Docosahexaenoic Acid as Master Regulator of Cellular Antioxidant Defenses: A Systematic Review

Docosahexaenoic acid (DHA) is a polyunsaturated fatty acid that benefits the prevention of chronic diseases. Due to its high unsaturation, DHA is vulnerable to free radical oxidation, resulting in several unfavorable effects, including producing hazardous metabolites. However, in vitro and in vivo investigations suggest that the relationship between the chemical structure of DHA and its susceptibility to oxidation may not be as clear-cut as previously thought. Organisms have developed a balanced system of antioxidants to counteract the overproduction of oxidants, and the nuclear factor erythroid 2-related factor 2 (Nrf2) is the key transcription factor identified for transmitting the inducer signal to the antioxidant response element. Thus, DHA might preserve the cellular redox status promoting the transcriptional regulation of cellular antioxidants through Nrf2 activation. Here, we systematically summarize the research on the possible role of DHA in controlling cellular antioxidant enzymes. After the screening process, 43 records were selected and included in this review. Specifically, 29 studies related to the effects of DHA in cell cultures and 15 studies concerned the effects of consumption or treatment with DHA in animal. Despite DHA’s promising and encouraging effects at modulating the cellular antioxidant response in vitro/in vivo, some differences observed among the reviewed studies may be accounted for by the different experimental conditions adopted, including the time of supplementation/treatment, DHA concentration, and cell culture/tissue model. Moreover, this review offers potential molecular explanations for how DHA controls cellular antioxidant defenses, including involvement of transcription factors and the redox signaling pathway.


Introduction
Docosahexaenoic acid (DHA) is a highly polyunsaturated fatty acid (PUFA) of the n-3 series with 22 carbon atoms and 6 cis double bonds. DHA plays a crucial role in lipid metabolism [1], in membrane structure [2], in cell signaling [3], and in inflammation [4]. Epidemiological studies have demonstrated that diets rich in DHA have a positive effect against several types of disease [5][6][7]. Marine-based fish and fish oil are the most popular and well-known sources of DHA.
Plasma non-esterified DHA derived from chilomicrons and VLDLs enters the cells via passive diffusion or transporters such as fatty acid transport protein or fatty acid transporter CD36 [8]. Inside the cells, non-esterified DHA is converted by acylCoA synthases to DHA-CoAs, which are substrates for β-oxidation, desaturation/elongation and assimilation into complex lipids, i.e., phospholipids in the plasma membrane [9].
The physicochemical properties of the membrane bilayer and the chemical reactivity of the fatty acids that compose the membrane are two inherent traits of the membrane phospholipids that regulate their fluidity and determine their susceptibility to oxidative damage. The first property is related to the fact that oxygen and reactive species are more soluble in the fluid lipid bilayer than in the aqueous solution. Consequently, membrane lipids become primary targets of oxidative damage. The second and more significant property is related to the fact that PUFA residues of phospholipids are extremely sensitive to oxidation [10]. PUFAs are usually oxidized by a well-known mechanism called "free radical oxidation". This theory involves an attack of oxygen at the allylic position with the formation of unsaturated hydroperoxides. These hydroperoxides also take part in the auto-oxidation and thus initiate a chain reaction [11].
Due to its high unsaturation, DHA susceptibility to free radical oxidation may represent the other side of the coin. This uncontrolled oxidation of DHA may have a variety of metabolic and physiological repercussions, such as altering the lipid bilayer's structure and function [12] or producing harmful byproducts such malondialdehyde and alkenals [13]. Therefore, a theoretical concern remains on using DHA for preventing chronic diseases whenever oxidative stress is one of the underlying mechanisms.
The human body implemented several strategies to counteract the effects of excess free radicals based on antioxidant molecules [14]. Endogenous antioxidants, which are products of the body's metabolism, may be enzymatic or non-enzymatic. Enzymatic antioxidants playing an essential role in the first line of defense are superoxide dismutase (SOD), catalase (CAT), glutathione peroxidases (GPx), and peroxiredoxins (PRx) [15,16]. The second line of defense involves non-enzymatic antioxidants such as glutathione (GSH) and thioredoxin (TRx), characterized by the ability to rapidly inactivate ROS and oxidants [17].
This articulated mechanism is regulated at the cellular level through a cis-acting element called antioxidant or electrophile response elements (ARE/EpRE) [18]. Nuclear factor erythroid 2-related factor 2 (Nrf2) is the key transcription factor for transmitting the inducer signal to AREs, and many food bioactive compounds were identified as Nrf2 inducers [18]. As DHA has been shown to modulate transcription of genes related to lipid metabolism, such as stearoyl-Coenzyme A desaturase 2 and 3-hydroxy-3-methylglutaryl-coenzyme A reductase [19,20], by interacting with several nuclear receptors, such as peroxisome proliferator-activated receptor (PPAR) and sterol regulatory element binding protein (SREBP) [21][22][23], it is reasonable to assume that DHA could maintain the cellular redox status promoting the transcriptional regulation of antioxidant expression through Nrf2 activation.
In light of these considerations, we tried to overview studies on the potential effect of DHA in regulating cellular antioxidant enzymes. This review highlights DHA's healthrelated potential and hypothesizes possible molecular scenarios between DHA and Nrf2 in regulating cellular antioxidant defenses.

Methods
This systematic review was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines (PRISMA) [24]. The protocol for this systematic review was registered on INPLASY (INPLASY202360017) and is available in full at inplasy.com (https://inplasy.com/inplasy-2023-6-0017/). The search was carried out by using the PubMed database in December 2022, and was conducted using the following keywords and Boolean operators: "docosahexaenoic acid" OR "DHA" OR "C22:6" AND "antioxidant" NOT "review". The initial search yielded 1063 hits. During the screening process (reviewing titles), 941 records were excluded. After abstract analysis, another 84 articles were ousted. Altogether, 43 records were selected and included in this review. Specifically, 29 studies related to the effects of DHA in cell cultures and 15 studies concerning the effects of consumption or treatment with DHA in animal models were deeply analyzed. One study concerned cell culture and animal models and was allocated to each group. Chosen studies were published between 1998 and 2021 without restriction regarding period or publication status. Exclusion criteria were: (i) titles irrelevant to the research topic; (ii) abstract inappropriate or not related to the research topic; (iii) studies that used n-3 PUFA rich oils which would not allow us to discriminate the effect of DHA from other n-3 PUFAs; (iv) studies that co-administrated DHA with other compounds; (v) studies that used DHA oxidation products to better reflect normal nutritional conditions; and (vi) studies or data with inadequate statistical analysis or inappropriate controls. Reviews, letters, abstracts, and articles without a complete text in the English language were also excluded. Two independent investigators (S.M.B and M.D.N.) checked the titles and abstracts of the studies, and disagreements between the two reviewers were resolved through a mediator (S.I.). Primary outcomes include the most relevant variables to answer the research question (the modulation of cellular antioxidant defenses by DHA), while secondary outcomes include additional variables to help the interpretation of the results of the primary outcomes. The detailed selection process is presented in Figure 1.

Effect of DHA Supplementation on Antioxidant Defenses in Cultured Cells
Although studies on humans remain the gold standard for evaluating the relationship between nutrients and health, the development of reliable in vitro/ex vivo models allow the investigation of the cellular/molecular mechanisms and represents a first-and undoubtedly necessary-step when investigating the health-promoting properties of food components [25]. Table 1 summarizes the data published on the effect of DHA supplementation on antioxidant defenses in cultured cells (Table 1).  The number of available studies is limited to 29 cases. Eight of them used primary cells, while twenty-one relied on cell lines. Primary cells included human fibroblast [28] and peripheral blood mononuclear cells [26], bovine endothelial cells [27], carp brain cells [31], and rat thymocytes [26], hepatocytes [29], hippocampal neurons [30], and astrocytes [32,33]. Among cell culture studies, six were conducted on hepatocytes [38][39][40][41][42][43], four on nervous system cells [44][45][46][47], four in adrenal cells [48][49][50][51], two on pancreatic [52,53] and breast cells [37], and only one study was conducted on ovarian [34], skeletal muscle [35], adipocyte [36], and monocyte cells [54].

Effect of DHA Treatment on Antioxidant Defenses in Animal Models
Again, the DHA treatment spanned a range from 2.5 [59,61] to 1500 mg/d/kg bw [55] or a diet containing DHA from 5 [56] to 10 g/kg diet [65], with a time of somministration ranging from 1 day (single administration) [64,67] to 90 days [62].

Role of DHA in the Intracellular Redox Homeostasis Mechanism
To maintain their energy metabolism, mammalian cells have evolved to use oxygen as a final electron acceptor. Consequently, they must deal with a collection of undesired oxygenated byproducts produced due to these oxygen-dependent metabolic processes. These oxygenated byproducts are collectively referred to as ROS [69]. At low levels, ROS can undergo reactions with biological macromolecules contributing to redox signaling and biological function [69], but-at supraphysiological concentrations-they may undergo aspecific reactions that generate other reactive species with potentially toxic consequences [70].
The susceptibility of fatty acids to oxidation is thought to depend directly on their degree of unsaturation. Reportedly, the oxidation rate of a fatty acid or its esters is typically increased by at least one or two factors for every additional double bond in a fatty acid [71], thus placing DHA in the highest ranks among oxidizable species. However, various in vitro [26,31,40,42,[49][50][51] and in vivo studies [43,57,58,63,64,67,68] suggest that the relation between the chemical structure of DHA and, even from a theoretical standpoint, its vulnerability to ROS oxidation is not as easy to predict.
Most of the selected in vitro and in vivo studies considered in this review reported a general promoting effect at the transcriptional or post-transcriptional level. Given the evidence for the role of DHA in the development of chronic diseases, it appears necessary to assess the molecular mechanism at the basis of the protective role of DHA.
The cap'n'collar basic-region leucine zipper transcription factor Nrf2, encoded by NFE2L2, is a master regulator of intracellular redox homeostasis because, in response to oxidative stress, it orchestrates induction of a battery of genes such as SOD, NAD(P)H quinone dehydrogenase 1, and heme oxygenase-1 that serve to increase the antioxidant and detoxifying capacity of the cell [72]. This tight control of Nrf2 is achieved by a repressor protein Keap1, a cysteine redox-sensitive factor, that, under normal conditions, serves as a Nrf2-specific adaptor protein for the Cullin-3 ubiquitin ligase complex and perpetually targets Nrf2 in the cytoplasm for degradation by the 26S proteasome [73]. Dissociation of Keap1/Nrf2 complex by oxidants leads to transportation and accumulation of Nrf2 to the nucleus where it binds ARE/EpRE sequences in the promotor region of several genes related to phase II drug conjugation, scavenging of H 2 O 2 , and GSH-and Trx-based antioxidant systems [74]. Since oxidative stress is associated with several diseases [75][76][77] and many of these ailments have been demonstrated to be prevented by DHA [78][79][80], it might not surprising that DHA could reversibly activate Nrf2 and promote induction of cellular antioxidant defenses. DHA itself is not a ligand for Nrf2, so one crucial issue that remains to be determined is the endogenous activation underlying the transcriptional responses elicited by DHA.
Under physiological conditions, DHA can be oxidized enzymatically or non-enzymatically. In the enzymatic oxidation pathway, cyclooxygenase and lipooxygenase catalyze the conversion of DHA to produce a large variety of oxidation metabolites, including hydroperoxide and hydroxide positional isomers [81]. In addition, ROS can oxidize DHA through nonenzymatic reactions that release highly electrophilic species, including neuroprostane and hydroperoxide break-down products such as 4-hydroxy-2-hexenal (HHE) [82,83]. As a result, oxidized DHA, or its derived electrophilic species, may react with Keap1 sulfhydryls, altering Keap1 secondary structure which is followed by a loss of association between Keap1 and Cullin-3. This in turn inhibits Nrf2 ubiquitination, leading to stabilization and nuclear translocation of Nrf2 and to the subsequent induction of Nrf2 target genes [84]. In support of this hypothesis, one recent observation indicates that low but significant levels of HHE are generated upon DHA supplementation [85] just before changes in gene expression are observed. Furthermore, 15-deoxy-∆12,14-prostagandin [86], F4 neuroprostanes [83], and HHE [87] have been demonstrated to be activators of Nrf2 and to induce expression of cytoprotective enzymes [88].
After incorporation in the plasma membrane, DHA may profoundly influence cellular membrane composition affecting membrane fluidity, phase behavior, permeability, fusion, flip-flop, and protein function [89]. DHA acyl chains have been shown to affect bilayer properties including lateral pressure, microviscosity, curvature, permeability, elasticity, microdomain formation, and hydrophobic match due its high conformational flexibility arising from the low potential energy barriers to rotation about the single carbon-carbon bonds [90]. Moreover, DHA infiltrates rafts and non-raft membrane microdomains, disrupts raft clustering, and increases the size of rafts [91]. The rearrangement of membrane microdomains may have implications for the raft platform signaling and collocation of the transmembrane protein into or out of rafts. Membrane physical properties are mainly affected by lipid composition, and as previous studies have indicated, the activity of G protein-coupled receptors (GPCRs) located in plasma membranes are influenced by the surrounding fluidic membranes [92,93]. Recently, Yoshida et al. demonstrated, using nanodisc technology to control membrane properties, that increased membrane fluidity shifted the equilibrium toward an active form of the receptor through conformational changes [94]. After activation, GPCR transmits signals by downstream pathways leading to the regulation of physiological processes, including antioxidant response [95]. DHA-containing lipids enhance the function of the prototypical GPCR rhodopsin, which simulation studies have explained takes place as a result of the high conformational flexibility of DHA chains [96]. This provides hybrid lipids with a high affinity for the rough surface of GPCRs, further promoting protein-protein interactions.
PPARs, including α, δ, and γ isoforms, comprise a subfamily of the nuclear receptor superfamily that is highly expressed in mammalian tissues [104]. Each PPAR subtype is located in the cytoplasm [105] and, after activation by specific ligands, enters to the nucleus heterodimerizing with the retinoid X receptor (RXR) before binding to the PPAR responsive element (PPRE) of specific target genes [106]. Natural products, including DHA and its metabolites, serve as endogenous PPARs ligands, which exert adaptive metabolic responses to changes in metabolic status in various tissues [107,108]. Several studies have shown that PPARs ligands can transcriptionally modulate antioxidants such as TRx [109], GPx [110,111], CAT [112], and SOD [21] due the presence of a PPRE in the promotor regions of the coding gene sequence [113,114]. In addition, evidence indicates that PPARs may be phosphorylated by several kinases such as protein kinase A, MEK/ERK, and p38 kinase [115][116][117][118], all of which affect the A/B domain of the receptor and modulate its ligand-independent AF-1 transactivating function [119]. Several studies also strongly support a reciprocal regulation of the Nrf2 and PPARγ pathways to reinforce the expression of one another [120][121][122][123]. In this sense, Nrf2 and PPARγ pathways seem to be connected by a positive collaborative feedback loop, which maintains the expression of both transcription factors and their target antioxidant genes in a simultaneous manner [124].

Conclusions
Considering the results obtained in the present review and the extensive search of relevant information available in the literature, we have proposed a scheme to give a logical explanation regarding the potential mechanisms of action of DHA in the modulation of cellular antioxidant defenses ( Figure 2). Despite DHA's promising and encouraging effects at modulating the cellular antioxidant response, differences observed among the reviewed studies may be accounted for by the different experimental conditions adopted. In fact, unlike the genetically predetermined protein profile, the diet profoundly influences the acyl composition, and several studies have shown a time-and concentration-dependent effect on incorporating DHA into cellular lipids [125,126]. In this line, a very recent paper conducted in football players has demonstrated a dose-response incorporation of DHA into red blood cell membranes up to 6 g·d −1 , which can be used to rapidly achieve a desired omega-3 index (>8%) in only 8 weeks [127].
In addition, although various cells readily take up DHA, its accumulation is organspecific, with a higher content in the brain, liver, and heart respecting plasma, pancreas, and erythrocytes [128,129]. At a cellular level, the uptake of long chain fatty acids is mainly regulated by the fatty acid transporter CD36, a transmembrane glycoprotein highly expressed in tissue with high fatty acid uptake [130] with a pivotal role in cellular lipid homeostasis [131].
Moreover, DHA bioavailability depends on the chemical form it conveys. Recent evidence indicates that DHA esterified in phospholipid and triglyceride is more readily absorbed by the body than in ethyl ester form [132]. Taken together, discrepancies in these terms in the studies selected in this review may have determined substantial differences in DHA accumulation and should be deeply considered in future studies to evaluate the minimum effective treatment times and concentrations of DHA according to the cellular models adopted. In any case, further studies including pharmacokinetic/pharmacodynamic modeling and human trials which consider not only DHA in particular but n-3 PUFA in general should be taken into deep consideration.
To conclude, the identification of molecular mechanisms involved in redox metabolism is an important issue for the development of therapies for chronic disorders, and although further studies are needed, the comprehensive vision offered here may help to address future studies toward specific pathways and to provide molecular-based support to any recommendation pertaining the food/health relationship.