Inflammation of adipose tissue (AT) plays an important role in the development of many chronic diseases associated with obesity [1
]. Both animal and clinical investigations suggest that inflammation and dysfunction of AT in obesity, resulting in the aberrant production of inflammatory mediators, are key processes linking obesity to comorbidities, including insulin resistance and type 2 diabetes, cardiovascular and respiratory diseases, osteoarthritis, and cancer [2
Until recently, adipocytes, the main cellular components of AT, were mainly considered as an energy depot, but there is now clear evidence pointing to an active role of adipocytes as endocrine cells producing hormones, growth factors, cytokines, chemokines, collectively called adipo(cyto)kines. These proteins can act in an autocrine, paracrine or endocrine fashion to control various functions, including energy homeostasis and metabolism in the AT itself, the liver, the skeletal muscle and the vasculature [3
]. Altered adipocyte function, as it occurs in obesity and overweight, especially increased intra-abdominal adiposity, is accompanied by tissue remodeling and oxidative stress, and deleteriously changes the expression and production of pro- and anti-inflammatory, metabolic and proliferative factors, which leads to a tissue and systemic low-grade proinflammatory state and the development of obesity comorbidities [3
]. Therefore, targeting adipocyte dysfunction and inflammation is an attractive perspective for the treatment of obesity-related metabolic and vascular diseases. With the increase of AT mass, the production of insulin-sensitizing and anti-inflammatory adipokines, such as adiponectin, drops. Contrarily, enlarged adipocytes start producing adipokines, such as chemokines and cytokines, affecting the AT infiltration of immune cells, mainly monocytes/macrophages, granulocytes and lymphocytes. In particular, a key event in the induction of obesity-induced AT inflammation and development of insulin resistance appears to be the polarization of AT macrophages from an anti-inflammatory M2 phenotype to a pro-inflammatory M1 phenotype [4
]. During obesity, signals mostly from hypertrophic and dysfunctional adipocytes drive M2-to-M1 transition of AT macrophages, which then become a major source of pro-inflammatory factors, such as tumor necrosis factor (TNF)-α and interleukin (IL)-1β, thus amplifying adipocyte dysfunction and recruiting and activating more pro-inflammatory cells in a vicious cycle, and ultimately contributing to local and systemic chronic low-grade inflammation and insulin resistance [1
A main player in inflammation is nuclear factor-κB (NF-κB). NF-κB is a family of inducible transcription factors, which regulate the expression of genes, such as chemokines, adhesion molecules, cytokines, involved in processes of immune and inflammatory responses [5
], and has been implicated in the initiation and progression of metabolic diseases, connecting inflammation and dysmetabolism in the AT [6
]. In resting cells, NF-κB is sequestered in the cytosol bound to Inhibitor of κB (IκB) proteins, which prevents the nuclear localization and transcriptional function of NF-κB. After stimulation with cytokines, microbial factors, fatty acids and other stimuli such as those in obesity, the IκB kinase (IKK) complex containing two catalytic subunits (IKKα and IKKβ) is activated, which triggers the phosphorylation and degradation of IκB, and promotes NF-κB translocation to the nucleus and the transcription of target genes. Either genetic deficiency or chemical inhibition of components of the NF-κB pathway including IKKβ has been found to significantly decrease obesity-induced insulin resistance in humans and mice [7
]. Indeed, stimulation of the NF-κB pathway has been reported to induce insulin resistance through both direct serine phosphorylation of insulin receptor substrate (IRS)-1/-2 mediated by IKKβ as well as the upregulation of pro-inflammatory cytokines, which in turn activate serine kinases such as IKKβ and mitogen-activated protein kinase (MAPK) causing IRS inhibition, and induce other inflammation-related negative regulators of insulin signaling [10
]. Therefore, targeting NF-κB-driven inflammation may translate into beneficial modulatory effects on metabolic diseases.
Other players increasingly recognized in obesity-associated inflammation and comorbidities are microRNAs (miRNAs), small non-coding RNAs that exert regulatory effects on gene expression by degrading complementary mRNA targets and inhibiting translation [11
]. These miRNAs play essential roles in many biological processes, including survival, proliferation, differentiation, apoptosis, metabolism, and immunity. Recently, several miRNAs have emerged as involved in pathways related to obesity such as adipocyte differentiation, adipokine expression, glucose and lipid metabolism, insulin signalling, oxidative stress, and inflammation [12
]. Such miRNAs can be secreted by adipocytes, and transferred by membrane-bound extracellular vesicles, namely exosomes (from intracellular endosomes and typically 30–100 nm in diameter) and microvesicles (from plasma membrane and 0.1–1 μm in diameter) [14
], to neighbouring and distant cells, where they may exert both local and systemic regulatory effects, including immune cell activation, insulin resistance, inflammation and fibrosis [15
]. Both the AT and circulating miRNAs are deregulated in human obesity [16
]. The miRNA profiling in adipocytes and their supernatants in inflammatory conditions has revealed sustained and concordant changes in the expression of a range of miRNAs related to adipogenesis, oxidative stress, the insulin pathway, and inflammation [13
Among AT-related miRNAs, miR-155 [19
], miR-34a [16
], and let-7 [22
] appear to be involved in inflammatory responses and linked to the NF-κB pathway [24
] Consequently, the ability to control NF-κB and its sub-network activation could contribute to modulating pathogenic processes of various inflammatory diseases, also connected to adipocytes inflammation.
Accumulating evidence suggests that extra virgin olive oil (EVOO), an integral ingredient of the Mediterranean diet, exerts health benefits that include the modulation of inflammatory and immune responses, the prevention of cancers and the reduction in the risk of coronary heart disease and metabolic disease [30
]. Contributory roles have been ascribed to its main components including monounsaturated fatty acids (particularly oleic acid) and polyphenols. In particular, polyphenols include phenolic alcohols such as hydroxytyrosol (3,4-dihydroxyphenylethanol), tyrosol (p
-hydroxyphenylethanol), as well as secoiridoids that contain either elenolic acid or elenolic acid derivatives in their molecular structure, such as the aglycone of oleuropein, the aglycone of ligstroside, and their respective decarboxylated dialdehyde derivatives oleocanthal (p
-hydroxyphenylethanolelenolic acid dialdehyde or decarboxymethyl ligstroside aglycone, OC) and oleacein (3,4-dihydroxyphenylethanolelenolic acid dialdehyde, OA) [31
]. EVOO is a rich source of phenolic compounds (40–1000 mg/kg), which not only are strong antioxidants and radical scavengers both in vitro and in vivo, but also possess other potent biological activities, including anti-inflammatory, immunomodulatory, hypotensive, anti-microbial and metabolic effects, that could partially account for the observed health effects of the Mediterranean diet [32
Among these phenolic constituents, two key secoiridoids, OC and OA, are attracting clinical and nutritional interest due to their proved bioactivity. Early research conducted by Beauchamp and colleagues demonstrated that OC inhibits the pro-inflammatory cyclooxygenase (COX) enzymes in a dose-dependent manner, mimicking the anti-inflammatory action exerted by ibuprofen [33
]. Subsequent investigations demonstrated OC efficacy against inflammatory diseases, including joint degenerative disease, neurodegenerative disease and specific cancers [34
]. The oleuropein derivative OA is another recently emerged, potentially healthful polyphenol, exerting anti-inflammatory, antioxidant and vasculoprotective effects [36
Polyphenols from several food sources have been increasingly found to favorably affect AT physiology [38
], and their modes of action include interaction with cell signaling pathways, modulation of transcription factors activity and gene expression, but also the post-transcriptional regulation of genes via modulation of miRNAs expression [39
Data concerning the action of EVOO polyphenols in the context of human obesity-related AT inflammation and dysfunction are scarce, and mainly regard oleuropein and its derivative hydroxytyrosol, demonstrating potential anti-obesity and anti-diabetic effect [42
]. Recently, beneficial properties of in vivo administration of OA have been reported, showing protective effects against weight gain, insulin resistance, liver steatosis, and lipid metabolism in high fat diet-fed mice [47
]. Data on the effect of OC and OA on obesity-associated inflammatory responses in adipocytes are lacking. In the light of their anti-inflammatory action, we hypothesized an effect of OC and OA on adipocyte inflammatory dysfunction as a mechanism underlying the beneficial effects of EVOO in the prevention of cardio-metabolic risk. We here assessed the effect of OC and OA on the expression of inflammation-related genes (transcripts) and miRNAs in human adipocytes, under an obesity-mimicking inflammation induced by the cytokine TNF-α.
This study shows that OC and OA, two secoiridoid polyphenols typically present in EVOO, a basic component of the Mediterranean diet, present anti-inflammatory effects in human adipocytes challenged with TNF-α, a prototypic inflammatory stimulus. TNF-α is overexpressed in AT during obesity, and is causally linked to AT inflammation [70
]. We showed that OC and OA: (1) attenuated TNF-α-induced expression of genes and miRNAs involved in AT inflammation and related adipocyte dysfunction; (2) reduced the activation of NF-κB pathway; and (3) may prevent NF-κB activity by directly interacting with the p65 subunit. These newly discovered effects of OC and OA expand the potential health-promoting properties of these compounds, and could contribute to explaining the cardiometabolic benefit of EVOO consumption.
A dysregulated pattern of adipocyte gene expression is involved in the initiation and progression of AT inflammation associated with obesity and its metabolic and vascular complications. During the course of obesity, AT is characterized by adipocyte hypertrophy, oxidative stress, and subsequent increased angiogenesis, immune cell infiltration, and tissue remodeling. Mechanistically, adipocyte hypertrophy is accompanied by the upregulation of proinflammatory cytokines and chemokines, such as TNF-α, IL-1, IL-6, MCP-1, CXCL-10, M-CSF, angiogenic factors such as VEGF, matrix-degrading and proinflammatory enzymes such as MMPs and COX-2, respectively, and pro-oxidant enzymes, including NADPH oxidase. Conversely, adipocyte hypertrophy is also accompanied by an opposite downregulation of anti-inflammatory and insulin-sensitizing adipokines, such as adiponectin, and impaired antioxidant defense. The increased oxidative stress and altered pattern of secretory products by adipocytes participate in the establishment of an autocrine/paracrine loop, with further adipocyte and macrophage pro-inflammatory activation and derangement, as well as of an “endocrine” signaling, propagating AT dysfunction to distant metabolic and vascular organs [40
Increased production of cytokines and chemokines in obese AT has been implicated in the regulation of monocyte recruitment to the AT and pro-inflammatory activation, that further exacerbates AT inflammation and insulin resistance [71
]. In particular, the obligate role in AT macrophage recruitment and inflammation of MCP-1 [73
], a potent chemoattractant for monocytes/macrophages, as well as M-CSF, a critical regulator of macrophage development and survival [72
], has been demonstrated in obesity in animal models. Furthermore, overproduction of MCP-1 by AT may also exert an endocrine role via the systemic circulation, inducing insulin resistance in the skeletal muscle and the liver [74
], and also negatively impacting on the vascular wall, eventually relating obesity and related cardiovascular diseases [76
]. Other chemokines, including CXCL-10 [77
], play a role in orchestrating monocyte/macrophage chemotaxis.
We here show that OC and OA almost totally prevented the mRNA upregulation of IL-1β, MCP-1, CXCL-10 and, for OC only, M-CSF as well as MCP-1 protein secretion in inflamed adipocytes, thus highlighting the potential for these compounds to blunt adipocyte dysfunction and the associated pro-inflammatory cascade occurring during obesity.
A similar anti-inflammatory effect on gene expression of cytokines, chemokines and other pro-inflammatory markers has been previously found for both compounds in different cell models. OC was able to inhibit gene expression of IL-6, MIP-1α in murine chondrocytes and macrophages challenged with lipopolysaccharide (LPS), and IL-1β, IL-6, TNF-α, GM-CSF, IL-8, CCL3, LCN2, MMP-13, ADAMTS-5 in LPS-stimulated macrophages [78
]. OA has been shown to inhibit TNF-α- and lipopolysaccharide (LPS)-induced MCP-1 and adhesion molecules (ICAM-1, VCAM-1, and E-selectin) mRNA expression in endothelial cells and consequent monocyte adhesion to the endothelium [36
]. It also prevented the expression of the adhesion molecule CD11b/CD18, and the release of elastase, MMP-9, and IL-8 in human neutrophils [80
], and enhanced the anti-inflammatory potential of human macrophages, by upregulating the expression of CD163, as well as IL-10 and heme oxygenase (HO)-1 secretion [81
], with a potential atherosclerotic plaque-stabilizing effect [82
In accordance with the early demonstration of an anti-inflammatory action of OC and OA by inhibiting the activity of the pro-inflammatory COX [33
] and 5-lipoxygenase (LOX) [83
] enzymes in cell-free enzymatic assays, as well as the expression of COX-2 in LPS-challenged human monocytes (with a 96% and 88% inhibition for OC and OA, respectively) [84
] and chondrocytes [79
], we here show that OC and OA were able to abolish COX-2 mRNA upregulation induced by TNF-α in human SGBS adipocytes. This effect could have therapeutic implications for obesity-related inflammation because COX-2-derived prostanoids, mainly prostaglandin E2
), play a significant role in mediating macrophage infiltration and inflammation in the visceral fat and the development of AT and systemic insulin resistance and fatty liver [85
]. From a mechanistic point of view, using specific COX-2 and PGE2
/EP3 receptor signaling inhibitors, the adipocyte COX-2/PGE2
/EP3 receptor-mediated pathway during hypertrophy and hypoxia was involved in the activation of NF-κB and hypoxia-inducible factor (HIF)-1α, with consequent increased production of MCP-1, TNF-α, leptin and reactive oxygen species (ROS), as well as decreased adiponectin expression [86
]. Accordingly, the COX-2 and EP3 mRNA levels were positively correlated with the body mass index (BMI) in human subjects, corroborating animal and in vitro results [86
]. Moreover, COX-2-derived prostanoids have been implicated in MMP induction in vascular and inflammatory cells, thus promoting angiogenesis and tissue remodeling [31
Fat accumulation is a complex process involving adipogenesis, angiogenesis and proteolytic remodeling of the extracellular matrix, where a critical role is exerted by adipocyte expression and release of MMPs, including the gelatinase MMP-2 [87
], as well as the angiogenic growth factor VEGF [88
]. In vitro and in vivo studies suggest that MMP-2 may be a key regulator of adipocyte differentiation and AT expansion, through the degradation of the extracellular matrix and basement membrane components, the activation of latent growth factors, the stimulation of angiogenesis and adipogenesis [87
]. Moreover, VEGF expression in AT, inducible by inflammatory cytokines in both the stromal-vascular fraction and mature adipocytes, could contribute to the vascularization preceding AT growth [88
]. Although metabolic consequences of manipulating VEGF expression in mice are not consistent in the literature, a recent study has found that adipose-specific VEGF repression in mice leads to white AT browning, uncoupling protein(UCP)-1 upregulation accompanied by reduced lipid accumulation in AT, resistance to obesity, and increased insulin sensitivity. VEGF-A, the isoform here measured, binds both the tyrosine kinase receptors KDR (VEGF-R2) and VEGF-R1, expressed mainly on the vascular endothelium. However, only blockade of KDR, and not VEGF-R1, restricts AT expansion and limits diet-induced fat tissue expansion. Although a previous study failed to show KDR expression in murine pre-adipocytes and ruled out an autocrine stimulation of VEGF on adipocyte function, we could detect the mRNA expression of KDR as inducible by TNF-α in human SGBS adipocytes, but the potential downstream effect(s) of VEGF/KDR signaling activation in adipocytes is (are) unknown and should be further investigated. Relevant to these issues, we here demonstrated that OC and OA also strongly attenuated the TNF-α-induced upregulation of MMP-2 as well as VEGF and KDR mRNA levels in human adipocytes. A similar modulatory effect on MMP-2 and VEGF expression has been recently reported for OC in melanoma cells, via the suppression of STAT3 transcriptional pathway.
Obesity development is associated with unbalanced overproduction of ROS [90
], which are implicated in the development of adipocyte inflammation, recruitment of macrophages to AT, and whole-body insulin resistance, [91
]. In particular, a role for NOX, a membrane-bound enzyme complex that produces superoxide by transferring electrons from NAD(P)H to molecular oxygen, has recently emerged [92
]. NOX-4 is the major NOX isoform in cultured murine and human adipocytes, and NOX-4-derived ROS mediate the increased expression of chemotactic factors in cultured adipocytes exposed to nutrient excess [93
]. Diet-induced obesity in mice is accompanied by increased AT NOX-4 activity, the inhibition of which leads to decreased ROS and AT-inflammation [91
]. A role for the NOX-2 isoform of NADPH oxidase has also been inferred, as deficiency of NOX-2 attenuated AT inflammation and insulin resistance in mice fed a high-fat diet [94
]. We here provide the first demonstration of the suppression by OC and OA of the TNF-α-induced upregulation of both NOX-2 and, more strongly, NOX-4 mRNA in human adipocytes. This previously unappreciated effect on pro-oxidant gene expression could complement the antioxidant properties demonstrated for both compounds [95
]. Indeed, OC and OA have been found to inhibit NADPH oxidase activity and intracellular superoxide anion production in human monocytes [84
]. This occurred at a relatively high concentration (100 µmol/L) that, in the case of OC, partially reduced cell viability [84
], at variance from our experimental condition in which the reduced expression of NOX-4 by 25 µmol/L OC or OA was not associated with any effect on adipocyte viability. Furthermore, OA was seen to prevent H2
-induced DNA damage in monocytes [97
], and reduced ROS production and myeloperoxidase release by human neutrophils [80
]. OA was also found to prevent angiotensin II-induced ROS production and senescence in endothelial progenitor cells via the activation of the antioxidant transcription factor nuclear factor-E2-related factor 2 (Nrf2) and the increased expression of HO-1 [98
]. These antioxidant effects of OC and OA are concordant with the ability of polyphenol extracts from EVOO to inhibit the activity of NADPH oxidase and the expression of NOX-4 and NOX-2 in endothelial cells [99
]. Concordantly, EVOO was able to counteract postprandial oxidative stress by reducing NOX-2 activity in healthy subjects [100
]. In agreement with a similar previously reported effect by the EVOO simple polyphenol hydroxytyrosol [31
], we here also show that OC and OA prevented the TNF-α-induced upregulation of antioxidant enzymes, such as GPX and mitochondrial SOD, which play a role in detoxifying superoxide anion and hydrogen peroxide (and organic hydroperoxides), respectively. This modulatory effect suggests the attenuation of cell oxidative stress and a restoration of the basal antioxidant defense in the presence of anti-inflammatory and antioxidant polyphenols.
Although not directly tested in the present study, the modulation of gene expression associated with processes involved in adipocyte dysfunction and inflammation could lead to an improved adipocyte secretory profile, reduced production of ROS, improved metabolic function, as well as decreased chemotaxis and activation of macrophages. Collectively, these data may lay down the basic foundations for further animal and human studies on the cardiometabolic benefit of OC and OA as either isolated compounds or as food ingredients.
To gain further molecular insight into the anti-inflammatory activity of OC and OA, we studied the effect of these polyphenols on the activation of NF-κB, which transcriptionally regulates a plethora of pro-inflammatory genes including those here assessed, and functions as a link between inflammation and dysmetabolism in obesity [6
]. The NF-κB family is a heterogeneous set of transcription factors that dimerize in different combinations at a common dimerization domain called Rel homology domain [102
]. Five mammalian NF-κB proteins have been identified, including p65 (also named as RelA), RelB, c-Rel, p50, and p52, which share a conserved DNA-binding domain. With the exception of p52, these proteins are constitutively present in the cytoplasm, bound to inhibitory proteins, i.e., inhibitor of κB (IκB)-α and IκB-β. Activation of the NF-κB pathway by most stimuli, including TNF-α, involves ROS-sensitive pathways and leads to IκB-α and IκB-β degradation and the consequent translocation of p50/p65 heterodimers into the nuclei, where they bind to promoter regions of genes encoding for effectors of the inflammatory response, such as cytokines, chemokines, adhesion molecules, acute-phase proteins, inducible enzymes (COX-2, inducible nitric oxide synthase, etc.), and proangiogenic growth factors [5
]. NF-κB is an attractive pharmacological and nutritional target in the prevention and treatment of inflammatory diseases, and its inhibition by dietary polyphenols has emerged as a common denominator of most of the anti-inflammatory properties of these compounds [103
Previous studies have reported the suppression of the NF-κB pathway by hydroxytyrosol and oleuropein in several cell models [31
], and by OC in LPS-challenged human chondrocytes [104
]. Concordantly, we here found, for the first time, that OC and OA are able to inhibit NF-κB activation in TNF-α-stimulated human adipocytes, as demonstrated by reduced DNA binding activity of the p65 subunit from OC- and OA-treated nuclear extracts. This may represent a mechanism underlying the general downregulation by these compounds of genes involved in AT inflammation. The interference with NF-κB activation may occur through (a) direct targeting of the DNA binding activity of individual NF-κB proteins; (b) blocking the nuclear translocation of NF-κB by inhibiting the nuclear import system; (c) stabilizing IκBα protein; (d) targeting cytosolic signaling nodes upstream of NF-κB activation, such as IκB kinase (IKK)β [102
]. We cannot rule out the involvement of oxidative stress inhibition by OC and OA in the observed attenuation of NF-κB activation. Moreover, polyphenols might also act by directly interacting with protein targets to inhibit their function [106
]. Previous studies demonstrated that small-molecule ligands can inhibit both the DNA binding activity and nuclear translocation of NF-κB by interacting with a specific region of the p65 subunit encompassing the cysteine residues C38 and C120. Therefore, we applied a molecular modelling protocol including docking studies, MD simulations and binding energy evaluations with the aim of investigating the reliability of a direct interaction between EVOO polyphenols and NF-κB, which could thus represent a potential molecular mechanism at the basis of the activity of these compounds. For these studies, we focused on OC, which is a prototypical secoiridoid structurally similar to—but more active than—OA. Our computational protocol, which evaluated multiple potential binding modes of OC in complex with the human NF-κB p50/p65 heterodimer, suggested that the compound would most likely interact with the p65 subunit of NF-κB, binding to a pocket located in the proximity of C38 and C120, as reported for known NF-κB inhibitors. Moreover, additional computational studies suggested that OC could favorably interact with the catalytic site of IKKβ and act as an IKKβ inhibitor, thus hampering the phosphorylation of IκB proteins and the release of activated NF-κB ready to be translocated to the cell nucleus. Although not formally tested in the present study, we could infer similar NF-κB binding modes by OA. The results of our molecular modeling studies suggest novel possible modes of action at the basis of the anti-inflammatory protective effect of EVOO polyphenols. Further studies will be aimed at addressing a more in-depth evaluation of these as well as other possible mechanisms of action of OC, OA and other EVOO polyphenols.
Although improvements in AT inflammation in obesity can translate into beneficial metabolic effects, cellular and animal studies suggest that dietary polyphenols may be effective in contrasting obesity and its metabolic derangements by directly modulating energy metabolism, lipid/lipoprotein metabolism, glucose homeostasis, and insulin sensitivity, in relevant tissues including the AT and the liver [107
]. We therefore examined the effect of OC and OA on the gene expression of a metabolic effector, i.e., PPARγ, primarily expressed in the AT and intertwined with adipogenesis and insulin sensitivity. We found that OC and OA counteracted the inhibition of PPARγ expression in response to TNF-α. PPARγ, a member of the PPAR nuclear receptor subfamily, is a transcription factor controlling the expression of genes involved in lipid metabolism, glucose homeostasis, insulin signaling, and cellular differentiation [108
]. PPARγ agonists, such as rosiglitazone and pioglitazone, are therapeutic agents for type 2 diabetes [108
]. Besides energy metabolism, another well-established function of PPARγ is the inhibition of inflammatory gene expression through the interference with specific pro-inflammatory transcription factors including NF-κB, by inducing its negative regulator IκBα or interacting with NF-κB and its regulatory proteins leading to NF-κB inactivation (i.e., trans-repression) [109
]. Therefore, PPARγ improvement by OC and OA offers an additional mechanism for their anti-inflammatory action and the inhibition of NF-κB signaling. A similar improvement of PPARγ expression has been previously reported for OC-rich EVOO in the brain of Alzheimer’s disease mice model [110
], and for the EVOO simple phenol hydroxytyrosol, a precursor of OA, in TNF-α-stimulated SGBS cells [45
]. A very recent study documented an OA-mediated down-regulation of PPARγ in murine differentiating 3T3-L1 pre-adipocytes and in vivo in high-fat diet-fed mice, which was accompanied by reduced adipogenesis and weight gain and improved insulin resistance [111
]. Here we provide the first demonstration of PPARγ upregulation by OC and OA in human mature adipocytes under inflammatory condition, with implication for attenuating inflammation and adipocyte dysfunction. In order to further evaluate OC and OA role in obesity and dysmetabolism, our following studies will be aimed at assessing the effects of OA and OC on other major molecular regulators of metabolic functions, including adipogenesis (e.g., C/EBPα, fatty acid binding protein), lipogenesis (e.g., SREBP1c, fatty acid synthase [FAS], acetyl-CoA carboxylase [ACC], etc.), lipolysis (e.g., adipose triglyceride lipase [ATGL] and hormone sensitive lipase [HSL]), insulin signaling and glucose uptake/metabolism (e.g., IRS-1/2, GLUT-4), as well as fatty acid oxidation and energy expenditure (e.g., PPARα, 5′ adenosine monophosphate-activated protein kinase [AMPK], PPARγ coativator [PGC]-1α, uncoupling protein-1/2).
Another notable observation of the present study is the involvement of miRNAs in the molecular anti-inflammatory response observed in adipocytes following OC and OC treatment. Currently, miRNAs have been found to be regulated by some natural compounds. By altering the expression of miRNAs and influencing the downstream signaling pathways or target genes, several natural compounds exhibit their bioactivity in the prevention, diagnosis, therapy, prognosis and drug resistance of human diseases [112
]. The here-tested EVOO polyphenols demonstrated the ability to modulate miRNAs linked to inflammation, i.e., miR-34a, miR-155 and let-7c. Indeed, the expression of these miRNAs is significantly modulated by TNF-α in human adipocytes toward a pro-inflammatory mode, and is closely associated with NF-κB signaling.
Emerging evidence suggests that miR-34a levels are elevated by NF-κB through directly binding to its promoter [26
]. Furthermore, miR-34a directly interacts with sirtuin 1 (Sirt1), which leads to the rapid suppression of NF-κB [29
]. Therefore, it is conceivable the activation of a biological autocrine loop, whereby NF-κB increases miR-34a levels and miR-34a induces the activation of NF-κB.
Further, miR-155 is starkly connected to the inflammatory process and the NF-κB pathway [25
]. A recent article reported that NF-κB-p65 interacts with the promoter site of miR-155, so that NF-κB activation increases levels of miR-155. Moreover, miR-155 may regulate the expression of IκB kinase (IKKβ and IKKε), which leads to the repression or, at least, the limitation of NF-κB activation) [27
]. miR-155 has also been seen to repress B-cell leukemia/lymphoma(BCL) 6, a negative regulator of NF-κB signaling, thus promoting inflammatory pathways in macrophages [113
]. Taken together, these data indicate the existence of a NF-κB/miR-155 pathway [114
]. In our model of TNF-α-stimulated adipocytes, the phenolic compounds OC and OA showed the ability to counteract the increased expression of miR-34a and miR-155, potentially interrupting the pro-inflammatory NF-κB/miR-34a/miR-155 loop. Furthermore, PPARγ is a recognized direct target of both miR-155 [19
] and miR-34a [115
]. Therefore, the downregulation of mir-155 and mir-34a by OC and OA could mediate, at least in part, the here observed recovery of PPARγ expression by both compounds.
Another miRNA tightly connected with the NF-κB pathway is let-7 [24
]. Indeed, NF-κB directly activates Lin28B, a strong inhibitor of let-7 expression at the transcriptional and post-transcriptional levels [24
]. The attenuated NF-κB signaling induced by OC and OA led to increased let-7c levels, thus interfering with the feedback loop. Moreover, increased levels of let7-c are in line with the decreased levels of its direct target, COX-2 (miRTarBase ID: MIRT051845), confirming the anti-inflammation potential of let-7c modulation by OC and OA.
Notably, adipocyte-derived miRNAs and exosome-derived miRNAs isolated from conditioned media showed an overlapping expression pattern. Indeed, miRNAs are also present in the extracellular environment in membrane-covered microvesicles as exosomes [117
]. Recent studies have revealed that microvesicles containing miRNAs can be secreted by adipocytes and act as endocrine and/or paracrine signaling molecules inside recipient cells [12
]. By this way, miRNAs deliver effects at a distance after being vehicled in body fluids [118
]. Therefore, the beneficial modulation by OC and OA of exosomal miR-34a, miR-155 and let-7c suggests the ability for these compounds also to control cell-to-cell communication.
The bioinformatic analysis of biological pathways predicted to be associated with the regulation of miR-34a, miR-155 and let-7c confirms the involvement of pathways belonging to the inflammatory process, including, regarding miR-155 and miR-34a: cytokine-mediated signaling, response to TNF-α pathway, angiogenesis and vascular remodeling, and positive regulation of chemotaxis and NF-κB activity; and regarding let7-c: cell communication, regulation of cell cycle and response to cytokines, and extracellular matrix organization. Since these processes are also profoundly implicated in the AT pathophysiology during obesity development, through the regulation of miR-34a, miR-155 and let-7c, OC and OA could exert an overall blocking effect of inflammatory and dysmetabolic pathways in human adipocytes. Therefore, the miRNAs here investigated may cooperate in the regulation of the observed anti-inflammatory effects of OC and OA in human adipocytes.
An important limitation of the present study is the use of OC and OA parent compounds that may not be the major bioavailable forms in vivo. OC and OA—the latter also deriving from oleuropein degradation—are stable in the acidic condition of the stomach, reach the small intestine in an unchanged form and enter the systemic circulation mainly as glucuronides [119
]. Of note, the inflammatory cells may secrete lysosomal β-glucuronidase into the extracellular space [121
], thus potentially increasing the local concentration of the free forms of polyphenols. A recent study reports that OA is readily absorbed and metabolized to hydroxytyrosol, homovanillic acid, and homovanillyl alcohol after oral administration in rats [122
]. However, metabolism and bioavailability of these compounds, mostly OC, need to be firmly established to substantiate their health effects. Although only three miRNAs genes were here investigated, our results provide preliminary evidence that OC and OA could exert anti-inflammatory efficacy targeting miRNAs in adipocytes, thus paving the way for further studies.
Overall, our findings outline a working model for OC and OA, demonstrating that polyphenols isolated from EVOO, OC and OA, counteract inflammation by attenuating NF-κB activation, the expression of inflammatory cell and exosomal miRNAs, and concomitantly modulate the expression of pro-inflammatory genes toward a protective profile in inflamed adipocytes (Figure 11
). These results provide evidence of unappreciated protective effects and active ingredients mediating the cardio-metabolic benefits of EVOO consumption, suggesting OC and OA as novel dietary tools to limit metabolic inflammation associated with obesity.