Norepinephrine Inhibits Lipopolysaccharide-Stimulated TNF-α but Not Oxylipin Induction in n-3/n-6 PUFA-Enriched Cultures of Circumventricular Organs

Abstract

Sensory circumventricular organs (sCVOs) are pivotal brain structures involved in immune-to-brain communication with a leaky blood-brain barrier that detect circulating mediators such as lipopolysaccharide (LPS). Here, we aimed to investigate the potential of sCVOs to produce n-3 and n-6 oxylipins after LPS-stimulation. Moreover, we investigated if norepinephrine (NE) co-treatment can alter cytokine- and oxylipin-release. Thus, we stimulated rat primary neuroglial sCVO cultures under n-3- or n-6-enriched conditions with LPS or saline combined with NE or vehicle. Supernatants were assessed for cytokines by bioassays and oxylipins by HPLC-MS/MS. Expression of signaling pathways and enzymes were analyzed by RT-PCR. Tumor necrosis factor (TNF)α bioactivity and signaling, IL-10 expression, and cyclooxygenase (COX)2 were increased, epoxide hydroxylase (Ephx)2 was reduced, and lipoxygenase 15-(LOX) was not changed by LPS stimulation. Moreover, LPS induced increased levels of several n-6-derived oxylipins, including the COX-2 metabolite 15d-prostaglandin-J2 or the Ephx2 metabolite 14,15-DHET. For n-3-derived oxylipins, some were down- and some were upregulated, including 15-LOX-derived neuroprotectin D1 and 18-HEPE, known for their anti-inflammatory potential. While the LPS-induced increase in TNFα levels was significantly reduced by NE, oxylipins were not significantly altered by NE or changes in TNFα levels. In conclusion, LPS-induced oxylipins may play an important functional role in sCVOs for immune-to-brain communication.

Keywords: circumventricular organs; cytokines; immune-to-brain communication; lipopolysaccharide; oxylipins.

Figure 1

Bioassay cytokine measurements from supernatants: Release of the proinflammatory cytokines tumor necrosis factor α (TNFα, (a) and interleukin (IL)-6 (b) into the supernatants of primary sensory circumventricular organ (sCVO) cultures. Cells were stimulated with lipopolysaccharide (LPS, 10 µg/mL) or saline and simultaneously treated with norepinephrine (NE, 1 µmol/L) or vehicle for 4 h. For every sample set, cytokine concentrations are plotted as fold change to the LPS-treated sample (LPS+vehicle) from six independent experiments. TNFα and IL-6 secretion increased after LPS stimulation (* main effect LPS stimulation), whereas NE treatment lowered basal and LPS-induced secretion of both cytokines into the supernatants (+ main effect NE treatment). LPS-induced TNFα release was significantly dampened by NE (# LPS+vehicle vs. LPS+NE). n = 6 per group; two-factorial ANOVA, Bonferroni post hoc test, *** p < 0.001; ⁺ p < 0.05, ⁺⁺ p < 0.01; ## < 0.01.

Figure 2

mRNA expression of inflammatory markers and lipid mediator metabolism: mRNA expression was assessed for markers of inflammatory pathway activation (a–d) and lipid mediator metabolism enzymes (e–g). Cells were stimulated with lipopolysaccharide (LPS, 10 µg/mL) or saline and simultaneously treated with norepinephrine (NE, 1 µmol/L) or vehicle for 4 h. The 2-ΔΔct method was applied to present relative quantities as fold change of the lowest expression. Pro-inflammatory pathway-activation-marker inhibitor (I)κB, (a) and suppressor of cytokine signaling (SOCS)3, (b) and the anti-inflammatory cytokine interleukin (IL)-10 (d) increased due to LPS stimulation. The rate-limiting enzyme for prostaglandin production, i.e., cyclooxygenase (COX)-2 (e), showed an LPS-induced increase, whereas there were reduced epoxide hydrolase (Ephx)2 (g) mRNA levels in LPS-stimulated groups. Peroxisome proliferator-activated receptor gamma coactivator ((PGC)1α, (c)) and arachidonate 15-lipoxygenase (ALOX15, (f)) mRNA expression were not significantly affected by LPS-stimulation or NE treatment. n = 6 per group; two-factorial ANOVA, Bonferroni post hoc test, * main effect LPS stimulation; * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 3

n-6 lipid mediators and metabolites from supernatants: detection and quantification of n-6 lipid mediators and metabolites derived from arachidonic acid (AA) (a–h) or linoleic acid (LA) (i–k) from supernatants using LC-MS/MS. Primary sensory circumventricular organ (sCVO) cell cultures were stimulated with lipopolysaccharide (LPS 10 µg/mL) or saline and simultaneously treated with norepinephrine (NE, 1 µmol/L) or vehicle for 4 h. For every sample set, metabolite concentrations are plotted as fold change to the LPS-stimulated sample (LPS+vehicle) for each of the six independent experiments. AA-derived metabolites elevated in LPS-stimulated samples: (a) 15-HETE (hydroxy-eicosatetraenoic acid); (b) 8,15-DiHETE (dihydroxy-eicosatetraenoic acid), (c) 15-OxoETE (oxo-eicosatetraenoic acid); (d) 5-OxoETE; (f) 11,12-DHET (dihydroxy-eicosatrienoic acid); (g) 14,15-DHET; (h) 15d-PGJ2 (15-Deoxy-Delta-12,14-prostaglandin-J2); (e) 18-HETE was reduced in LPS-stimulated samples. LPS-induced LA metabolites showed higher levels in supernatants compared to controls: (i) 13-HODE (hydroxy-octadecadienoic acid); (j) 9,10-DiHOME (dihydroxy-octadecenoic acid); (k) 12,13-DiHOME; n = 6 per group, two-factorial ANOVA, Bonferroni post hoc test, * main effect LPS stimulation, * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 4

n-3 lipid mediators and metabolites from supernatants: detection and quantification of n-3 lipid mediators and metabolites derived from eicosapentaenoic acid (EPA, (a–e)) or docosahexaenoic acid (DHA, (f)) from supernatants using LC-MS/MS. Primary sensory circumventricular organ (sCVO) cell cultures were stimulated with lipopolysaccharide (LPS, 10 µg/mL) or saline and simultaneously treated with norepinephrine (NE, 1 µmol/L) or vehicle for 4 h. For every sample set, metabolite concentrations are plotted as fold change to the LPS-stimulated sample (LPS+vehicle) for each of the six independent experiments. The EPA metabolites LXA5 (Lipoxin A5, (a)) and 18-HEPE (hydroxy-eicosapentaenoic acid, (e)) were elevated in supernatants of LPS-stimulated cells, whereas 8-HEPE (b), 9-HEPE (c), 12-HEPE (d) levels were reduced due to LPS-stimulation when compared to controls. DHA-derived NPD1 (neuroprotectin D1, (f)) showed LPS-induced increase. n = 6 per group, two-factorial ANOVA, Bonferroni post hoc test, * main effect LPS stimulation; * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 5

Summary of results and lipid mediator metabolism pathways in lipopolysaccharide (LPS)-stimulated CVO cultures: simplified pathway models of polyunsaturated fatty acids’ (PUFA) metabolism with literature-based “checkpoint” enzymes; n-6 fatty acids (linoleic acid, yellow; arachidonic acid, red) and n-3 fatty acids (eicosapentaenoic acid, green; docosahexaenoic acid, blue) are metabolized by the same enzymes: cyclooxygenases (COX), lipoxygenases (LOX), cytochrome P450 monoxygenases (CYP450), and epoxide hydrolase 2 (Ephx2). Other pathway enzymes are not displayed for better visualization. Increased or decreased levels of metabolites in supernatants or cells’ mRNA expression of enzymes are displayed by arrows (increased ↑, decreased ↓). Substrates (fatty acids) and enzymes not measured by us are visualized in color (fatty acids) or in italic with an asterisk (*). Colored lines depict metabolic pathways via enzymes to produce oxylipins from their substrates. Sensory circumventricular organ (sCVO)-derived primary cell cultures were pre-incubated with PUFA (each 50 µmol/L). LPS-induced (10 µg/mL) inflammatory stimulation of cell cultures for 4 h induced mRNA expression of COX-2 and decreased mRNA expression of Ephx2. ALOX15 expression was not affected by LPS stimulation. Enzymes’ activity led to altered release of lipid metabolites into supernatants, as depicted by arrows. Arachidonic acid-derived metabolites were elevated in LPS-stimulated samples: 15-HETE (hydroxy-eicosatetraenoic acid); 8,15-DiHETE (dihydroxy-eicosatetraenoic acid), 15-OxoETE (oxo-eicosatetraenoic acid); 5-OxoETE; 11,12-DHET (dihydroxy-eicosatrienoic acid); 14,15-DHET; 15d-PGJ2 (15-Deoxy-Delta-12,14-prostaglandin-J2). The arachidonic acid metabolite 18-HETE was reduced in LPS-stimulated samples when compared to controls. Linoleic acid metabolites were increased in supernatants by LPS: 13-HODE (hydroxy-octadecadienoic acid); 9,10-DiHOME (dihydroxy-octadecenoic acid); 12,13-DiHOME. Eicosapentaenoic acid metabolites LXA5 (Lipoxin A5) and 18-HEPE (hydroxy-eicosapentaenoic acid) were elevated in supernatants of LPS-stimulated cells, whereas 8-HEPE, 9-HEPE, and 12-HEPE levels were reduced due to LPS stimulation versus controls. Docosahexaenoic-derived NPD1 (neuroprotectin D1) was observed to be increased by LPS stimulation. Metabolic pathways are visualized according to [6,99,100].