Female C57BL/6 mice (6–8 weeks) were purchased from Jackson laboratories. All mice were housed under pathogen-free conditions in the Animal Resource Facility (ARF) at the University of South Carolina with a maximum of five animals per cage. All requirements for the experiments were performed under the policy approved by the Institutional Animal Care and Use Committee (IACUC). Mice were housed under a 12 h light/dark cycle at 18–23°C and 40–60% humidity.

All chemicals and reagents were purchased as follows: N-arachidonoyl ethanolamine or Anandamide (AEA) from Cayman Chemicals (Ann Arbor, Michigan, United States); Staphylococcus Enterotoxin B (SEB): from Toxin Technologies (Sarasota, FL); RPMI 1640 supported with L-glutamine: from Corning (New York, NY, United States); Fetal Bovine Serum, Penicillin and Streptomycin: from Invitrogen Life Technologies (CA, United States). True Nuclear Transcription Factor Buffer Set was purchased from Biolegend (San Diego, CA) while Fc Block reagent was purchased from BD Pharmingen (San Diego, CA). EasySep mouse MDSCs (CD11bGr1) selection kit was purchased from STEMCELL technologies (Seattle, WA) for purification of MDSCs. miScript SYBR green PCR kit, miScript primer assays kit, RNA easy, and miRNA easy kit were purchased from Qiagen (Valencia, CA). Taq DNA polymerase kit was purchased from Invitrogen Life Technologies (Carlsbad, CA).

ARDS was induced in mice as described previously (Rieder et al., 2011). Mice were randomly divided into either vehicle or treatment groups before exposure to SEB. Mice were exposed to SEB intra-nasally (I.N.) with a single dose at a concentration of 50 µg/mouse in 25 µl of Phosphate Buffer Saline (PBS) as previously described (Rieder et al., 2011) on day 0. On day-1, AEA or VEH was given into these mice by the Intra Peritoneal (I.P.) route at a dose of 40 mg/kg body weight. AEA dissolved in ethanol (50 mg/ml) was diluted further in PBS. Each mouse received 0.1 ml consisting of 84 μl of PBS and 16 μl of ethanol containing AEA. The vehicle controls received 0.1 ml consisting of 84 μl of PBS and 16 μl of ethanol without AEA. The dose of AEA was based on our previous studies demonstrating that 40 mg/kg body weight of AEA attenuated T cell-mediated delayed-type hypersensitivity response (Jackson et al., 2014b). The treatment with AEA was repeated on day 0 (SEB exposure day), and day 1. Mice were euthanized on day 2 (48 h after SEB exposure) for various studies.

To evaluate the effect of AEA to attenuate the effects of SEB, clinical parameters of the lungs were measured using whole-body plethysmography (Buxco, Troy, NY, United States), as described (Elliott et al., 2016). A single mouse from each group of Naïve, SEB + VEH, and SEB + AEA was first restrained in a plethysmographic tube and was allowed to acclimatize as previously described (Elliott et al., 2016). The lung function was calculated for clinical parameters such as Specific Airway Resistance (sRAW), Specific Airway Conductance (sGAW) and Minute per Volume (MV).

Lung tissue was excised from each mouse and directly fixed in 10% formalin without inflation. The lung tissues were then embedded in paraffin and sections were cut at our core facility. The sections were then processed for H & E staining. Briefly, lung tissue sections mounted on slides were first transferred in xylene to deparaffinized the tissue sections. The tissue sections were then rehydrated in alcohol (100, 95, and 90%). The sections were finally stained with Hematoxylin and Eosin (H & E) and dehydrated. H & E stained sections were analyzed using KEYENCE digital microscope VHX-7000 (IL-United States).

These cells were isolated as described previously (Rieder et al., 2012). Briefly, mice were given heparinized PBS to perfuse the lungs. A stomacher 80 biomaster blender (Seward, Fl) was used to homogenize the lung tissue suspended in 10 ml of 1× PBS and 10% Fetal Bovine Serum (FBS). The homogenized lung tissues were centrifuged at 1,200 rpm at 4°C for 12 min. The pellet was then washed with PBS and resuspended in PBS post centrifugation. We then used a density gradient centrifugation method to purify lung mononuclear cells (MNCs). In brief, the cells resuspended in sterile PBS were carefully layered using Histopaque-1077 purchased from Sigma-Aldrich (St. Louis, MO) and centrifuged at 500 × g for 30 min at room temperature (25°C). The mononuclear cells were collected at the interface. MNCs mixed with trypan blue were then enumerated using a Biorad TC 20-Automated cell counter.

Blood samples from each group of mice were collected under general anesthesia. The collected blood samples were then centrifuged to collect the sera. The sera were either used immediately or stored at −80°C. The detection of the cytokines was performed using ELISA MAX standards kits from Biolegend.

Spleen cells and MNCs isolated from the lungs were stained with various fluorescent–conjugated antibodies purchased from Biolegend (SanDiego, CA, United States), which included: For T cell subsets: Allophycocyanin (APC) conjugated anti-CD3, Phycoerythrin (PE) conjugated anti-CD4, PerCP/Cyanine 5.5 (PerCP-CY5.5) conjugated anti-CD8, Fluorescein Isothiocyanate (FITC) conjugated anti-Vβ8, and PE/Dazzle Conjugated anti-NK1.1. For lymphocyte activation markers, T cell memory markers, MDSCs, and Tregs, the following Abs were used: Phycoerythrin (PE) Conjugated anti-CD69, Brilliant Violet-510 Conjugated anti-CD25, Fluorescein Isothiocyanate (FITC) Conjugated anti-CD44, Brilliant Violet-785 Conjugated anti-CD3, Allophycocyanin (APC) Conjugated anti-CD62L, Alexa Fluor 488 Conjugated anti-FoxP3, Phycoerythrin (PE) Conjugated anti-IL10, Fluorescein Isothiocyanate (FITC) Conjugated anti-CD11b, Phycoerythrin (PE) Conjugated anti-Gr1, Alexa Fluor 700 conjugated anti-LY6C, and Phycoerythrin cy5 (PECY5) conjugated anti-LY6G.

The staining of cells for dual markers was performed as described in our previous publications (Al-Ghezi et al., 2019; Becker et al., 2020; Dopkins et al., 2020; Mohammed et al., 2020b). In brief, MNCs and spleen cells were treated with Fc Block reagent (BD Pharmingen, San Diego, CA) followed by staining cells with various antibodies by incubation at 4°C for 20–30 min. Stained cells were washed twice with cold PBS containing 2% FBS (Staining buffer). For IL 10 and FOXP3 staining, we used the Intracellular Cytokine Staining Kit (BD biosciences) consisting of fixation/permeabilization buffer. In brief, the cells were fixed using fixation/permeabilization buffer, followed by washing the cells with PBS and then staining the cells using anti–Foxp3 and anti–IL10 antibodies. The cells were finally suspended in 0.5 ml staining buffer and analyzed using Beckman Coulter FC500 or BD Bioscience FACSCelesta^™, which was followed by analysis on the FlowJo V10 version software.

Mice were injected with Anandamide and SEB, as described earlier. Forty eight hrs later, mice were euthanized and the peritoneal exudate was collected and washed with PBS. The cells were resuspended in 1 ml, treated with Fc block for 10 min at RT. Next, the EasySep CD11b Gr1 selection kit (STEMCELL Technologies, Seattle, WA; catalog number 19867) was used to isolate CD11bGr1 based on the guidelines provided in the kit.

To test the immunosuppressive effects of MDSCs on T cell proliferation, naive C57BL/6 mice were euthanized and splenic T cells were isolated. Splenic T cells at a concentration of 5 × 10⁵ cells/well were cultured in 96-well tissue culture plates in the presence of Concanavalin A (ConA), a T cell mitogen, at a concentration of (2.5 μg/ml) together with different ratios of AEA induced purified MDSCs for 48 h. (³H)thymidine (2 μCi per well) was added to the cell culture in the last 18 h, and the radioactivity was measured using a liquid-scintillation counter (MicroBeta TriLux; PerkinElmer).

Micro-RNA analysis was performed as described previously (Tomar et al., 2015; Alghetaa et al., 2018). Total RNA including miRNA was isolated from lung infiltrating MNCs using miRNeasy kit from Qiagen. A Nanodrop 2000 spectrophotometer (Thermo Fisher Scientific, Wilmington, DE) was used to determine RNA purity and concentration. Next, Affymetrix gene chip miRNA 4.0 array platform was used (Affymetrix Inc., Santa Clara, CA) to determine the miRNA expression profile of lung MNCs and only those miRNAs that were altered more than 1.5-fold or higher were considered for further analysis. Ingenuity Pathway Analysis (IPA) was used for identifying the role of miRNA in various biological pathways using the software available at http://www.ingenuity.com. Also, the microRNA.org database was used to examine the sequence alignment regions between miRNA-23a-3p and miRNA-34a-5p and their respective target genes.

This was performed as described previously (Mohammed et al., 2020a). Briefly, to determine the expression of the selected miRNA and respective genes, qRT-PCR was performed using cDNA synthesized from total RNA and miRNA. miScript II RT kit from Qiagen was used to synthesize cDNA using and following the protocol of Qiagen company.

Snord96A and GAPDH were used as endogenous controls. 2^ΔΔCt was used to determine fold change in expression the following primers were used for qRT-PCR (Table 1).

Transfection was performed as described previously (Busbee et al., 2015). In brief, splenocytes were cultured in complete RPMI 1640 medium (10% FBS, 10 mM L-glutamine, 10 mM Hepes, 50 μM β-mercaptoethanol, and 100 μg/ml penicillin). 2.5 × 10⁵ cells/well were seeded in a 24-well plate and transfected with mock control, synthetic mimic mmu-miRNA-23a-3p MSY0000532, synthetic mimic mmu-mmiRNA-34a-5p MIN0000532 using a HiPerFect transfection reagent kit from Qiagen.

Graph Pad Prism 6.0 Software (San Diego, CA, United States) was used for statistical analyses. Student’s t-test was used for comparison among two groups. One way-ANOVA with post hoc Tukey’s test was used to compare between more than two groups. The results were expressed as mean ± SEM. A p-value ≤ of 0.05 was considered statistically significant.

We used the following groups of mice to study the effect of AEA on SEB-mediated ARDS: Naïve, SEB + VEH, and SEB + AEA. After 48 h of SEB exposure, the lung function of all groups was evaluated using a Buxco instrument system. We observed that the clinical parameters for lung function including specific airways resistance, specific airway conductance, and minute per volume to be significantly improved in SEB + AEA groups and similar to the naïve group, in comparison to SEB + VEH mice (Figure 1A). These data demonstrated that AEA was able to rescue impairments to the lung function caused by SEB exposure. Further, the total number of mononuclear cells (MNCs) in the lung was significantly decreased in SEB + AEA mice when compared to SEB + VEH mice (Figure 1B). Histological sections of the lungs are shown in Figure 1C that are representative pictures of the data presented in Figure 1B.

Analysis of sera obtained from mice showed that there was a significant decrease in the pro-inflammatory cytokines TNFα, IL-2, and IFN-γ in SEB + AEA group when compared to SEB + VEH mice (Figure 1D).

We then assessed the presence of T cells in the lungs (Figures 2A–D) and spleens (Figures 2E–H) by staining with fluorescein-conjugated antibodies and analyzing using flow cytometry. The data have been depicted as a representative flow cytometric analysis followed by percentage of cells/mouse, and total number of cells/mouse in vertical bars. Results showed that there was a significant increase in T cell subset populations, both in percentages and numbers, including CD4⁺ T cells, CD8⁺ T cells, Vβ8+ T cells, and NK T cells in the lungs and spleens of SEB + VEH group when compared to the naïve mice, thereby demonstrating that SEB caused activation and proliferation of these cells (Figures 2A–H). Interestingly, SEB + AEA group when compared to the SEB + VEH group, showed a significant decrease in the percentage and numbers of all T cell subsets, including the Vβ8+ T cells that are specifically activated by SEB. This combined with the fact that SEB + AEA group had a decrease in the total number of infiltrating MNCs in the lungs (Figure 1B) when compared to SEB + VEH group, clearly demonstrated that AEA decreases both the percentage and total numbers of T cell subsets induced by SEB.

To determine whether the AEA suppresses SEB-activated proliferation of T cells directly, we performed in vitro assays in which splenic lymphocytes, isolated from naïve mice, were pretreated with a dose of 20 μM of AEA and activated 30 min later with 1 μg/ml of SEB. The cells were then incubated for 24 and 48 h. We assessed the effect of AEA on T cell subpopulations, including CD4⁺ T cells, CD8⁺ T cells, Vβ8 + T cells, and NKT cells by flow cytometry. We observed that there was a significant decrease in the percentage of all T cell subsets and NKT cells in the SEB + AEA group when compared to the SEB + VEH group at 24 h (Figures 3A–D), and this decrease was further augmented at 48 h (Figures 3E–H). Next, we investigated activation markers such as CD3, CD69, CD25, and CD44, as well as CD62L (L-slectin) which is expressed on naïve CD4⁺ T cells used for recirculation and binding to high endothelial venules and on central memory T cells. We observed that the percentages of CD3⁺CD69⁺, CD69⁺, CD44⁺, and CD62L + cells from the SEB + Veh group were significantly increased when compared to naïve mice indicating that SEB had activated a significant proportion of T and other immune cells. However, in the SEB + AEA group, there was a significant decrease in percentages of these cells when compared to the SEB + VEH group except CD62L + cells which were enhanced in SEB + AEA mice when compared to the SEB + VEH group (Figure 4). It should be noted that when we looked at CD25 ⁺ cells that were CD69-, these cells were significantly increased SEB + AEA group (33.6%) when compared to SEB + VEH (7.9%) (Figure 4). Because CD25 is also expressed on Tregs, these data suggested that AEA may increase the proportion of Tregs which is shown in subsequent Figs. Together, these in vitro studies demonstrated that AEA directly worked on SEB-activated immune cells thereby significantly suppressing their activation.

We next analyzed miRNA profiles of lung infiltrating MNCs to examine the extent to which AEA attenuates inflammation via miRNA induction. The data were analyzed using Ingenuity Pathway Analysis (IPA) software as described earlier (Rao et al., 2015a). The heat map of miRNAs (>3,000) was first generated which showed marked alterations in miRNA expression in SEB + VEH mice when compared to the SEB + AEA mice. Upon analysis of a ≥1.5 fold increase or decrease in miRNA expression, a marked difference in the upregulation/downregulation of miRNAs between SEB + VEH and SEB + AEA mice was observed (Figure 5A). Specifically, there were at least 77 miRNAs that were upregulated and 59 miRNAs that were downregulated in SEB + AEA mice when compared to SEB + VEH mice (Figure 5B). Next, we analyzed select miRNAs for their targets especially those involved in the regulation of inflammation, using the IPA software from Qiagen and observed that some of the downregulated miRNAs, specifically miRNA-23a-3p targeted TGF-β2, ARG1, and miRNA-34a-5p (synonym: miRNA-449c based on IPA classification) targeted FoxP3 (Figure 5C). Also, we found miRNA-34a-5p to target GATA3 involved in Th2 differentiation and miRNA-30c-5p which was down-regulated to target SOCS1, a suppressor of cytokine signaling. These data suggested that AEA, by downregulating these miRNAs, may induce several of these anti-inflammatory molecules such as TGF- β2, ARG1, GATA3, and SOCS1.

We further confirmed miRNA-23a-3p and miRNA-34a-5p specific target genes using miRNA.org or TargetScan software. miRNA-23a-3p showed a strong binding affinity with its complementary 3′UTR region of TGF-β2 gene and miRNA-34-5pa showed a strong binding affinity with the 3′UTR region of the FoxP3 gene (Figures 6A,E). We then validated the expression of miRNA-23a-3p and miRNA-34a-5p and their respective genes (ARG1, TGF-β2, and FoxP3) by performing qRT-PCR using MNCs isolated from the lungs. Both miRNA-23a-3p and miRNA-34a-5p were significantly downregulated in SEB + AEA mice when compared to SEB + VEH counterparts (Figures 6B,F). Upon analysis of their respective target genes including ARG1, TGF-β2, and FoxP3 in the lung MNCs, there was a significant upregulation of both genes in cells from SEB + AEA mice when compared to SEB + VEH mice (Figures 6C,D,G).

To corroborate that miRNA-23a-3p and miRNA-34a-5p specifically regulate the respective target genes (ARG1, TGF-β2, and FoxP3), we performed a series of transfection assays. To this end, SEB at a concentration of 1 μg/ml was used to activate splenocytes following overnight incubation. The following day, activated cells were then transfected with mock, mimic or inhibitor of miRNA-23a-3p, or miRNA-34a-5p. 48 h after transfection, we collected the cells for qRT-PCR. Based on the qRT-PCR analysis, we observed that there was a significant increase in miR-23a (Figure 7A) and miR-34a-5p expression (Figure 7D) in cells transfected with their respective mimics. In contrast, the cells that were transfected with respective inhibitors showed a significant decrease in miRNA-23a and miRNA-34a-5p expression. Upon analysis of their target genes, we noted that there was significant suppression of TGF-β2 and ARG1 in miR-23a mimic-transfected cells (Figures 7B,C) as well as FoxP3 expression in miRNA-34-5p mimic-transfected cells (Figure 7E), whereas there was complete reversal (increase) in their expression (Figures 7B,C,E) following inhibition of their respective miRNAs. These findings indicated a direct relationship with miRNA-23a-3p and miRNA-34a-5p expression with their target genes ARG1, TGF-β2, and FoxP3.

The above studies indicated that AEA-regulated miRNAs might induce anti-inflammatory genes leading to the generation of suppressor cells such as MDSCs and T regulatory cells (Tregs). To test this, MNCs from the lungs of both SEB + AEA and SEB + VEH mice were assessed for the generation of MDSCs and their subsets (CD11b + Gr1+ and LY6C + LY6G+). Flow cytometry data showed that there was a significant increase in the percentage of MDSCs (CD11b + Gr1+) in SEB + AEA mice when compared to SEB + VEH mice in both the lungs and spleen (Figures 8A,E). Additionally, we noted that AEA caused increased percentages of both Ly6G⁺Ly6C^low granulocytic and Ly6G⁻Ly6C^high monocytic MDSCs (Figure 8B). Furthermore, upon analysis of Tregs, there was a significant increase in CD4 + FoxP3 + cells in SEB + AEA mice (Figure 8C) when compared to the SEB + VEH group. A significant increase of CD4 + IL10 + cells, a phenotype of regulatory Type 1 (Tr1)-like cells, was observed as well in SEB + AEA mice when compared to SEB + VEH mice, (Figure 8D). To confirm that the MDSCs induced by AEA were immunosuppressive, we performed an in vitro assay in which spleen T cells from naïve mice were activated with ConA in the presence of varying proportions of purified MDSCs. The Data shown in Figure 8F demonstrated that the MDSCs caused a dose-dependent decrease in T cell proliferation. These data together suggested that AEA was inducing immunosuppressive MDSCs and Tregs, consistent with the data shown in miRNA experiments.