with PR8 with marked inflammatory cell infiltrate (arrow) in the peribronchiolar/perivascular submucosa and alveolar airspace.(D) ⌬9-THC-treated wild-type mouse instilled with PR8 with modest to no inflammatory cell infiltrate of the submucosa and alveolar airspace. (E) Corn oiltreated CB1⫺/⫺/CB2⫺/⫺ mouse instilled with PR8 with marked inflammatory cell infiltrate in the peribronchiolar/perivascular submucosa and alveolar airspace. (F) ⌬9-THC-treated CB1⫺/⫺/CB2⫺/⫺ mouse instilled with PR8 with severe inflammation within the submucosa and alveolar airspace. Original bars ⫽ 50 ␮m

cell infiltrate of the perivascular/peribronchiolar compartment; 2, moderate, inflammatory cell infiltrate of the perivascular/ peribronchiolar space with modest extension into the alveolar parenchyma; and 3⫽severe, inflammatory cell infiltrate of the perivascular/peribronchiolar space with large, inflammatory foci found in the alveolar parenchyma) and compared between treatment groups and wild-type and CB1⫺/⫺/CB2⫺/⫺ mice at 7 dpi (Fig. 8). There was no inflammation observed in the lungs of wild-type mice intranasally instilled with saline. However, two out of eight CB1⫺/⫺/CB2⫺/⫺ mice instilled with saline had a modest inflammation present. Corn oil-treated mice infected with PR8 had marked inflammation scores in wild-type and CB1⫺/⫺/CB2⫺/⫺ mice as compared with uninfected control mice. ⌬9-THC-treated wild-type mice challenged with PR8 had a marked suppression of the inflammatory response in the conducting airways. In contrast, ⌬9-THC-treated CB1⫺/⫺/ CB2⫺/⫺ mice infected with PR8 had a trend toward increased (P⫽0.079) inflammation of the pulmonary airways.

Effects of CB1 and CB2 deficiency on the numeric cell densities of apoptotic cells and metaplastic goblet cells Bronchiolar epithelial cell apoptosis in influenza-infected mice is directed by cell-mediated immune responses to virally infected cells [26]. The numeric cell density for the apoptotic cell marker CAS-3 (Fig. 9) was measured in the epithelium lining generation 5 of the main axial airway. CAS-3 numeric cell densities were not distinguishably different between treatment groups in wild-type mice. In PR8-infected CB1⫺/⫺/ CB2⫺/⫺ mice, CAS-3-labeled cells were modestly increased 792


Fig. 8. Combined histopathology-based inflammation scores for generations 5 and 11 of the main axial airway of the left lung lobe. Effects of ⌬9-THC on the inflammatory response gathered within the subepithelial interstitium and alveolar parenchyma following influenza challenge of the pulmonary airways in CB1⫺/⫺/CB2⫺/⫺ and wild-type mice. Scores were tabulated and averaged for the two sections taken from generations 5 and 11 of the left lung lobe. Scores: 0 ⫽ No inflammation; 1⫽ mild, inflammatory cell infiltrate of the perivascular/ peribronchiolar compartment; 2 ⫽ moderate, inflammatory cell infiltrate of the perivascular/peribronchiolar space with modest extension into the alveolar parenchyma; and 3 ⫽ severe, inflammatory cell infiltrate of the perivascular/ peribronchiolar space with a greater magnitude of inflammatory foci found in the alveolar parenchyma. Data represent the mean ⫾ SEM presented as the inflammation score. *, A significant difference between corn oil-treated mice infected with PR8 and respective controls instilled with saline; #, a significant effect elicited by ⌬9-THC-treated mice infected with PR8 when compared with corn oil-treated mice infected with PR8; ⴙ, a significant difference between CB1⫺/⫺/CB2⫺/⫺ and wild-type mice within a respective treatment group.


⫺/⫺

⫺/CB2

mice, in the presence or absence of PR8 infection, had significant increases in the numeric cell density for AB/ PAS-positive epithelial cells as compared with wild-type mice.

DISCUSSION

Fig. 9. Quantification of bronchiolar epithelial cell apoptosis at generation 5 by numeric cell density of CAS-3-positive epithelial cells. The effects of ⌬9-THC on apoptotic cell death in response to influenza challenge. Numeric cell densities were determined by enumerating CAS-3-positive nuclei per length of basal lamina. Data represent the mean ⫾ SEM, presented as the number of CAS-3-positive nuclei per length of basal lamina. *, A significant difference between corn oil-treated mice infected with PR8 and respective controls instilled with saline.

above saline-instilled controls but were not different than cell densities enumerated for CB1⫺/⫺/CB2⫺/⫺ mice infected with PR8 and treated with ⌬9-THC. During the recovery from viral infection, a metaplastic change occurs in the epithelium lining the bronchi that includes increased numbers of mucus-producing goblet cells [27]. Consistent with our previous finding [5], there was no difference in the numeric cell density of AB/PASpositive mucous cells (Fig. 10) quantified at 7 dpi in wild-type mice with any treatment. Interestingly, ⌬9-THC-treated CB1⫺/

⌬9-THC administration has been reported to decrease host resistance to bacterial and viral pathogens in mouse models [2–5]. However, the role of the CB receptors CB1 and CB2 in altered immune competence to pathogen challenge remains poorly understood. The CB receptor CB1 antagonist SR141716A and CB2 receptor antagonist SR144528 have been used to study CB1/CB2-specific effects elicited by CB treatment, including efforts to uncover the CB1/CB2 receptors involvement in ⌬9-THC suppression of cytokine production elicited by pathogen challenge [28], but the specificity of these pharmacologic agents and their putative off-target effects are always of concern. The objective of the present study was to use CB1/CB2 receptor null mice to rigorously evaluate previously characterized immunologic and pathologic responses to PR8 infection in the presence and absence of ⌬9-THC treatment [5]. Results from this study indicated that ⌬9-THC treatment of wild-type and CB1⫺/⫺/CB2⫺/⫺ mice affected epithelial and immune cell responses to PR8 that were dependent and independent of the CB1/CB2 receptors as summarized in Table 1. Perhaps even more important was the observation that CB1/ CB2 null mice exhibited a significantly enhanced response to PR8 infection regardless of CB treatment, strongly suggesting an endogenous role for CB1 and/or CB2 in immune regulation. TABLE 1. Summary of the Effects of PR8 Infection on Airway –/– Epithelial Cells and Immune Cells in Wild-Type and CB–/– 1 /CB2 9 Mice Treated with Corn Oil or ⌬ -THC –/– CB–/– 1 /CB2

Wild-type Effect of corn oil on: Epithelium MCM Lesion Inflammatory cells BALF leukocytes BALF cytokines Inflammation score

Saline

PR8

Saline

PR8

— —

— ⫹

— —

— ⫹

— — —

⫹⫹ ⫹⫹ ⫹⫹

— — —

–/– CB–/– 1 /CB2

Wild-type Effect of ⌬9-THC on:

Fig. 10. Quantification of bronchiolar epithelial MCM at generation 5 by numeric cell density of mucosubstances. The effects of ⌬9-THC on the development of MCM in the epithelium lining the main axial airway following influenza challenge and subsequent inflammatory cell responses. Numeric cell densities were determined by enumerating AB/PAS-positive secretory vesicles in epithelial cells per length of basal lamina from stained sections. Data represent the mean ⫾ SEM, presented as the number of AB/PAS-positive nuclei per length of basal lamina. *, A significant difference between corn oil-treated mice infected with PR8 and respective controls instilled with saline; ⴙ, a significant difference between CB1⫺/⫺/CB2⫺/⫺ and wild-type mice within a respective treatment group.

Epithelium MCM Lesion Inflammatory cells BALF leukocytes BALF cytokines Inflammation score

⫹⫹⫹ ⫹⫹ ⫹⫹

Saline

PR8

Saline

PR8

— —

— ⫹

⫹⫹⫹ —

⫹ ⫹⫹

— — —

⫹ ⫹ ⫹

— — —

⫹ ⫹ ⫹⫹⫹

Summary of the effects of corn oil and ⌬9-THC on BALF and histochemistry measurements taken for epithelial and inflammatory cells in wild-type and –/– CB–/– mice. Dashed line (—) denotes no effect. A single plus (⫹) 1 /CB2 denotes a mild effect, a double plus (⫹⫹) indicates a modest effect, and a triple plus (⫹⫹⫹) signifies a marked response.


793

This finding provides a novel extension to our current understanding of host immunity and the relationship between a normal, “homeostatic” immune response regulated by CB receptor-linked signaling pathways and hyper-responsive, “unchecked” immunity compromised by the absence of the CB1/ CB2 receptors. In the current study, ⌬9-THC was administered by oral gavage. As we have reported previously, the oral route of administration was selected as a result of the fact that ⌬9-THC is a highly lipophilic molecule requiring a nonaqueous diluent for drug delivery, which has the potential for inducing irritation and damage to the airways. Conversely, oral administration of ⌬9-THC in corn oil is well tolerated but poorly absorbed from the gastrointestinal tract. Oral administration of 75 mg/kg ⌬9-THC for 5 consecutive days resulted in a serum concentration of 66.2 ng/ml of the parent compounds, 446.5 ng/ml of the 9-carboxyl metabolite, and 10.5 ng/ml of the 11-hydroxyl metabolite, 4 h after the last ⌬9-THC dose [5]. The levels of ⌬9-THC observed systemically correlate well with a previous report by Azorlosa and co-workers [29], in which peak human plasma levels ranged from 57 to 268 ng/ml. The rationale for the dosing paradigm of PR8 and ⌬9-THC was to investigate the putative effects of ⌬9-THC treatment on the immune response to PR8 during the early stages of the primary infection. To assess whether the CB receptors CB1 and CB2 are critical to the cell-mediated immune response that clears influenza from the lungs, we compared the levels of H1 mRNA in lungs from wild-type and CB1⫺/⫺/CB2⫺/⫺ mice. In the current study, we observed H1 mRNA levels in the lungs of ⌬9-THC-treated CB1⫺/⫺/CB2⫺/⫺ and wild-type mice challenged with PR8 that were greater than H1 mRNA levels in CB1⫺/⫺/CB2⫺/⫺ and wild-type mice challenged with PR8 alone. Interestingly, H1 mRNA levels were reduced in the lungs of CB1⫺/⫺/CB2⫺/⫺ mice by two orders of magnitude when compared with H1 mRNA levels observed for wild-type mice. The measurement of H1 mRNA serves as a marker of transcriptional activation of the influenza virus genome that could result from increases in viral copy number or increases in transcriptional activation. Whether ⌬9-THC is exerting an effect on copy number or transcriptional activation or both is not presently known. It is tempting to speculate, however, that H1 mRNA levels are related to copy number, as there are marked differences in the magnitude of the inflammatory response to PR8 observed by histopathology. The profound reduction in H1 mRNA levels also suggests that the kinetics of the cell-mediated immune response to PR8 is markedly enhanced in CB1⫺/⫺/CB2⫺/⫺ mice. The suggested hyper-responsive immunity toward PR8 infection brought about by the absence of functional CB receptors CB1 and CB2 is supported by our findings, in the current study, of increased background levels of CD4⫹ T cells in CB1⫺/⫺/CB2⫺/⫺ mice in combination with a more vigorous and extensive inflammation of the alveolar parenchyma. Recently, Karsak and co-workers [24] reported a more vigorous, allergic response to the contact allergen 2,4-dinitroflourobenzene in CB1⫺/⫺/CB2⫺/⫺ mice when compared with wild-type mice. The current study and that reported by Karsak and co-workers [24] suggest that the CB receptors CB1 and CB2 may play a crucial role in maintaining immune homeostasis and controlling the magnitude of the immune response through 794


negative regulation. It is tempting to speculate that endoCBs may naturally provide a mechanism to temper the immune response. The absence of CB1 and CB2 receptors appears to potentially result in an enhanced immune response under certain conditions. We have previously observed decreased recruitment of monocytes and lymphocytes to the airways in ⌬9-THC-treated mice challenged with PR8 [5]. This observation was consistent with the suppressive effects of CBs on leukocyte chemotaxis reported by others [30 –33]. To clarify whether mice lacking functional CB1/ CB2 receptors exhibit patterns of leukocyte recruitment to the airways in response to viral challenge that are modulated by ⌬9-THC in a CB1/CB2 receptor-independent manner, total and differential cell counts from BALF were assessed. In the current study, the magnitude of total leukocyte recruitment to the pulmonary airways and the cellular subsets present were similar between CB1⫺/⫺/CB2⫺/⫺ and wild-type mice challenged with PR8, suggesting that mice lacking functional CB1/CB2 receptors remained capable of mounting an aggressive immune response toward PR8. Differential cell counts provided evidence that the immune response to PR8 by CB1⫺/⫺/CB2⫺/⫺ mice was more robust than the immune response mounted against PR8 in wildtype mice. More specifically, in the BALF, monocytes and neutrophils were decreased in ⌬9-THC-treated CB1⫺/⫺/CB2⫺/⫺ mice infected with PR8. The decreased recruitment of monocytes and neutrophils to the airways might point to a CB receptor-dependent effect on the chemotaxis of these leukocytes. Contrary to findings with monocytes and neutrophils, there were remarkable similarities in the total number of lymphocytes in BALF among all treatment groups for CB1⫺/⫺/CB2⫺/⫺ and wild-type mice. Upon further examination of the T cell subsets retrieved in BALF, there was evidence that the T cell response to PR8 challenge with ⌬9-THC treatment involved predominantly CD4⫹ T cells in CB1⫺/⫺/CB2⫺/⫺ mice. As mentioned previously, increased numbers of CD4⫹ T cells might suggest hyper-responsive immunity in CB1⫺/⫺/CB2⫺/⫺ mice. However, the balance of the CD4⫹ TH subset (TH1 cell-mediated immunity vs. TH2 humoral immunity) remains unclear. We have previously shown that PR8 challenge of the pulmonary airways leads to marked increases in total protein and chemokine/cytokines MCP-1, IL-6, TNF-␣, and IFN-␥ in BALF [5]. ⌬9-THC and other CBs have been reported to modulate the expression and/or secretion of these chemokines and cytokines [20, 33–36]. In the current study, we observed increased total protein in the BALF of wild-type and CB1⫺/⫺/CB2⫺/⫺ mice challenged with PR8. Comparatively, a marked difference existed between ⌬9-THC-treated CB1⫺/⫺/CB2⫺/⫺ and wild-type mice infected with PR8, suggesting an increased severity in vascular leakiness or inflammatory cell secretions in CB1⫺/⫺/CB2⫺/⫺ mice. Consistent with increased amounts of total protein in BALF, inflammation scores assessed by histopathology accurately reflected changes observed in vascular leakiness. In addition to total protein, we evaluated chemokines/cytokines in BALF. Although there were trends toward increases in BALF chemokines and cytokines (e.g., MCP-1, TNF-␣, and IFN-␥) from CB1⫺/⫺/CB2⫺/⫺ mice infected with PR8, in the absence of ⌬9-THC, the collective findings did not provide evidence for CB1 and/or CB2 receptordependent or -independent regulation of the secretion of these cytokines. Chemokine and cytokine levels should be interpreted http://www.jleukbio.org

carefully, as they represent a concentration in the airways at a given moment in time with respect to an ever-changing, dynamic inflammatory process. It is difficult to speculate in the current study as to whether differences observed between treatment and control groups represent trends toward increased or decreased expression or secretion of chemokines or cytokines. Moreover, the cytokines measured in BALF are pleiotropic and redundant, as they may be derived from multiple immune cell types. Not surprisingly, cytokines and chemokines are regulated in their own unique manner with their own distinct kinetics, and the molecular mechanism by which certain cytokines and chemokines are further modulated by CBs is only partially understood. It has been suggested that ⌬9-THC treatment modulates cytokine production, resulting in decreased TH1 cell-mediated immunity and increased TH2 humoral immunity [37, 38]. As CB1⫺/⫺/CB2⫺/⫺ mice had a marked, lymphocytic response consisting of greater numbers of CD4⫹ T cells, and ⌬9-THC treatment of CB1⫺/⫺/CB2⫺/⫺ mice challenged with PR8 resulted in decreased concentrations of IFN-␥, the concentrations of the TH2 cytokines IL-2, IL-4, and IL-5 were also evaluated in BALF. Only the cytokine IL-5 was influenced by PR8 infection in wild-type mice in the presence or absence of ⌬9-THC. IL-5 concentrations were reduced in PR8infected CB1⫺/⫺/CB2⫺/⫺ mice in the presence or absence of ⌬9-THC, and ⌬9-THC treatment of CB1⫺/⫺/CB2⫺/⫺ or wild-type mice infected with PR8 exhibited no effect on the levels of IL-5 when compared with CB1⫺/⫺/CB2⫺/⫺ or wild-type mice infected with PR8 alone. Our results suggest that the increased numbers of CD4⫹ T cells observed in CB1⫺/⫺/CB2⫺/⫺ mice were not actively secreting more TH2-type cytokines than wild-type mice and that there did not appear to be an imbalance between TH1-type and TH2-type cytokines secreted as a result of ⌬9-THC treatment in response to PR8. To evaluate the extent of the inflammatory response to PR8 in CB1⫺/⫺/CB2⫺/⫺ mice as compared with wild-type controls, lung sections were taken at the levels of the 5th (proximal) and 11th (distal) airway generations, bifurcating away from the main axial airway of the left lung lobe and scored for the magnitude and severity of inflammation. Within saline-instilled wild-type mice, there was no inflammation observed. Conversely, in two out of eight CB1⫺/⫺/CB2⫺/⫺ mice instilled with saline, in the presence or absence of ⌬9-THC treatment, there was modest inflammation. The CB1⫺/⫺/CB2⫺/⫺ mice were subjected to an extensive battery of serological screens prior to their use and were found to be negative for all pathogens tested. In a separate, unpublished study, two untreated CB1⫺/⫺/CB2⫺/⫺ mice were necropsied and examined for microscopic evidence of lesions in the brain, eye, conjunctiva, myocardium, lung, stomach, small intestine, large intestine, liver, spleen, testicle, and nasal cross sections. Within the lungs, there was a mild interstitial, inflammatory infiltrate composed of lymphocytes and plasma cells. In addition, there were multifocal regions of atelectasis, which is the incomplete expansion of the lungs or the collapse of lung parenchyma, producing an area of relatively airless pulmonary parenchyma. Atelectatic parenchyma is prone to developing superimposed infections, which may partially account for the more vigorous, inflammatory responses observed within CB1⫺/⫺/CB2⫺/⫺ mice. Regardless, inflammation scores were similar for lung sections from CB1⫺/⫺/CB2⫺/⫺ and wild-type mice challenged with PR8, in the absence of ⌬9-THC treatment, and the histopathology for

PR8-infected CB1⫺/⫺/CB2⫺/⫺ and wild-type mice had a similar pattern of inflammatory cells infiltrating the lung. In ⌬9-THCtreated wild-type mice challenged with PR8, there was a marked reduction in the inflammation score, suggesting that ⌬9-THC altered the kinetics of leukocyte chemotaxis to the airways in response to PR8. This is supported by the marked reduction or absence of inflammatory cells in the submucosa and alveolar parenchyma. In contrast, inflammation scores for lung sections from ⌬9-THC-treated CB1⫺/⫺/CB2⫺/⫺ mice challenged with PR8 were modestly enhanced with respect to scores assessed for lung sections from CB1⫺/⫺/CB2⫺/⫺ mice challenged with PR8 alone. Indeed, CB1⫺/⫺/CB2⫺/⫺ mice treated with ⌬9-THC and infected with PR8 represented some of the most severely affected mice in the study and exhibited more vigorous, inflammatory responses than even observed in CB1⫺/⫺/CB2⫺/⫺ mice challenged with PR8 alone. We have previously demonstrated that epithelial changes in the pulmonary airways of mice challenged with PR8 included increased evidence for apoptosis and MCM following PR8 challenge, which was suppressed by ⌬9-THC [5]. Cell death pathways initiated by cytotoxic T cells or NK cells or by the virus itself influence CAS-3 activation [39, 40]. In the comparison of wildtype and CB1⫺/⫺/CB2⫺/⫺ mice, PR8 challenge led to increased CAS-3 staining; however, there were no effects observed with ⌬9-THC administration. In a similar manner, MCM was quantified by counting AB/PAS-positive epithelial cells lining the main axial airway at G5 of the left lung lobe. We have previously shown that goblet cells producing mucosubstances in the airway epithelium increase substantially by 10 dpi in wild-type mice. In the current study, we found that CB1⫺/⫺/CB2⫺/⫺ mice had significantly greater numbers of goblet cells lining the airways at G5 when treated with ⌬9-THC alone at 7 dpi. This finding suggests that ⌬9-THC is interacting directly with a signaling pathway tied with increased mucin production in CB1⫺/⫺/CB2⫺/⫺ mice or that exposure of CB1⫺/⫺/CB2⫺/⫺ mice to ⌬9-THC is handled by the host as an allergen leading to an alternative, hyper-responsive CD4⫹ TH2 T cell response that is capable of eliciting MCM. In conclusion, ⌬9-THC-treated CB1⫺/⫺/CB2⫺/⫺ mice infected with PR8 exhibited more robust immune responses than wild-type mice receiving the respective treatment. Moreover, CB1⫺/⫺/CB2⫺/⫺ mice infected with PR8 had markedly decreased levels of viral mRNA as compared with wild-type mice. Furthermore, ⌬9-THC treatment increased viral load when compared with mice infected with PR8 alone, irrespective of the presence or absence of CB1 and CB2 receptors. Therefore, the current study demonstrated that ⌬9-THC influenced the immune response to influenza infection in a manner that was dependent and independent of the CB1/CB2 receptors. Future studies will examine the effect of ⌬9-THC on viral infection of lung parenchymal cells and establish the kinetics of lymphocyte, monocyte, and neutrophil chemotaxis to the lungs of PR8-infected CB1⫺/⫺/CB2⫺/⫺ mice.

ACKNOWLEDGMENTS This work was supported by the Michigan State University Foundation Strategic Partnership Grant, National Institutes of Health Grant DA07908, and the National Institutes of Environmental Immune responses of CB1/CB2 receptor-null mice to PR8

795

Health Sciences Training Grant T32 ES07255. We acknowledge Dr. Allen Harmsen for providing the influenza virus used in this study and Robert Crawford and Lori Bramble for providing technical assistance with flow cytometry. In addition, the authors thank Kim Hambleton for her assistance in the submission of this manuscript.

REFERENCES 1. Klein, T. W., Cabral, G. A. (2006) Cannaboinoid-induced immune suppression and modulation of antigen-presenting cells. J. Neuroimmune Pharmacol. 1, 50 – 64. 2. Morahan, P. S., Klykken, P. C., Smith, S. H., Harris, L. S., Munson, A. E. (1979) Effects of cannabinoids on host resistance to Listeria monocytogenes and herpes simplex virus. Infect. Immun. 23, 670 – 674. 3. Fischer-Stenger, K., Updegrove, A. W., Cabral, G. A. (1992) ⌬ 9-Tetrahydrocannabinol decreases cytotoxic T lymphocyte activity to herpes simplex virus type 1-infected cells. Proc. Soc. Exp. Biol. Med. 200, 422– 430. 4. Klein, T. W., Newton, C., Friedman, H. (1994) Resistance to Legionella pneumophila suppressed by the marijuana component, tetrahydrocannabinol. J. Infect. Dis. 169, 1177–1179. 5. Buchweitz, J. P., Karmaus, P. W., Harkema, J. R., Williams, K. J., Kaminski, N. E. (2007) Modulation of airway responses to influenza A/PR/8/34 by {⌬}9-tetrahydrocannabinol in C57BL/6 mice. J. Pharmacol. Exp. Ther. 323, 675– 683. 6. Lu, T., Newton, C., Perkins, I., Friedman, H., Klein, T. W. (2006) Role of cannabinoid receptors in ⌬-9-tetrahydrocannabinol suppression of IL12p40 in mouse bone marrow-derived dendritic cells infected with Legionella pneumophila. Eur. J. Pharmacol. 532, 170 –177. 7. Klein, T. W., Newton, C. A., Nakachi, N., Friedman, H. (2000) ⌬ 9-Tetrahydrocannabinol treatment suppresses immunity and early IFN-␥, IL12, and IL-12 receptor ␤ 2 responses to Legionella pneumophila infection. J. Immunol. 164, 6461– 6466. 8. Jan, T. R., Farraj, A. K., Harkema, J. R., Kaminski, N. E. (2003) Attenuation of the ovalbumin-induced allergic airway response by cannabinoid treatment in A/J mice. Toxicol. Appl. Pharmacol. 188, 24 –35. 9. Kawakami, Y., Klein, T. W., Newton, C., Djeu, J. Y., Specter, S., Friedman, H. (1988) Suppression by ⌬-9-tetrahydrocannabinol of interleukin 2-induced lymphocyte proliferation and lymphokine-activated killer cell activity. Int. J. Immunopharmacol. 10, 485– 488. 10. Ouyang, Y., Hwang, S. G., Han, S. H., Kaminski, N. E. (1998) Suppression of interleukin-2 by the putative endogenous cannabinoid 2-arachidonylglycerol is mediated through down-regulation of the nuclear factor of activated T cells. Mol. Pharmacol. 53, 676 – 683. 11. Yea, S. S., Yang, K. H., Kaminski, N. E. (2000) Role of nuclear factor of activated T-cells and activator protein-1 in the inhibition of interleukin-2 gene transcription by cannabinol in EL4 T-cells. J. Pharmacol. Exp. Ther. 292, 597– 605. 12. Ronni, T., Sareneva, T., Pirhonen, J., Julkunen, I. (1995) Activation of IFN-␣, IFN-␥, MxA, and IFN regulatory factor 1 genes in influenza A virus-infected human peripheral blood mononuclear cells. J. Immunol. 154, 2764 –2774. 13. Bot, A., Reichlin, A., Isobe, H., Bot, S., Schulman, J., Yokoyama, W. M., Bona, C. A. (1996) Cellular mechanisms involved in protection and recovery from influenza virus infection in immunodeficient mice. J. Virol. 70, 5668 –5672. 14. Pettit, D. A., Anders, D. L., Harrison, M. P., Cabral, G. A. (1996) Cannabinoid receptor expression in immune cells. Adv. Exp. Med. Biol. 402, 119 –129. 15. Schatz, A. R., Lee, M., Condie, R. B., Pulaski, J. T., Kaminski, N. E. (1997) Cannabinoid receptors CB1 and CB2: a characterization of expression and adenylate cyclase modulation within the immune system. Toxicol. Appl. Pharmacol. 142, 278 –287. 16. McCoy, K. L., Matveyeva, M., Carlisle, S. J., Cabral, G. A. (1999) Cannabinoid inhibition of the processing of intact lysozyme by macrophages: evidence for CB2 receptor participation. J. Pharmacol. Exp. Ther. 289, 1620 –1625. 17. Buckley, N. E., McCoy, K. L., Mezey, E., Bonner, T., Zimmer, A., Felder, C. C., Glass, M., Zimmer, A. (2000) Immunomodulation by cannabinoids is absent in mice deficient for the cannabinoid CB(2) receptor. Eur. J. Pharmacol. 396, 141–149. 18. Zhu, L. X., Sharma, S., Stolina, M., Gardner, B., Roth, M. D., Tashkin, D. P., Dubinett, S. M. (2000) ⌬-9-Tetrahydrocannabinol inhibits antitumor immunity by a CB2 receptor-mediated, cytokine-dependent pathway. J. Immunol. 165, 373–380.

796


19. Kaplan, B. L., Rockwell, C. E., Kaminski, N. E. (2003) Evidence for cannabinoid receptor-dependent and -independent mechanisms of action in leukocytes. J. Pharmacol. Exp. Ther. 306, 1077–1085. 20. Kaplan, B. L., Ouyang, Y., Rockwell, C. E., Rao, G. K., Kaminski, N. E. (2005) 2-Arachidonoyl-glycerol suppresses interferon-␥ production in phorbol ester/ionomycin-activated mouse splenocytes independent of CB1 or CB2. J. Leukoc. Biol. 77, 966 –974. 21. Costa, B., Colleoni, M., Conti, S., Parolaro, D., Franke, C., Trovato, A. E., Giagnoni, G. (2004) Oral anti-inflammatory activity of cannabidiol, a non-psychoactive constituent of cannabis, in acute carrageenan-induced inflammation in the rat paw. Naunyn Schmiedebergs Arch. Pharmacol. 369, 294 –299. 22. Watzl, B., Scuderi, P., Watson, R. R. (1991) Influence of marijuana components (THC and CBD) on human mononuclear cell cytokine secretion in vitro. Adv. Exp. Med. Biol. 288, 63–70. 23. Srivastava, M. D., Srivastava, B. I., Brouhard, B. (1998) ⌬9 Tetrahydrocannabinol and cannabidiol alter cytokine production by human immune cells. Immunopharmacology 40, 179 –185. 24. Karsak, M., Gaffal, E., Date, R., Wang-Eckhardt, L., Rehnelt, J., Petrosino, S., Starowicz, K., Steuder, R., Schlicker, E., Cravatt, B., Mechoulam, R., Buettner, R., Werner, S., Di Marzo, V., Tuting, T., Zimmer, A. (2007) Attenuation of allergic contact dermatitis through the endocannabinoid system. Science 316, 1494 –1497. 25. Plopper, C. G., Mariassy, A. T., Lollini, L. O. (1983) Structure as revealed by airway dissection. A comparison of mammalian lungs. Am. Rev. Respir. Dis. 128, S4 –S7. 26. Topham, D. J., Tripp, R. A., Doherty, P. C. (1997) CD8⫹ T cells clear influenza virus by perforin or Fas-dependent processes. J. Immunol. 159, 5197–5200. 27. Buchweitz, J. P., Harkema, J. R., Kaminski, N. E. (2007) Time-dependent airway epithelial and inflammatory cell responses induced by influenza virus A/PR/8/34 in C57BL/6 mice. Toxicol. Pathol. 35, 424 – 435. 28. Newton, C. A., Lu, T., Nazian, S. J., Perkins, I., Friedman, H., Klein, T. W. (2004) The THC-induced suppression of Th1 polarization in response to Legionella pneumophila infection is not mediated by increases in corticosterone and PGE2. J. Leukoc. Biol. 76, 854 – 861. 29. Azorlosa, J. L., Heishman, S. J., Stitzer, M. L., Mahaffey, J. M. (1992) Marijuana smoking: effect of varying ⌬ 9-tetrahydrocannabinol content and number of puffs. J. Pharmacol. Exp. Ther. 261, 114 –122. 30. Coopman, K., Smith, L. D., Wright, K. L., Ward, S. G. (2007) Temporal variation in CB2R levels following T lymphocyte activation: evidence that cannabinoids modulate CXCL12-induced chemotaxis. Int. Immunopharmacol. 7, 360 –371. 31. Ghosh, S., Preet, A., Groopman, J. E., Ganju, R. K. (2006) Cannabinoid receptor CB2 modulates the CXCL12/CXCR4-mediated chemotaxis of T lymphocytes. Mol. Immunol. 43, 2169 –2179. 32. Nilsson, O., Fowler, C. J., Jacobsson, S. O. (2006) The cannabinoid agonist WIN 55,212-2 inhibits TNF-␣-induced neutrophil transmigration across ECV304 cells. Eur. J. Pharmacol. 547, 165–173. 33. Smith, S. R., Denhardt, G., Terminelli, C. (2001) The anti-inflammatory activities of cannabinoid receptor ligands in mouse peritonitis models. Eur. J. Pharmacol. 432, 107–119. 34. Lavon, I., Sheinin, T., Meilin, S., Biton, E., Weksler, A., Efroni, G., Bar-Joseph, A., Fink, G., Avraham, A. (2003) A novel synthetic cannabinoid derivative inhibits inflammatory liver damage via negative cytokine regulation. Mol. Pharmacol. 64, 1334 –1341. 35. Molina-Holgado, F., Molina-Holgado, E., Guaza, C. (1998) The endogenous cannabinoid anandamide potentiates interleukin-6 production by astrocytes infected with Theiler’s murine encephalomyelitis virus by a receptor-mediated pathway. FEBS Lett. 433, 139 –142. 36. Sacerdote, P., Martucci, C., Vaccani, A., Bariselli, F., Panerai, A. E., Colombo, A., Parolaro, D., Massi, P. (2005) The nonpsychoactive component of marijuana cannabidiol modulates chemotaxis and IL-10 and IL-12 production of murine macrophages both in vivo and in vitro. J. Neuroimmunol. 159, 97–105. 37. Newton, C. A., Klein, T. W., Friedman, H. (1994) Secondary immunity to Legionella pneumophila and Th1 activity are suppressed by ⌬-9-tetrahydrocannabinol injection. Infect. Immun. 62, 4015– 4020. 38. Yuan, M., Kiertscher, S. M., Cheng, Q., Zoumalan, R., Tashkin, D. P., Roth, M. D. (2002) ⌬ 9-Tetrahydrocannabinol regulates Th1/Th2 cytokine balance in activated human T cells. J. Neuroimmunol. 133, 124 –131. 39. Takizawa, T., Tatematsu, C., Ohashi, K., Nakanishi, Y. (1999) Recruitment of apoptotic cysteine proteases (caspases) in influenza virus-induced cell death. Microbiol. Immunol. 43, 245–252. 40. Wurzer, W. J., Planz, O., Ehrhardt, C., Giner, M., Silberzahn, T., Pleschka, S., Ludwig, S. (2003) Caspase 3 activation is essential for efficient influenza virus propagation. EMBO J. 22, 2717–2728.
