Effects of kynurenic acid and choline on lipopolysaccharide-induced cyclooxygenase pathway

Introduction

Bacterial endotoxin lipopolysaccharide (LPS) initiates an inflammatory response that produces pro-inflammatory cytokines by activating a member of toll-like receptors, TLR4, among “pathogen-associated molecular pattern (PAMP) recognition receptors” expressed in immune cells [1]. Although acting against infections, the uncontrolled release of cytokines causes widespread inflammatory response (as in the case of cytokine storm).

The cholinergic anti-inflammatory pathway (CAP) can control inflammatory response endogenously. Alpha-7 nicotinic acetylcholine receptors (α7nAChRs) expressed on immune cells, including macrophages, are highly permeable to calcium (Ca²⁺) and, upon activation, the increased intracellular Ca²⁺ levels trigger a variety of downstream signalling cascades which provides interactions between parasympathetic and the immune systems leading to activation of CAP [2, 3]. Evidence suggests that α7nAChR-activating agents modulate LPS-induced cytokine release in different in-vivo and in-vitro experimental models [4] and produce anti-inflammatory effects [5, 6]. Choline, a precursor of ACh, enhances cholinergic activity by directly interacting with cholinergic receptors at high concentrations [7]. Several studies showed that activation of CAP by choline via α7nAChRs reduces the levels of pro-inflammatory factors from endotoxin-induced mononuclear cells via nuclear factor kappa B (NF-κB) pathway [8].

The majority of tryptophan, a precursor of serotonin and kynurenic acid (KYNA), is metabolized via the kynurenine pathway while the remaining 5 % participates in the serotonin/melatonin pathway [9]. Kynurenic acid produces prophylactic effects in different neurodegenerative disorders [10]. Moreover, KYNA has been shown to activate G-protein-coupled receptor 35 (GPR35) to produce anti-inflammatory and antioxidant actions [11, 12]. Regarding the interaction with the cholinergic system, there is controversy about the action of KYNA on α7nAChRs as KYNA was reportedly defined as an α7nAChR antagonist earlier [13]. However, its antagonistic property has never been confirmed [14].

Prostaglandins (PGs) are produced from arachidonic acid by two cyclooxygenase (COX) enzymes. The COX-1 enzyme is mainly associated with homeostasis, whereas COX-2, the inducible isoform, is stimulated by cytokines in inflamed tissues, eventually contributing PG synthesis in inflammation [15]. We previously showed that choline downregulates LPS-induced COX pathway along with PG levels in central and peripheral tissues of endotoxemic rats [16]. However, the interaction between endogenous choline and KYNA on inflammatory pathways is unclear. Thus, this study aimed to examine the role of choline and KYNA in the LPS-induced COX-2 pathway via α7nAChR along with changes in intracellular Ca²⁺ levels, pro-inflammatory mediator TNFα and NF-κB.

Materials and methods

Results

Effects of choline and KYNA on cell viability

RAW 264.7 cells were incubated with increasing choline and KYNA (1–3–10–30–100 μM) for 24 h, and results were normalized against the control (n=5). The lack of cytotoxicity confirmed the suitability of the concentration range for the following experiments (Figure 1).

Figure 1:

Effects of choline and KYNA on viability of RAW264.7 cells. Shown are the effects of choline (A), KYNA (B) on cell viability assessed by MTT assay, n=5. KYNA, kynurenic acid.

Effects of choline and KYNA on LPS-induced COX-2 levels

COX-2 mRNA expressions concentration-dependently increased in LPS-treated groups (p<0.001, n=5) comparable to control (Figure 2A). The minimally effective LPS concentration, which induced COX-2-mediated response (both COX-2 and PGE₂) was determined as 1 μg/mL (n=3, Figure 3).

Figure 2:

Effects of choline and KYNA on LPS-induced COX-2 mRNA expressions. Shown are the effects of LPS (A), choline (B), nicotinic receptor antagonists (C), KYNA (D), choline + KYNA (E) on COX-2 expressions. ***p<0.001, vs. control; ^††p<0.01, ^†††p<0.001 vs. LPS group; ^‡p<0.05, ^‡‡p<0.01 vs. LPS + choline, one-way ANOVA with post-hoc Tukey-Kramer multiple comparison test or Student’s t-test, n=5. MEC, mecamylamine; MLA, methyllycaconitine; KYNA, kynurenic acid.

Figure 3:

Effects of choline and KYNA on LPS-induced COX-2 protein expressions. Shown are the effects of LPS, choline, nAChR antagonists, MEC and MLA, KYNA, choline + KYNA on COX-2 protein expressions. **p<0.01, vs. control; ^†p<0.05 vs. LPS group, one-way ANOVA with post-hoc Tukey-Kramer multiple comparison test or Student’s t-test, n=3. MEC, mecamylamine; MLA, methyllycaconitine; KYNA, kynurenic acid; CH, choline.

Choline decreased LPS-induced COX-2 mRNA (p<0.001, n=5, Figure 2B) and COX-2 protein concentration-dependently (at 10 μM or higher concentration) comparable to LPS control group (n=3, Figure 3). COX-2 mRNA and protein expressions increased in MEC and MLA-treated groups (p<0.05 and 0.01, respectively) comparable to choline-treated group (n=5, Figure 2C and n=3, Figure 3). KYNA decreased LPS-induced COX-2 mRNA in a concentration-dependent manner (p<0.001, n=5, Figure 2D) which were also confirmed by WB data (n=3, Figure 3). COX-2 mRNA expression levels decreased significantly in KYNA- and choline treated groups (p<0.01 in each, n=5) comparable to choline-only treatment (Figure 2E). Equieffective concentrations of KYNA- and choline (10 μM) treatment decreased COX-2 protein levels comparable to choline- and LPS-treated groups (n=3, Figure 3).

Discussion

The present study shows that choline significantly inhibited LPS-induced COX-2 and PGE₂ levels by modulating the cholinergic anti-inflammatory pathway (CAP) via α7nAChR activation. Additionally, the effects of choline were potentiated in the presence of a tryptophan metabolite, KYNA, suggesting its possible interaction with cholinergic control of LPS-induced response in macrophages.

The anti-inflammatory effects α7nAChR agonists have been widely investigated in LPS-stimulated RAW264.7 cells in vitro [5, 8]. Earlier, significant decrease in pro-inflammatory cytokine levels including TNF-α, interleukins (IL-1β, -6, -18) and HMGB1 by α7nAChR-mediated activation of CAP in LPS-activated macrophages suggested that α7nAChRs would be identified as target receptors in the treatment of inflammatory conditions [6]. Furthermore, COX-2 expression in macrophages has been shown to be increased in response to LPS [16, 19] besides elevated pro-inflammatory cytokine, TNF-α, which also induces PLA₂ (phospholipase A2) and COX-2 activities [20, 21].

Earlier studies showed conflicting results regarding the role of cholinergic agonists and COX-2 pathway in several cell types. In one study, tacrine, an ACh esterase inhibitor, has been shown to decrease LPS-induced COX-2 and PGE₂ levels in RAW 264.7 cells [22], while in the other, nicotine, an α7nAChR agonist, increased the same parameters in microglial cells [23]. Choline and a non-selective COX inhibitor, acetylsalicylic acid, showed potentiated anti-inflammatory effects by decreasing PG and cytokine levels via α7nAChR activation in an acute inflammation model in mice [24]. Our previous study showed that intraperitoneal administrations of CAP-acting agents, choline and CDP-choline, decreased LPS-stimulated COX-2, PGE₂, PGI₂, TXA₂ along with the TNF-α levels in the endotoxemic rats [16]. In the present study, choline administration significantly reduced COX-2 expression and the PGE₂ levels via α7nAChR activation in LPS-activated murine macrophage cells indicating that α7nAChR-mediated activation of CAP has a major role in choline’ effectiveness on COX pathway under inflammatory conditions.

Changes in cytosolic Ca²⁺ levels may affect different cellular sensors including Ca²⁺/calmodulin-dependent protein kinase II, protein kinase C (PKCs), and the p21ras/phosphatidylinositol 3-kinase (PI3K)/Akt pathways, which are involved in NF-κB activity [25]. Activation of Ca²⁺-permeable α7nAChRs and the changes in downstream signalling pathways alters inflammatory response. Studies showed that α7nAChR activation elevates intracellular Ca²⁺ concentration, which leads to secondary Ca²⁺ release from intracellular stores (Ca²⁺-induced Ca²⁺ release, CICR) and activates ERK/MAPK cascade in neurons and astrocytes [26]. Additionally, α7nAChR activation with AChEI donepezil mediated elevation of the phosphorylation of Akt, effector of PI3K, has been demonstrated under inflammatory conditions [27]. On the other hand, Ca²⁺-dependent protein phosphatase, calcineurin, dephosphorylates calcium-dependent transcription factor, nuclear factor of activated T cells (NFAT)c1, then, NFAT and the NFAT:calcineurin complex is rapidly imported to the nucleus where it is transcriptionally active [28]. The level of activated NFAT can be altered by oscillating concentrations of intracellular Ca²⁺ [29]. Dual roles of α7nAChRs proposed by recent studies suggest that G proteins might also be involved in inositol-triphosphate(IP₃)-induced Ca²⁺ release and metabotropic actions of this receptor in immune system cells; which may also account for their immunomodulatory actions [30], [31], [32]. In addition, Ca²⁺ elevation by α7nAChRs may be categorized to couple Ca²⁺ to downregulation of NF-κB by activating Ca²⁺-dependent phosphatases within these restricted subsarcolemmal areas similar to that we proposed earlier for vascular smooth muscle cells [33]. Choline treatment downregulated NF-κB and cytokine levels via α7nAChR [8]. Consistent with previous data, our results showed that intracellular Ca²⁺ concentration was significantly elevated by choline (>3 µM) via α7nAChR which downregulated LPS-induced NF-κB and COX-2 expressions along with that of PGE₂ and TNFα.

KYNA has been shown to activate G-protein-coupled receptor 35 (GPR35) to produce anti-inflammatory actions [11, 34]. Studies demonstrated that under inflammatory conditions, KYNA decreases the levels of cytokines, including TNF, interleukins and HMGB1 in different inflammatory cells [11, 35]. GPR35-mediated action of KYNA has been shown to reduce LPS-induced TNF-α levels in mononuclear cells obtained from human peripheral blood samples [11]. The present study shows that KYNA decreased LPS-induced TNF-α levels leading to decreased COX-2 expression and PGE₂ levels. However, co-administration of equieffective concentrations (10 µM) of KYNA showed only an additive interaction on COX-2 inhibition as this combination did not yield any further downregulation. Induction of PLA₂ and COX-2 activities by pro-inflammatory cytokines produced by macrophages in response to an inflammatory stimulus may counteract the inhibitory effects of choline and KYNA [20, 21]. Therefore, a possible involvement of GPR35 in the anti-inflammatory effects of KYNA on COX pathway should also be investigated further as a decrease in TNF-α levels might mediate interaction between KYNA and choline on COX-2 pathway based on our current observations.

α7nAChR antagonistic property of KYNA has been extensively debated as the data was not reproduced [13, 14]. In another study, KYNA did not antagonize ACh on nicotinic receptors [14]; however, it upregulated central nAChR expression [13]. Consistent with previous findings, our data showed that elevated intracellular Ca²⁺ followed by choline-induced α7nAChRs activation decreased partly in the presence of KYNA (>10 µM), suggesting an interaction between KYNA and choline on α7nAChR only at higher concentrations. The reversal of the antagonistic effect of KYNA on choline-induced Ca²⁺-elevation at the highest KYNA concentration (100 µM) might be due to the mobilization of Ca²⁺ from internal Ca²⁺ stores or to an additional Ca²⁺ influx from extracellular space.

Conclusions

Inflammation comprises complex reactions which require treatment strategies that produce comprehensive effects. This study demonstrates that choline exerts its anti-inflammatory effect on COX pathway via activating α7nAChRs and downregulating NF-κB signaling pathway in LPS-induced inflammation model in murine macrophage cell lines. Additionally, KYNA affects choline’s actions on COX pathway, suggesting a possible interaction of kynurenine and cholinergic pathways under inflammatory conditions. In addition to choline’s effects on different systems affected by inflammatory response via nicotinic receptors, its downregulating effects on COX-2 pathway also increase its potential value as a treatment option. Moreover, this study demonstrated for the first time that the choline supplements may enhance anti-inflammatory effects of endogenous KYNA, possibly affecting neuroinflammatory disorders. Further studies investigating the possible interaction between choline and KYNA are required for the development of novel anti-inflammatory strategies.