Agricultural dust derived bacterial extracellular vesicle mediated inflammation is attenuated by DHA

Abstract

Dietary long-chain omega-3 polyunsaturated fatty acids (n-3 PUFA) and their pro-resolving metabolites are protective against atherosclerotic disease, and ameliorate systemic inflammatory conditions including lupus erythematosus, psoriasis, and bronchial asthma. Organic bioaerosol inhalation is a common and injurious hazard associated with agricultural occupations such as work in swine concentrated animal feeding operations (CAFOs) and is known to increase the risk for developing respiratory conditions such as asthma and COPD. Nearly all cells secrete membrane-bound vesicles (extracellular vesicles, EVs) that have the capacity to transmit protein, nucleic acid, and lipid signaling mediators between cells. Using a polymer-based isolation technique (ExoQuick, PEG) followed by ultracentrifugation, EVs were isolated from CAFO dust extracts, and were quantified and partially characterized. Here, we investigated the role of the n-3 PUFA docosahexaenoic acid (DHA) as a component of n-6 to n-3 PUFA mixtures used to recapitulate physiologically relevant dietary ratios in the resolution of inflammatory injury caused by exposure to EVs carried by agricultural organic dust in vitro. Primary human bronchial epithelial cells, fibroblasts and monocyte-derived macrophages were exposed to EVs isolated from swine CAFO dust. Cells were treated with mixtures of n-6 and n-3 PUFA during recovery from the EV-induced injury. CAFO dust extract (DE) was found to contain EVs that contributed significantly to the overall consequences of exposure to complete DE. DHA-rich PUFA ratios inhibited DE-derived EV-induced proinflammatory cytokine release dose-dependently. DHA-rich PUFA ratios also reversed the damaging effects of EVs on recellularization of lung matrix scaffolds, accelerated wound healing, and stimulated the release of pro-resolution mediators. These results underscore the importance of n-3 PUFA as anti-inflammatory compounds during recovery from EV-laden environmental dust exposure in the context of cellular responses in vitro, warranting future translational studies.

Figure 1

EVs are abundant in DE-derived dust samples. Nanoparticle tracking analysis of 100% DE-derived EVs indicated a concentration range of 2.5–3.8 × 10¹⁰ particles/mL and a mean size of 215 nm (A). Representative transmission electron micrographs illustrate the concentration and size distribution of EV preps (B). In all subsequent experiments the Sch D DE-derived EVs were used.

Figure 2

Dust-derived EVs contain DNA from both Gram (+) and Gram (−) bacterial species, as well as viral (bacteriophage) signatures. The 50 most prevalent signatures originated predominantly from Gram (+) bacteria (65%) commonly found in feces, soil, and as part of the normal swine gut flora. Several pathologic species were also identified.

Figure 3

DE-derived EVs dose-dependently stimulated inflammatory mediator release from bronchial epithelial cells, and detergent lysis of EVs reduced their potency. BEAS-2B cells were treated with various concentrations of DE-derived EVs, or 5% complete DE, and incubated for 24 h. IL-6, IL-8, and AREG were measured in culture supernates (A). Soluble protein release from HBEC challenged with EVs recovered from DE treated with doses of Triton X-100 is shown in (B). The medians ± quartiles of 3 independent experiments are shown (n = 12 technical replicates per condition for (A) and 8 replicates for (B). *p < 0.05, **p < 0.01, ***p < 0.0001 vs control (or for indicated comparisons in (B) (ANOVA, Tukey’s post-hoc test).

Figure 4

Stimulatory effects of EVs are sensitive to heating but are not due to LPS contamination and the effects of complete DE are partially due to the EV cargo. Saturated (100%) DE was subjected to heat (90 °C for 15 min) or LPS ablation (polymyxin B, 12 h @ 4 °C). HBEC were challenged with EVs isolated from these preparations (5% DE equivalents) or with heat-or polymyxin B-treated 5% DE for 24 h (A). Polymyxin B treatment significantly depleted LPS from EV suspensions (B). Alternatively, cells were challenged with complete 5% DE, an equivalent concentration of DE-derived EVs, or DE devoid of EVs, for 24 h. Supernatant medium was assessed for IL-6, IL-8 and AREG levels by ELISA (C). EV abundance (NTA) as a proportion of total protein for 5 different preps is shown in (D). Data shown for (A) and (C) are means of 3 independent experiments (8 or 14 technical replicates per condition). *p < 0.05, **p < 0.01, ***p < 0.0001 vs control (AVOVA, Tukey’s post-hoc test). For (B) and (D), 6 technical replicates per condition, ***p < 0.001, Student’s t test.

Figure 5

DHA-rich PUFA ratios restored the EV-induced suppression of recellularization on lung matrix scaffolds. Human lung scaffolds were seeded with HBEC, challenged with 50 × 10⁹ EVs/mL for 5d (“exposure,” solid bars) then allowed to recover in the presence of PUFA ratios for an additional 3d (“recovery,” hatched bars). DE-derived EVs hindered the repopulation of scaffolds during both exposure and recovery, while DHA-rich mixtures reversed the recellularization deficit. *p < 0.05, ***p < 0.0001 (ANOVA, Tukey’s post-test, n = 3 experiments, 12 or 16 technical replicates per condition).

Figure 6

EV-induced wound resolution deficit was reversed in the presence of DHA-rich PUFA ratios. Circular wounds made in confluent monolayers of HBEC (A) or HLF (B) were allowed to heal in the presence of control medium, DE-derived EVs, or EVs with one of three PUFA ratios. The wound area was analyzed over the course of 24 h, compared to the size of the initial wound, and percent closure was calculated. DHA-rich PUFA ratios (1:1) significantly reversed the DE-mediated wound closure deficit. Results shown are means for 3 independent experiments, 12 (A) or 9 (B) technical replicates per condition. **p < 0.01, ***p < 0.0001 vs. Control, ^##p < 0.001 vs. EVs alone at corresponding time points, (Student’s t test for pairwise comparisons).

Figure 7

DHA-rich PUFA ratios dampened DE-derived EV-induced inflammatory cytokine release, but augmented AREG release during recovery. HBEC were stimulated with concentrated DE-derived EVs (50 × 10⁹/mL) for 2 h, EVs were then removed, and cultures were allowed to recover in the presence of PUFA ratios for 22 h. Release of EV-mediated proinflammatory cytokines IL-6 and IL-8 was attenuated, and pro-repair AREG was augmented by treatment with the DHA-rich (1:1) PUFA ratio and by DHA alone during the recovery phase. Data shown are means of 4 independent experiments with 24 technical replicates per condition. *p < 0.05, **p < 0.01, ***p < 0.0001 (ANOVA, Tukey’s posttest).

Figure 8

The DHA-rich PUFA ratio (1:1) inhibited EV-mediated proinflammatory cytokine release but augmented pro-repair mediator release from MDM during recovery. Sub-confluent monocyte-derived macrophages were challenged with EVs for 2 h, then allowed to recover in the presence of PUFA ratios for 22 h. The DHA-rich ratio dampened the EV-induced release of inflammatory cytokines (IL-8, TNFα) and elevated the release of pro-repair mediators RvD1, AREG, IL-10. Data pooled from n = 5 experiments, 24 technical replicates. **p < 0.01, ***p < 0.0001, for indicated comparisons (ANOVA, Tukey’s post-hoc test).

Figure 9

Schematic overview. Mechanism of n-3 PUFA-modulated suppression of EV-induced inflammation. Exogenous DHA competes with the more proinflammatory n-6 PUFA arachidonic acid (AA) for incorporation into cell membranes (1). Inside the cell, free DHA inhibits enzymatic conversion of AA to eicosanoids by direct competition (2). Specialized pro-resolving mediators are synthesized from free cytoplasmic DHA (3). Binding of exogenous DHA to the GPR120 surface receptor interrupts NFκB-mediated inflammatory cytokines (4).