Phytochemical Composition, Anti-Inflammatory and ER Stress-Reducing Potential of Sambucus ebulus L. Fruit Extract

Sambucus ebulus L. (SE) fruits are used for their immunostimulation, hematopoietic and antiviral potential. Recently, we focused on analyzing the mechanism underlying SE fruit aqueous extract’s (FAE) immunomodulation and anti-inflammatory activities, with attention to its endoplasmic reticulum (ER) stress-reducing potential. J774A.1 macrophages were treated with SE FAE alone or in conditions of lipopolysaccharides (LPS) stimulation. Using GC–MS and LC–MS/MS, its phytochemical composition was analyzed. To measure transcription and protein levels, we used qPCR and Western blot, respectively. The prevailing phytochemicals in SE FAE were hydroxycinnamic acids, proanthocyanidins and anthocyanins. The content of some amino acids, organic acids, alcohols, fatty acids and esters were newly reported. Extracts exerted an immunostimulation potential by stimulating IL-6, TNFα, Ccl2, COX2 and iNOS transcription, without inducing ER stress. SE FAE suppressed the LPS-induced transcription of inflammation related genes (IL-1β, IL-6, TNFα, Ccl2, Icam-1, Fabp4, COX2, iNOS, Noxo1, IL-1ra, Sirt-1) and reduced the protein levels of iNOS, peIF2α, ATF6α and CHOP. The effects were comparable to that of salicylic acid. SE suppresses LPS-stimulated inflammatory markers on the transcription and translation levels. Targeting ER stress is possibly another mechanism underlying its anti-inflammatory potential. These findings reveal the potential of SE fruits as a beneficial therapeutic of inflammation and ER stress-related pathological conditions.


Introduction
Traditional medicine is a good source of knowledge about therapeutics, which are consequently researched and successfully implicated in modern pharmaceutical preparations. Sambucus ebulus L. (SE), also known as dwarf elder or dwarf elderberry, is a widely used as wound-healing, anti-nociceptive, anti-rheumatoid, anti-influenza, antibacterial and diuretic medicinal plant in Bulgaria, Turkey, Iran, Lebanon, Romania and Bosnia-Herzegovina [1][2][3][4][5]. Fresh fruits, jam, tea or decoction of SE fruits are used as immunostimulating and hematopoietic herbal preparations, as well as for the treatment of rheumatoid arthritis and gastrointestinal disorders [1,2,6]. The number of modern studies focusing on SE biological activities are growing, but there is still insufficient knowledge regarding molecular mechanisms of action of fresh or dry fruits and various fruit extracts.
Only ripe fruits are used in traditional medicine recipes and the chemical content varies depending on the types of the extract [3,7]. Data from phytochemical analyses in Table 1. List of polar phytochemicals identified in the analyzed polar fraction (A) of SE FAE using GC-MS technique. The concentration was given in µg/mL extract. Results are presented as mean ± standard deviation.
Pre-treatment with the SA as a known anti-inflammatory compound, significantly reversed the LPS stimulation of all genes except Fabp4, as follows: IL-1β (31%, p < 0.001); IL-6 (76%, p < 0.001); TNFα (53%, p < 0.01); Ccl2 (32%, p < 0.05); Icam1 (96%, p < 0.05) (Figures 1a-c and 2a,b). The inhibitory effect of SE FAE on LPS-stimulated transcription of pro-inflammatory genes was similar to the effect of the positive control SA. In the case of Icam1, both the extract and the SA reduced the LPS-induced mRNA levels back to normal. When applied in highest concentration (10% v/v) the herbal extract had a reducing effect on the LPS-stimulated gene expression of IL-1β, Ccl2, and of Fabp4, which was even stronger than that of the SA. SE FAE significantly inhibited the LPS-stimulated transcription levels of COX2, iNOS and of Noxo1, by up to 73% (p < 0.05), 93% (p < 0.01) and 78% (p < 0.05), respectively, and the protein levels of iNOS by up to 33% (p < 0.01) (Figure 3a-d). When SA was applied prior to LPS stimulation mRNA levels of COX2, iNOS and Noxo1 were reduced by 85% (p < 0.05), 92.9% (p < 0.01), and by 90.7% (p < 0.05), respectively (Figure 3a-c). The effect shown by SE FAE was similar to that of SA and they both independently reduced LPS-stimulated transcription of iNOS and of Noxo1 back to the normal levels. Pre-treatment with herbal extract showed a stronger iNOS mRNA-and protein levels-reducing effect than the SA did in LPS-challenged cells.
Application of SE FAE suppressed the LPS-induced transcription of IL-1ra by up to 88.95% (p < 0.01) in a dose-dependent manner and that of Sirt-1 by up to 54% (p < 0.05) ( Figure 4). Similar effect was observed in the SA pre-treated cells, where LPS-induced IL-1ra and Sirt-1 mRNA transcription levels were decreased by 46% (p < 0.05) and by 82% (p < 0.01), respectively ( Figure 4). SE FAE exerted stronger reducing activity than that of SA on LPS-stimulated IL-1ra transcription, decreasing it to the normal levels.

Investigation of ER Stress-Related Biomarkers in a Model of LPS-Stimulated J744A.1 Macrophages
Regarding the well-known relationship between inflammation and ER stress, we have analyzed intracellular protein levels of ER stress-related proteins: activating transcription factor 6 alpha (ATF6α), phosphorylated eukaryotic translation initiation factor 2 alpha (peIF2α), and their downstream target gene's product C/EBP homologous protein (CHOP, growth arrest and DNA damage-inducible gene 153 (GADD153)) in a model of LPS-stimulated J744A.1 macrophages ( Figure 5). Cells were pre-treated with increasing concentrations of 2.5%, 5% and 10% v/v (0.25 mg DW/mL, 0.5 mg DW/mL, 1 mg DW/mL respectively) SE FAE or SA for 24 h followed by LPS-stimulation for additional 24 h, and respective control treatments were performed as well.
LPS-stimulated J744A.1 macrophages ( Figure 5). Cells were pre-treated with increasing concentrations of 2.5%, 5% and 10% v/v (0.25 mg DW/mL, 0.5 mg DW/mL, 1 mg DW/mL respectively) SE FAE or SA for 24 h followed by LPS-stimulation for additional 24 h, and respective control treatments were performed as well.   Applied alone, SE FAE in concentration of 2.5% v/v slightly increased peIF2α (by 16%, p < 0.05) (Figure 5a), but also slightly decreased the expression of ATF6α (by 11%, p < 0.05), similarly to the effect of SA (Figure 5b). The higher concentrations of the extract, 5% v/v and 10% v/v, reduced, in a dose-dependent manner CHOP levels by 20% (p < 0.05) and by 34% (p < 0.01), respectively (Figure 5c). The effect of SE FAE on CHOP levels was opposite to the effect of SA in non-stimulated with LPS macrophages. 2.3.3. The effect of SE FAE on ER stress-related biomarkers in LPS-stimulated J744A.1 macrophages.

The Effect of SE FAE on ER Stress-Related Biomarkers in LPS-Stimulated J744A.1 Macrophages
Pre-treatment of macrophages with 10% v/v SE FAE significantly altered the LPSstimulated levels of peIF2α (by 30%, p < 0.01), similarly to the effect of the SA (by 16%, p < 0.05). The same concentration of the extract similarly affected the expression of ATF6α (reduced by 27%, p < 0.05) (Figure 5b). Both, the 5% v/v and 10% v/v of the extract in the culture media significantly reduced in a dose-dependent manner the LPS-stimulated levels of CHOP by 19% (p < 0.05) and by 36%, respectively ( Figure 5c). LPS-stimulated ATF6α and CHOP levels were not reduced by the pre-treatment with SA.
The original western blot gels presenting the changes in protein levels of iNOS, peIF2α, ATF6α and CHOP in J774A.1 mouse macrophages pre-treated with increasing concentrations (2.5%, 5%, 10% v/v) of SE FAE or with SA for 24 h and subsequently stimulated or not with LPS, are given in Figure S12

Newly Detected Phytochemicals in SE Fruit Aqueous Extract
In the recent study we recognized as newly reported 10 AAs including 3 essential ААs, 8 OAs, 12 sugar acids and their phosphates, 16 alcohols, alcohol phosphates and their glycosides, 11 saccharides (mono-, di-, and tri-), 6 saturated and unsaturated fatty acids and 3 of their esters and 3 anthocyanins.

Newly Detected Phytochemicals in SE Fruit Aqueous Extract
In the recent study we recognized as newly reported 10 AAs including 3 essential AAs, 8 OAs, 12 sugar acids and their phosphates, 16 alcohols, alcohol phosphates and their glycosides, 11 saccharides (mono-, di-, and tri-), 6 saturated and unsaturated fatty acids and 3 of their esters and 3 anthocyanins.

Amino Acids
AAs are important for building cellular proteins, nucleotides, for maintaining acidbase balance and as neurotransmitters. Especially the essential AAs are valuable food components for living organisms. Other authors have reported three essential (Val, Leu and Thr) and three non-essential AAs (Ala, Gln and Tyr) in SE extracts [13]. In our study we have identified all of them except Ala. In addition, we have found three newly detected essential AAs (Ile, Phe and Lys) and six non-essential AAs (Pro, Gly, Ser, Asp, Asn, Gln) and Orn. In total, the content of AAs in the tested SE extract was 126.30 µg/mL, where the essential AAs comprise 30% of all AA content. Considering these findings, it could be concluded that SE fruits are a good source of both essential and non-essential AAs.

Organic Acids
The total amount of all detected OAs was 110.66 µg/mL. We identified, for first time, seven OAs, including isocitric acid (16.37%) and succinic acid (11.42%), important substrates for the normal functioning of citric acid cycle and cell energy production. Fumaric and malic acid were previously reported in SE fruit extract [13]. Additionally, other authors reported the presence of citric acid in SE fruit extract, also important for energy metabolism [13]. Pyroglutamic acid (5-oxoproline) was found in the highest content (30.39% of all OA) in our samples. Pyroglutamic acid is very important as an intermediate in γglutamyl cycle involved in transmembrane amino acid transportation and for synthesis of the antioxidant glutathione. We may speculate that the high content of pyroglutamic acid in our samples may be due to a high amount of its keto derivative L-proline (15.84% of AAs) found in SE fruits [39].

Sugar Acids and Alcohols
Regarding the content of sugar acids and sugar alcohols, the presence of pectic acid and sorbitol have been reported in the fruit of S. nigra [40,41]. Data in the literature concerning SE fruits content of sugar acids and sugar alcohols are not available.
In our study, the total amount of sugar alcohols was 268.62 µg/mL. For first time, we identified 17 sugar alcohols and their derivatives: sorbitol and its phosphate form (34.47% of sugar alcohols), glycerol, glycerol-3-phosphate and digalactosylglycerol (22.65%) and arabinitol (12.90%). Summarizing, the newly identified sugar alcohols and their derivatives comprise more than 68% of all quantified sugar alcohols in our samples. Therefore, SE fruits should be considered as a good natural source of sorbitol, glycerol and their derivatives.
The total amount of sugar acids in our samples was 115.49 µg/mL. For the first time, 12 sugar acids were identified; of them, the highest content was found for glucuronic acid (21.98%), galacturonic acid (16.28%), and glyceric acid (14.76%). Glucuronic and galacturonic acids are the most abundant sugar acids in the tested SE fruit extract, as they are the major components of plant polysaccharides, like cellulose and pectin.

Phenolic Compounds
Compared with other species of the Sambucus genus, such as S. nigra, S. cerulea, and S. racemose, S. ebulus is the richest in total hydroxycinnamic acids, catechin, epicatechin and flavonols [15].
Flavanols, including catechin and epicatechin, were previously reported in SE fruits [15,16], as were proanthocyanidin dimers and trimers [55]. SE is the richest among the Sambucus sp. in catechins and epicatechins [15]. Epicatechin is the major proanthocyanidin monomer and a component of proanthocyanidin dimers and trimers. It is considered that one of the richest sources of proanthocyanidins are grape seeds [56]; proanthocyanidin dimer and trimer content in SE fruits is comparable to that in the grape seeds [55].
Resveratrol is the most abundant stilbene in plants. Grape peels are known as one of the best sources of resveratrol, containing on average 0.169 mg/g DW [57]. We found that trans-resveratrol-3-O-glucoside represents 5.19 mg/g DW. Thus, SE fruits and its FAE seems to be considerable sources of resveratrol. Resveratrol exerts a wide range of biological activities. It acts as calorie-restriction mimetic, increasing the levels of so-called survival protein SIRT1 and improves energy metabolism, decreases plasma glucose, triglycerides and inflammatory cytokines [58]. Its positive impacts on human heath are complemented by improved plasma antioxidant activity and reduced oxidative stress [59,60]. In obese individuals, resveratrol improves insulin sensitivity [61] and mitochondrial oxidative capacity when used in combination with epigallocatechin gallate [62].
Quinic acid is a compound conjugating with hydroxycinnamic acids to form their esters. Its presence in SE fruit tea (hot extraction) was reported previously by our group [55]. Recently, we also confirmed its availability in SE FAE (cold extraction).
Hydroxycinnamic acids are the most abundant phenolic acids in fruits, vegetables, and coffee beans [63]. They present as esters of hydroxycarboxylic acids, such as quinic acid or as glycosylated derivatives. Among them are caffeic acid, ferulic acid, chlorogenic acid, isoferulic acid and coumaric acid. There are data showing that SE fruits contain highest concentration of neochlorogenic acid and chlorogenic acid among all Sambucus sp., followed by Sambucus cerulea [15]. The same authors report also presence of p-coumaric acid-Oglucoside, 3-O-p-coumaroylquinic acid, and 4-O-p-coumaroylquinic acid in SE fruits. A study on SE fruit tea confirmed the presence of these hydroxycinnamic acids and those that we have also found in SE FAE [55]. There was no significant difference in hydroxycinnamic acid content between SE fruit tea prepared by hot and cold FAE extraction. In accordance with others, we have also found that the neochlorogenic acid followed by chlorogenic and 3-O-p-coumaroylquinic acid were the main hydroxycinnamic acids in SE FAE. The beneficial effects of hydroxycinnamic acids as potential chemo-preventives are associated to their antioxidant activity [64]. Coumaric and ferulic acid and their amides significantly reduce LPS-stimulated NO synthesis, iNOS protein content and mRNA levels in RAW 264.7 macrophages, thus presuming a mechanisms of their anti-inflammatory activity [65]. Plant extracts rich in neochlorogenic acid possess various biological activities, including antioxidant and anti-inflammatory [66][67][68]. As SE FAE is rich in hydroxycinnamic acids and their derivatives, it could be suggested that hydroxycinnamic acids are the main bioactive components determining its antioxidant, and anti-inflammatory.

SE FAE Modulates mRNA and Protein Levels of Inflammation-Related Biomarkers in LPS-Challenged J774A.1 Macrophages
The anti-inflammatory effect of polyphenols is due to the decreased activation of macrophages and T-lymphocytes and the suppressed production of cytokines and chemokines or their receptors. Polyphenols such as resveratrol, catechin and quercetin, found in SE fruits, inhibit NFκB-dependent production of ICAM and VCAM in endothelial cells, as well as the expression of MCP-1 receptors CCR1 and CCR2 [76,77]. Inhibition of the latter reduces the chemotaxis of leukocytes to the site of inflammation and the subsequent increased production of IL-6. Anthocyanin metabolites reduce TNFα-induced expression of MCP-1 and ICAM, and thus combat oxidative stress. In models of LPS-induced inflammatory response of macrophages, anthocyanidin-and anthocyanin-rich extracts inhibit iNOS transcription and iNOS and COX-2 translation by targeting the NFκB and MAPK kinase signaling pathways [78,79]. Karlsen et al. [80] reported that blackcurrant and blackberry polyphenols significantly inhibited NFκB in LPS challanged monocytes isolated from healthy adults.
In our previous study we found that SE FAE reduces LPS activated mRNA expression of NFκB, which correlated with decreased transcription levels of glutamate-cysteine ligase and glutathione peroxidase enzymes [18]. Neochlorogenic and chlorogenic acid, also found in SE FAE, suppress LPS-stimulated activation of NFκB patway resulting in reduced iNOS synthesis and activation of COX-2, thus decreasing NO, prostaglandin E2, TNFα, IL-1β, and IL-6 levels in RAW 267 macrophages [81]. Neochlorogenic and chlorogenic acid-rich plant extracts significantly reduce the carrageenan-induced paw edema in rats, in vivo [81]. Coumaric and ferulic acids were found to reduce LPS-stimulated iNOS protein and mRNA levels [65]. Few studies have reported the strong antioxidant, anti-inflammatory and antidiabetic properties of cyanidin-3-O-galactoside, one of the main anthocyanin in SE FAE [54]. Ursolic acid, found in SE leaves, reveals anti-inflammatory activity by reducing TNFα-induced expression of ICAM-1 and VCAM-1 in human umbilical vein endothelial cells [82]. Ursolic acid reduces LPS-stimulated NFκB [44] and JNK signaling, thus inhibiting inflammatory cytokine production [45]. It was reported, also, that it may combat ER stressand NFκB-related inflammation in animals on a high-fat diet [46]. We also found an antiinflammatory effect of SE FAE, which may be due to the presence of ursolic acid in SE fruits reported by others [12].
The effect was comparable to that of salicylic acid, a known anti-inflammatory agent, used in our study as a positive control. The possible mechanism behind the observed anti-inflammatory effect of SE extract might due to the presence of neochlorogenic acid, chlorogenicacid, ursolic acid, resveratrol, catechin and quercetin, by suppressing the NFκB signaling pathway. Additional mechanism for reducing COX2 activity and prostaglandin production might be the direct NO . radical-scavenging activity of SE FAE [8].
Moreover, when applied alone, the lower doses of SE FAE induce the transcription of IL-6, TNFα and MCP-1 by two-fold; COX2 and iNOS transcription by three-fold and iNOS protein expression (p < 0.05). These results support the traditional application of dwarf elderberries in folk medicine as an effective immunostimulant. Our previous study reported increased NFκB, glutamate-cysteine ligase and glutathione peroxidase transcription and thus confirms the immunostimulatory effect of SE FAE [18]. Immunostimulatory effect was also proven for S. nigra, another member of the genus Sambucus [90].
NADPH oxidase (NOX), is one of the major enzymes in vascular endothelial cells, catalyzing the formation of a superoxide radical anion [91]. Endothelial eNOS, as well as iNOS, produce NO, which reacts with a superoxide radical anion forming highly reactive ONOO [92,93] and contributing to the development of oxidative stress. NOX is highly active in activated macrophages, taking part in a respiratory burst for destroying bacterial cell walls [94]. NOX is among the newly established target molecules in the treatment of hypertension and atherosclerosis, and concomitant pathologies such as diabetes and cardiovascular diseases [91,95].
By suppressing the LPS-induced gene expression of NOX subunit Noxo1, SE FAE exhibits strong antioxidant and anti-inflammatory activity. The effect of the extract on LPS-induced Noxo1 transcription was similar to that of SA. Both SE FAE and SA are very effective in completely neutralizing LPS-induced Noxo1 overexpression. Compounds such as epigallocatechin gallate, quercetin and isorhamnetin, the derivatives of which are found in SE fruits, were shown to target Noxo1 [96][97][98][99][100][101], while resveratrol decreases NOX activity [102]. NOX, and in particular its Noxo1 subunit, has been suggested as playing an important role in the IL-1β-dependent activation of NF-κB [103]. Therefore, the inhibition of Noxo1 gene expression is one potential mechanism by which SE fruits suppress the NFκB-dependent expression of pro-inflammatory genes and proteins such as iNOS.
FABP4 one of the fatty acid-binding proteins is expressed in both adipocytes and macrophages [104,105]. Macrophages are specific target cells for Fabp4 and its deficiency prevents atherosclerosis [100]. High FFA contribute to the development of atherosclerosis, are proinflammatory and activate TLR4 signaling cascades. The same signaling pathway is also activated by LPS [25]. In animal models of obesity and insulin resistance it was shown that the inhibition of FABP4 protein in macrophages reduces inflammatory cytokines (MCP-1, IL-1β, IL-6 and TNFα) and the formation of atherosclerotic lesions and foam cells and improves insulin sensitivity [106,107]. Recently, we studied the effect of SE FAE on the transcription levels of the Fabp4 gene. SE FAE, applied alone, slightly stimulated Fabp4, as was observed for other studied inflammatory genes. On the other hand, in LPS-stimulated cells, pre-treatment with the same concentrations of SE FAE completely prevents LPS-stimulated Fabp4 transcription. These findings suggest another possible anti-inflammatory mechanism of SE FAE action.
SIRT-1 is among the most studied sirtuins from class III histone deacetylases, which activation improves obesity related insulin resistance [108] and possesses anti-inflammatory potential [25]. Sirt-1 activators, such as resveratrol, may inhibit ICAM1 and TNFα induction [109]. In our study, SE FAE induces the expression of Sirt-1. The same result is observed in macrophages pre-treated with the extract and stimulated by LPS. SIRT-1 decreases serine phosphorylation in IRS-1, improves insulin signaling and, as a consequence, increases glucose transport [110]. The same mechanism is involved in improving insulin sensitivity decreased by TNF-alpha. These findings suggest another possible anti-inflammatory and insulin sensing mechanism of SE FAE, although these mechanisms are unclear and require additional studies in future.

SE FAE Modulates Levels of ER Stress-Related Proteins in LPS-Challenged J774A.1 Macrophages
Activation of ER stress may lead to the phosphorylation of JNK and IKK, which is wellknown to promote NFκB signaling and the consequent inflammation [32] accompanied by JNK-mediated phosphorylation of IRS 1/2 [111] to impair insulin signaling. On the other hand, triggered by viral or bacterial infections, the production of TNFα, IL-6, IL-1β, and INFγ may amplify the ER stress in many cell types including macrophages, pancreatic β cells and hepatocytes [112,113]. Since both the processes of inflammation and ER stress may result from each other, we analyzed the expression of three important ER stress-related proteins as potential mechanisms to explain the observed anti-inflammatory potential of SE FAE.
We have observed a significant increase in protein levels of transcription factors ATF6α and peIF2α and their downstream target CHOP in LPS-stimulated macrophages. Applied alone SE FAE downregulated the synthesis of CHOP at a dose-dependent manner and slightly that of ATF6α. SE FAE significantly reduced LPS-stimulated CHOP levels. The decrease in ATF6α levels and the phosphorylated eIF2α in LPS-stimulated macrophages provides evidence for a possible mechanism by which the extract inhibits CHOP synthesis. This cytoprotective mechanism in the case of stimulated ER stress is confirmed by the established high significant correlation between protein levels of CHOP and transcription factors ( Figure 5). This may explain the previously reported cytoprotective effects of SE fruits as well [9,19]. SA reduced only the LPS-stimulated peIF2α protein levels. It should be noted that the SE FAE effect was in a similar direction, however it was stronger than that of SA. With regard to the well-known anti-inflammatory activities of SA [114] and the links between inflammation and the activation of ER stress, we expected that SA might have reducing effect on ER stress-related biomarkers. However, SA did not exhibit any protective effect against LPS-stimulated ATF6α and CHOP levels, as SE FAE, in contrast, did. According to this observation, we may suggest that the SE FAE uses mechanisms different from those of SA, resulting not only in reduced transcription of inflammatory markers but also in the translation of ER stress-related ones.
ER stress could be activated by high levels of FFAs, similarly to inflammation, excess nutrients, improperly folded proteins and local hypoxia, which is characteristic of obesity. This results in increased oxidative stress in the liver and in adipose tissue of obese animals [115]. An interesting fact is the established activated expression of Fabp4 in macrophages and its association with the development of ER stress and inflammation [106]. In accordance with previous analyses, the suppression of Fabp4 in macrophages protects cells from the FFA-induced inflammatory process, which may result in increased insulin sensitivity and glucose tolerance [106]. The ability of the SE FAE to inhibit LPS-induced transcription of Fabp4 suggests that it would also have a protective effect in combating ER stress. FFA and glucose activate PERK-mediated phosphorylation and activation of eIF2α and RNA splicing of Xbp-1 in obese rat and human adipocytes [116,117]. CHOP is induced predominantly by the PERK/eIF2α/ATF4 signaling cascade associated with ER stress, as well as by the IRE1α/Xbp-1 signaling pathway and the ATF6α transcription factor in different pathological conditions, including diabetes [118][119][120][121].
The induction of CHOP is associated with the activation of apoptosis and DNA damage. Its induction in humans and animal macrophages is associated with the detachment of atherosclerotic plaques in atherosclerosis [122]. The production of superoxide by NOX in atherosclerotic plaque-associated macrophages activates CHOP and subsequent ER stress-mediated cell death [123]. ER stress may stimulate NFκB, by Ca 2+ -and reactive oxygen species-dependent mechanisms [124] and the activation of PERK/eIF2α-mediated phosphorilation of IKK [125]. Another important mediator of the ER stress-related activation of NFκB signaling and the consequent TNFα, IL-6 and IL-1β cytokines production is iNOS [34]. This transforms iNOS enzyme into a cross point of inflammation and ER stress, and, consequently intoa possible therapeutic targets.
By preventing the LPS-induced transcription of iNOS and Noxo1 and the subsequent translation of iNOS protein, SE FAE may reduce superoxide radical and ONOOproduction, thus reducing the activation of ER stress-related inflammation; whereas, suppressing CHOP synthesis by suppression of peIF2 and ATF6α possess another important mechanism for combating ER stress-related activation of inflammation and cytokine production. These are the first results confirming that SE, and in particular SE fruits, could suppress the induction of inflammation by suppressing the activation of ER stress.
The complex phytochemical composition of SE FAE presumes that it acts at different levels on inflammation and ER stress regulatory cascades. SE FAE acts at different crosspoints starting from the direct scavenging of reactive oxygen species, through the regulation of gene transcription, to protein synthesis.
Our results contribute to the relationship between inflammation, ER stress and insulin resistance. We may speculate that SE FAE could have a beneficial effect in preventing atherosclerosis, insulin resistance and diabetes type 2, but this requires additional specific studies to confirm the possible insulin sensing effect of the SE fruits.

Plant Material
Well-ripened fruits of Sambucus ebulus L. were harvested from North-Eastern Bulgaria in the period August-September, 2014 and were dried in the dark at room temperature. SE FAE was prepared using 150 mg finely grounded dry fruits, extracted three times with 3 mL distilled water for 3 min in a vortex mixer (2000 rpm), at room temperature. After centrifugation (5 min, 3500 rpm) the supernatants were collected and diluted up to 15 mL with PBS buffer (pH = 7.4) for cell-culture experiments or with distilled water for phytochemical analyses. A specimen from S. ebulus L. fruits was deposited under No. 108144 in the Herbarium SO (by Index Herbariorum) of Sofia University St. Kliment Ohridski, Faculty of Biology.

Extraction
The samples were filtered through a 0.45-µm PTFE filter (Waters, USA) and the filtrates were loaded onto a reverse phase solid phase extraction column (Discovery ® DSC-18, 5 g, 20 mL) (Sigma-Aldrich Co. LLC, St. Louis, MO, USA). The SPE columns were activated rinsed with 40.0 mL distilled water and 1 mL of the filtered sample was loaded onto the SPE column. Three fractions were isolated: anthocyanin fraction (C)-eluted with 2 × 12 mL 0.1% (v/v) formic acid in acetonitrile; non-anthocyanin fraction (B) containing phenolic acids, flavonols, flavonols, eluted with a 2 × 12-mL ethyl acetate; polar fraction (A), containing organic acids, aminoacids and carbohydrates-eluted with 2 × 12-mL 0.2% (v/v) formic acid in water. The eluates were evaporated to dryness under reduced pressure at a temperature below 40 • C.

Analysis of Fractions B and C
Fractions B and C were analyzed using LC-PDA-ESI-MS/MS chromatographic system; in negative ESI mode for fraction B, and in positive ESI mode for fraction C as previously described [55].
For the analysis of polyphenolics, the dry residues of fraction B and C were dissolved in 200 µL metahnol:formic acid, (99:1 v/v), the solution was filtered through a 0.22 µm PTFE filter and 2 µL of the filtrate were injected into LC-PDA-ESI-MS/MS system. An LTQ Orbitrap mass spectrometer (Thermo Scientific, Hemel Hempstead, UK) equipped with an ESI source (in negative mode) was used for accurate mass measurements. Operation parameters were as follows: source voltage-4 kV; sheath, auxiliary and sweep gas −20, 10 and 2 arbitrary units, respectively; capillary temperature was275 • C. The samples were analyzed in full-scan mode at a resolution of 30,000 at m/z 400 and datadependent MS/MS events were acquired at a resolving power of 15,000. The most intense ions were detected during full-scan MS-activated data-dependent scanning. Ions that were insufficiently intense were analyzed in MS2 mode with a resolution power of 15,000 at m/z 400. An isolation width of 100 amu was used. Precursors were fragmented by a collision-induced dissociation with energy of 30 V and an activation time of 10 ms. The mass range in FTMS mode was from m/z 100 to 1000. The data analyses were performed using XCalibur software v2.0.7 (Thermo Fisher Scientific, Hemel Hempstead, UK).

Identification and Quantitative Analysis
The identification of compounds in fraction A was done either by comparison the retention times and Kovats indexes (RI) of the tested compounds with the same parameters of the corresponding pure standards or with mass spectra from the Golm Metabolome Database (http://csbdb.mpimp-golm.mpg.de/csbdb/gmd/gmd.html, 30 August 2021) and NIST'08 (National Institute of Standards and Technology, Gaithersburg, MD, USA) libraries.
The quantification of phenolics in fractions B and C was performed by the external standard method as previously described [55]. Fifteen phenolic compounds were confirmed by comparing their retention times, exact masses and fragmentation patterns with corresponding standards. The identification of the remaining compounds without available standards was based on accurate mass measurements of the [M − H] − ions and the fragmentation patterns, which was compared with the literature data.

Study Design
The experimental model involved macrophage cells seeded in 6-well plates (2 × 10 5 cells/well) and allowed to adhere overnight. The study design included the following experimental groups: (1) cells treated only with LPS; (2) cells treated with SE FAE; (3) cells pre-treated with SE FAE and consequently challenged with LPS. For control groups, we used untreated cells (blank); salicylic acid-treated cells (positive, antiinflammatory control) and cells pre-treated with salicylic acid and consequently challenged with LPS.
Cells were pre-treated with SE FAE with increasing concentrations of 2.5%, 5% and 10% v/v (0.25 mg DW/mL, 0.5 mg DW/mL, 1 mg DW/mL, respectively) or salicylic acid (100 µM) (Merck, Germany) dissolved in DMEM (with 4.5 g/L glucose, w/o phenol red and L-glutamine) supplemented with 10% heat-inactivated FBS, 100 U/mL penicillin/100 µg/mL streptomycin mixture and 2 mM L-glutamine. After 24 h cells were treated with 200 ng/mL LPS (Escherichia coli 026:B6, Sigma-Aldrich, Taufkirchen, Germany) or not, by the simple refreshing of culture media and incubated for additional 24 h. Following the last incubation period, the cells were lysed and total RNA or total protein were extracted and subjected to subsequent analyses. All treatments were performed in triplicate.

RNA Extraction and cDNA Synthesis
Total RNA was extracted using TRI reagent (Ambion, Waltham, MA, USA) according to the manufacturers' requirement. RevertAid First Strand cDNA Synthesis kit (Ther-moFisher Scientific, Waltham, MA, USA) was used to reversely transcribe 20 ng of total RNA using oligo (dT) 18 priming strategy. Following the manufacturers' protocol reaction conditions in final volumes of 10 µL were provided. cDNA synthesis was performed on GeneAmp PCR 7500 thermal cycler (Applied Biosystems, Waltham, MA, USA). After synthesis cDNA was diluted by adding of 30 µL nuclease-free distilled water to each sample and stored at −80 • C.

qPCR Analysis
Gene transcription levels were analyzed using the qPCR method and performed on an ABI PRISM 7500 (Applied Biosystems, Waltham, MA, USA). KAPA SYBR ®® FAST qPCR Master Mix (2X) with low ROX (KAPA Biosystems, Cape Town, South Africa) was used. The amplification reaction's final volume was 5 µL in 96-well plates, with 0.39 µL of cDNA template. Final concentration of primers' was 300 nM. Reaction conditions were as follows: 95 • C/5 min; 40 cycles at 95 • C/15 sec and 60 • C/1 min. A dissociation step was added to the instrument's protocol to check for nonspecific amplification. As an internal control, the β-actin gene was used. Relative gene expression levels were calculated using the 2 −∆∆Ct method [126]. The used primer sequences (Sigma-Aldrich, Taufkirchen, Germany) for each gene analyzed are presented in Table 3. Expression levels of mRNA are presented as relative units (RU) compared to the untreated control group of cells, where the levels of mRNA expression were considered to be equal to 1. Analyses were performed in triplicate and included all cell groups in one plate while testing the expression of each gene. Table 3. Oligonucleotide sequences of all used qPCR primer sets.

Statistical Analysis
GraphPad Prism v7.0 software (GraphPad Software, Inc.; La Jolla, CA, USA) was used to perform the statistical analyses (Student's t-tests, Spearman correlation, 95% CI). The values of p < 0.05 were considered as significant. Data were presented as mean ± SD (concentration of phytochemical) or ±SEM (mRNA and protein expression levels). All analyses and treatments were performed in triplicates.

Conclusions
The SE FAE is confirmed to be rich in phytochemicals, predominantly hydroxycinnamic acids, anthocyanins, proanthocyanidins and resveratrol, with strong antioxidant-, anti-inflammatory-and ER stress-reducing potential, as well as in AAs including essential ones, organic acids, alcohols and saturated and unsaturated fatty acids and esters, some of them reported for first time in SE fruits.
Considering the results, we may conclude that SE FAE, applied alone, possesses immunostimulating potential, without promoting any additional stress, such as ER stress. It may reduce the ER stress-related expression of the CHOP protein, independently from its immunostimulating potential. The SE fruit extract exerted significant antioxidant and anti-inflammatory action, decreasing the LPS-stimulated transcription of oxidative stress, inflammation, atherosclerosis and insulin resistance-related cytokines, chemokines and enzymes, as well as the translation of iNOS. The herb extract possesses significant ER stressreducing potential, by suppressing the LPS-stimulated synthesis of peIF2α, ATF6α and CHOP proteins. Taken together, these results reveal a new possible mechanism explaining the anti-inflammatory potential of SE fruits, by targeting ER stress and related biomarkers.
These findings are in concordance with the traditional usage of SE fruits and its FAE as potential natural immunomodulation preparation, beneficial in the prevention or treatment of oxidative stress-and inflammation-related conditions.  Figure S12: Original Western blot gels presenting the changes in protein levels of iNOS, peIF2α, ATF6α and CHOP in J774A.1 mouse macrophages pre-treated with increasing concentrations (2.5%, 5%, 10% v/v) of SE FAE or with SA for 24 h and subsequently stimulated or not with LPS, Table S1: Relative Kovat's retention index (RI) of analyzed polar compounds (fraction A) presented in Table 1, using GC-MS technique, Table S2: Precursor ion and fragment ion mass-to-charge ratios (m/z) of the analyzed polyphenols using the LC-PDA-ESI-MS/MS technique.

Data Availability Statement:
The data presented in this study are available in the article.