Normolipidemic vs. dyslipidemic subjects
Gene expression profiles of normo- and dyslipidemic subjects differed substantially after FO supplementation. In the dyslipidemic study group, considerably more genes were regulated after FO supplementation, whereas genes were mainly down-regulated. Although the reason for this finding cannot be clarified here, it might be partly explained by the interrelation between inflammation and dyslipidemia . Several studies have demonstrated that a dyslipidemic – in particular hypertriglyceridemic – state is accompanied by an induction of pro-inflammatory pathways [27, 28], while n-3 PUFAs are known to suppress these pathways . Indeed, a number of pro-inflammatory genes were down-regulated in dyslipidemic subjects after FO supplementation, including members of the TNFα pathway. However, besides the regulated genes involved in inflammatory pathways, it is apparent that a large number of other genes were regulated in dyslipidemic subjects in response to FO treatment, suggesting a strong regulative effect of FO in people with dyslipidemia. It needs to be clarified in future studies if some of these genes are associated with the known beneficial effects FO or n-3 PUFAs in dyslipidemia.
In contrast to dyslipidemic subjects, a higher number of genes were up-regulated in normolipidemic subjects in response to FO supplementation. This finding corresponds with the outcomes of two intervention studies investigating the effect of FO supplementation on whole genome expression in healthy subjects [16, 17]. The study of Gorjão and co-workers was conducted with ten men aged between 25 and 45 years supplemented with 1.62 g DHA and 0.78 g EPA per day over a period of eight weeks. The design of this study is comparable with our study, except for the sample material used for RNA isolation (fractionated lymphocytes vs. whole blood). Different RNA sources could yield different expression patterns, and cell fractioning itself could alter the gene expression pattern . Nevertheless, both studies provided similar results. Gorjão and co-workers discovered 71 up- and 6 down-regulated genes by macroarray analysis. Our normolipidemic FO group showed 627 total regulated genes, 383 of which were up-regulated and 244 down-regulated. Therefore, both studies indicate that genes in normolipidemic subjects are mainly up-regulated after FO supplementation. In the study by Bouwens and colleagues, a total of 111 men and woman aged between 66 and 80 years were supplemented with 1.09 g EPA and 0.85 g DHA per day over 26 weeks. There were great differences in the study design compared to our study with regard to gender (men and woman vs. men), subject age (older vs. middle aged), daily n-3 PUFAs intake (1.94 g vs. 2.7 g/d EPA + DHA), and sample material used for RNA isolation (peripheral blood mononuclear cells vs. whole blood), which makes it difficult to compare the results. However, the number of total regulated genes in both studies was in the same order of magnitude. Bouwens and co-workers identified a total of 1040 regulated genes, with 537 up-regulated and 503 down-regulated genes.
Fish oil vs. Corn oil
CO serves as placebo in many FO supplementation studies based on the assumption that CO is inert without biological effects. However, CO is rich in the n-6 PUFA LA, a precursor for AA. Thus, CO similarly contains FAs, capable of directly regulating gene expression or serving as precursors for other bioactive lipid mediators which regulate gene expression. It was, therefore, our intention to enlighten the possible regulative effects of CO on gene expression. The administration of six CO capsules provided three grams of LA per day, which is less than a quarter of the usual dietary intake of LA, and considerably less than a high consumption of vegetable oil. In view of this small amount and the low conversion rate of LA to AA, we assumed that the effect of LA on gene expression would be infinitesimal. Surprisingly, our data revealed that the administration of even small amounts of LA caused changes in gene expression patterns. However, during the evaluation of the study data, we focused on the specific effects of FO on gene expression.
Our data demonstrated substantial differences in gene expression regulation between FO and CO supplementation. Pathway analysis showed that considerably more pathways were dominantly regulated in the FO groups compared to the CO groups (in both normolipidemic and dyslipidemic subjects). Additionally, significantly more genes in these pathways and metabolisms were shown to be regulated in dyslipidemic subjects after FO supplementation compared to normolipidemic subjects, suggesting a more pronounced regulative potential of FO in dyslipidemic subjects. Metabolisms that were dominantly affected by FO include, for example, the immune system, lipid metabolism and CVD.
Immune system metabolism
Both FO and CO administration resulted in the regulation of genes associated with the immune system and infectious diseases, which is expected in view of the RNA source used. RNA was isolated from whole blood, including leukocytes, which are mainly involved in the immune system and pathogen defense. Therefore, it is not surprising that a great number of regulated genes are involved in immune system related pathways. However, comparing the effect of FO and CO supplementation on the expression of immune system related genes, it appears that FO regulates significantly more genes in specific pathways compared to CO, indicating the immune-modulatory capability of FO and its bioactive FAs, EPA and DHA. Generally, EPA and DHA have anti-inflammatory properties by increasing the production of potent bioactive lipid mediators (protecting and resolving), and inhibiting the formation of n-6 FA-derived pro-inflammatory eicosanoids (e.g. prostaglandine E2 and Leukotriene B4) . Together, these effects directly or indirectly suppress the activity of nuclear transcription factors, such as nuclear factor kappa b, which controls the expression of a variety of pro-inflammatory and pro-atherogenic genes, including those encoding for interleukin (IL)-1, IL-6, IL-8, tumor necrosis factor alpha (TNFα), E-selecting, vascular cell adhesion molecule-1, and cyclooxygenase-2 [11, 30].
With a few exceptions, our data revealed that none of the key regulators mentioned above were regulated after FO supplementation, either in dyslipidemic or in normolipidemic subjects (IL-8 was down-regulated after one week and twelve weeks in dyslipidemic FO-supplemented subjects, as well as IL-1 receptor, type II after one week). However, a number of inflammation and immune response related genes were regulated after twelve weeks of FO supplementation, especially in dyslipidemic subjects. The ratios demonstrate that more pro-inflammatory genes were down-regulated than up-regulated, whereas the rate of up- and down-regulation was balanced for anti-inflammatory genes. Moreover, pathway analyses showed that genes of the complement and coagulation cascades (complement component [3b/4b] receptor 1, coagulation factor III [thromboplastin, tissue factor], fibrinogen gamma chain) were mainly down-regulated in dyslipidemic subjects supplemented with FO. Since the blood coagulation cascade is evolutionarily closely related to the innate immune response, its pathways are assigned to the immune system. The coagulation pathway is essential for clot formation and the prevention of excessive bleeding. A dysregulation of the cascade activities can result in clinical manifestations of several diseases with critical thrombotic and/or inflammatory complications . Our results underline the well-known effect of FO to diminish the coagulant activity by the lowering of several coagulation factors and by reducing the capability of plasma to support thrombin generation, especially in patients with hyperlipidemia .
As expected, pathway analyses revealed that the effect of FO supplementation on the expression of genes involved in lipid metabolism was more pronounced compared to CO. Similarly, the regulatory effect of FO supplementation was markedly stronger among dyslipidemic subjects compared to normolipidemic subjects relating to the number of regulated genes. The liver plays a central role in lipid metabolism and n-3 PUFAs have been shown to regulate hepatic gene expression by targeting several transcriptional regulatory networks : For example, n-3 PUFAs regulate several inflammation molecules, including serum amyloid A, TNFα and IL-6 . These inflammation mediators modulate the expression of many lipid metabolism-related genes, for example, by suppressing the expression of perilipin, sterol regulatory element binding protein-1 and lipoprotein lipase. Together, these regulatory pathways result in induced lipolysis and reduced lipogenesis . Besides the down-regulation of FA synthesis gene expression, n-3 PUFAs up-regulate gene expression involved in FA oxidation, which is triggered by an activation of the transcription factor peroxisome proliferator activated receptors alpha (PPARα) .
Again, none of the lipid metabolism-related genes mentioned above were regulated in the present study after 12 weeks of supplementation, probably because the compartment examined (whole blood) is inappropriate to reflect the transcriptional profile of the liver. However, FO supplementation provoked a regulation of several lipid metabolism-associated pathways in dyslipidemic subjects in contrast to normolipidemic subjects, where only a few genes were regulated. This finding emphasizes the regulating effect of n-3 PUFAs on the lipid metabolism in dyslipidemic conditions. Genes, for example, in pathways related to FA metabolism and FA elongation in mitochondria, including genes coding several enzymes, were mainly down-regulated in FO supplemented dyslipidemic subjects. These enzymes (several dehydrogenises and an isomerase) are mainly promoters of mitochondrial FA oxidation. This down-regulation is in contrast to studies, which showed an increased peroxisomal FA oxidation in rats in response to FO administration, while the effect of FO on mitochondrial ß-oxidation was inconsistent . The observed reduced FA oxidation after 12 weeks of FO treatment in this study may be the result of increasing levels of oxidized n-3 PUFAs, which are prone to oxidation. Subsequently, the resulting oxidative stress may lead to an induction of antioxidative mechanisms, which in turn reduce oxidation. The finding that PPARα was up-regulated one week after FO capsule ingestion in dyslipidemic subjects (data not shown) support this hypothesis.
We observed a down-regulation of the gene coding for acetyl-CoA carboxylase beta, which is one of the key enzymes in FA biosynthesis, indicating a reduced lipogenesis. This finding may partly explain the reduced TG levels observed in this study. In the same group, several genes in pathways involved in FA metabolism were regulated. Interestingly, genes coding for Phospholipase A2, group IIE (PLA2G2E), a member of the secreted Phospholipase A2 [sPLA2] family) and Phospholipase B1 (PLB1) were up-regulated. The regulatory functions of PLA2G2E have not been completely uncovered. It is known that PLA2G2E catalyzes the hydrolysis of the 2-acyl groups in 3-sn-phosphoglycerides in membranes and could promote inflammation, since the release of AA is the first step in the AA breakdown pathway, which can be metabolized to several inflammatory and thrombogenic eicosanoids (prostaglandines and Leukotrienes) by the activity of cyclooxygenase and lipoxygenase . The up-regulation of PLA2G2E observed could also point to an intensified exchange of membrane bound AA in favor of EPA and DHA, which is indeed reflected in the decrease in AA levels and concomitant increase in EPA and DHA levels in RBC membranes observed.
In addition, an up-regulation of several genes related to glycerolipid metabolism was observed in FO supplemented dyslipidemic subjects. Glycerolipids are essential components of membranes and an up-regulation of the glycerolipid metabolism indicates remodeling activities of the membrane. Finally, several genes of the glycerophospholipid metabolism were up-regulated in FO supplemented dyslipidemic subjects. Glycerophospholipids, also referred to as phospholipids, are key components of the lipid bilayer of biological membranes and constitute the binding site for EPA and DHA, which are integrated in the membrane. An up-regulation of the glycerophospholipid metabolism results in an increased de novo phospholipid biosynthesis, enabling the incorporation of EPA and DHA into the membrane .
Pathway analysis showed that FO supplementation induced the regulation of pathways involved in specific CVD related metabolisms, especially among dyslipidemic subjects. Several genes from pathways involved in arrhythmogenic right ventricular cardiomyopathy (ARVC) and dilated cardiomyopathy (DCM) were mainly down-regulated. ARVC and DCM belong to cardiomyopathies, a group of diseases that primarily affect the myocardium. While ARVC is characterized by a fibro-fatty replacement of right ventricle myocardium, DCM is a myocardial disease with dilated left ventricle myocardium impairing the systolic pump function of the heart. ARVC and DCM may cause ventricular tachyarrhythmias, blood clots or sudden death. Although the aetiopathogenesis, including the role of dyslipidemia in ARVC and DCM, is largely unknown , inflammatory processes are likely to be involved [38, 39]. The down-regulating effect of FO on genes involved in ARVC and DCM pathways observed may be the result of diverse regulatory effects on lipid metabolism and anti-inflammatory processes. However, the effect of n-3 PUFA supplementation on the pathogenesis of ARVC and DCM in humans is unknown. An animal study showed that FO supplementation reduces arrhythmia in boxers (Canis lupus) with ARVC . In a recent intervention study with non-ischemic DCM patients, it was shown that n-3 PUFA treatment increased left ventricle systolic function and functional capacity .
Strengths and limitations
Strengths: The methodological approach of this study was carefully elaborated. The use of whole blood for RNA isolation is advantageous in view of the easy sample collection and the prevention of altered gene expression patterns which emerge during cell fractionation steps . In addition, the pooling of RNA samples reduces inter-individual variation, enabling one to focus on the characteristics of a population in contrast to an individual level . Moreover, only men were enrolled in the study population. Women are subjected to several hormonal changes which involve individual gene expression changes and hamper the attribution of observed effects to treatment.
Limitations: The study has a number of potential limitations, for example, the small sample size. Moreover, nine subject samples had to be excluded from microarray analysis. To minimize the already high effort for the participants, it was desisted from obtaining multiple baseline samples, which is recommended due to heavy fluctuations in lipid – especially TG – levels. The effect of FO on gene expression was compared to CO, which is often used in n-3 PUFA supplementation studies as a placebo control. It would have been instructive to examine the gene expression of a third untreated study group. However, this additional expenditure would have gone beyond the scope of our study.