n-3 PUFAs improve erythrocyte fatty acid profile in patients with small AAA: a randomized controlled trial[S]

Abdominal aortic aneurysm (AAA) is an important cause of death in older adults, which has no current drug therapy. Inflammation and abnormal redox status are believed to be key pathogenic mechanisms for AAA. In light of evidence correlating inflammation with aberrant fatty acid profiles, this study compared erythrocyte fatty acid content in 43 AAA patients (diameter 3.0–4.5 cm) and 52 healthy controls. In addition, the effect of omega-3 PUFA (n-3 PUFA) supplementation on erythrocyte fatty acid content was examined in a cohort of 30 AAA patients as part of a 12 week randomized placebo-controlled clinical trial. Blood analyses identified associations between AAA and decreased linoleic acid (LA), and AAA and increased Δ6-desaturase activity and biosynthesis of arachidonic acid (AA) from LA. Omega-3 PUFA supplementation (1.5 g DHA + 0.3 g EPA/day) decreased red blood cell distribution width (14.8 ± 0.4% to 13.8 ± 0.2%; P = 0.003) and levels of pro-inflammatory n-6 PUFAs (AA, 12.46 ± 0.23% to 10.14 ± 0.3%, P < 0.001; adrenic acid, 2.12 ± 0.13% to 1.23 ± 0.09%; P < 0.001). In addition, Δ-4 desaturase activity increased (DHA/docosapentaenoic acid ratio, 1.85 ± 0.14 to 3.93 ± 0.17; P < 0.001) and elongase 2/5 activity decreased (adrenic acid/AA ratio, 0.17 ± 0.01 to 0.12 ± 0.01; P < 0.01) following supplementation. The findings suggest that n-3 PUFAs improve fatty acid profiles and ameliorate factors associated with inflammation in AAA patients.

Omega-3 fatty acid supplementation in small AAA 1155 blood status and the development or growth of AAA (12). Omega-3 PUFA blood status is typically evaluated by omega-3 index measurement (13). The omega-3 index represents the ratio of EPA + DHA to all other fatty acids incorporated in red blood cell membrane phospholipids, and substantial published research suggests that this measurement is a reliable predictor of cardiovascular event risk (14)(15)(16)(17). In light of this and evidence correlating specific disease states with distinct fatty acid profiles (18)(19)(20), the present study was designed to compare full blood count data, baseline omega-3 index values, and fatty acid profiles of erythrocytes obtained from AAA patients with those of a healthy control cohort as part of an ongoing clinical trial. In addition, the impact of physiologically appropriate n-3 PUFA supplementation on full blood count values, the omega-3 index, and erythrocyte membrane fatty acid composition was examined in a cohort of AAA patients as part of a 12 week randomized double-blind placebo-controlled clinical trial. It was hypothesized that comparative analysis of erythrocyte fatty acid composition would yield a distinct AAA fatty acid profile that could direct future progress toward identification of novel therapeutic targets.

Observational (case-control) study
AAA patients were recruited from the Sunshine Coast University Hospital and a private clinic (Sunshine Vascular) and healthy control participants were recruited from the general population of the Sunshine Coast, Queensland, Australia with approval from the University of the Sunshine Coast (A/13/473 and A/16/833) and the Prince Charles Hospital Human Research Ethics Committees (HREC/16/QPCH/114 and HREC/12/QPCH/13). The study was conducted in accordance with the Declaration of Helsinki (1964). The patient group included 43 men with small AAA (diameter 3.0-4.5 cm), and the control group included 52 men without a documented AAA. It has previously been reported that there are significant differences in omega-3 index between men and women (21), and AAA is known to be much more common in men than women (22). In view of this, only men were included in the current study. Written informed consent was obtained for each participant. Maximum AAA diameter was measured by ultrasound prior to study entry. Exclusion criteria included age below 60 years or above 86 years, BMI above 39 kg·m 2 , uncontrolled hypertension, cardiac arrhythmia, heart failure, symptomatic aortic stenosis, chronic obstructive pulmonary disease, chronic inflammatory disease, and regular use of prescription anti-inflammatory medication. A family history of AAA or known aneurysmal disease served as additional exclusion criteria for control participants. All participants refrained from nonprescribed anti-inflammatory medications 72 h prior to blood collection and abstained from alcohol and caffeine for the 12 h leading up to their study visit.

Omega-3 clinical trial
The impact of n-3 PUFA supplementation on erythrocyte fatty acid composition was investigated in AAA patients as part of a parallel-design double-blind placebo-controlled trial (ANZCTR12616000483459), carried out between June 2017 and June 2018. Patients with small AAA (3.0-4.9 cm) (n = 32) were recruited from Nambour General Hospital, Sunshine Coast University Hospital, and a private clinic (Sunshine Vascular) and randomized ( Fig. 1) to receive either active fatty acids (three Blackmores Omega Brain capsules each containing 500 mg DHA and 100 mg EPA and delivering a total of 1.8 g n-3 PUFAs per day) or placebo fatty acids (three capsules each containing 490 mg corn oil, 490 mg olive oil, and 20 mg fish oil) for a period of 12 weeks. Capsules were of similar appearance and flavor. Participants were supplied with capsules at day 0, week 3, and week 8, with instruction to take three each morning. The final study visit was at week 12. Substantial research literature suggests that the intervention dose and duration selected for the clinical trial will be sufficient to raise the omega-3 index of supplemented subjects to protective levels associated with positive cardiovascular outcomes (23,24). Exclusion criteria for the clinical trial were: age below 55 years or above 86 years, consumption of three or more fish meals per week, the use of fish oil or krill oil supplements, and the use of anti-inflammatory medications. Written informed consent was obtained from each participant and information regarding medical history, physical activity, and dietary intake of n-3 PUFAs was collected. The protocol was approved by the University of the Sunshine Coast (Ethics approval number A/16/833) and the Prince Charles Hospital Human Research Ethics Committees (HREC/16/QPCH/114). The study was conducted in accordance with the Declaration of Helsinki (1964).

Randomization and evaluation of compliance
An Excel block randomization algorithm was used to assign participants to either an active or a placebo treatment protocol. Random selection of block size 4 or 6 during the computerized sort avoided selection bias (25). Sequentially numbered containers were used to implement the random allocation of capsules. The randomization allocation sequence was generated by the corresponding author. Compliance with treatment protocols was evaluated by monitoring capsule returns and by GC-MS measurement of red blood cell omega-3 index.

Blood sample collection
Fasting blood samples were collected from AAA patients and healthy control participants into EDTA-containing tubes and serum separator tubes. Erythrocytes from EDTA tubes were sedimented by centrifugation (1,500 g, 15 min, 15°C); the plasma and buffy coat fractions were removed and the packed red blood cells were stored at 80°C until analysis. Blood collected into serum separator tubes was allowed to clot at 22°C for 30 min prior to centrifugation (1,500 g, 15 min, 15°C). The serum was collected and centrifuged (4,000 g, 5 min, 4°C) and the supernatant was stored at 80°C until processing and analysis. For the omega-3 clinical trial, fasted blood samples were collected from AAA patients at day 0 (prior to initiation of the study), at day 21 (following 3 weeks of supplementation), and at day 84 (following 12 weeks of supplementation). Erythrocytes from EDTA tubes were processed and stored as described above.

Erythrocyte fatty acid analysis
Choice of erythrocytes as the preferred matrix for assessment of n-3 PUFA status was based on evidence suggesting that: 1) measurement of erythrocyte fatty acid levels assesses long-term dietary fatty acid intake; 2) erythrocyte fatty acid levels are less sensitive than plasma to day-to-day n-3 PUFA intake; 3) erythrocyte n-3 PUFA levels display one-fourth of the within-person variability that occurs with plasma; and 4) erythrocyte n-3 PUFA levels are highly correlated with those in a variety of tissues (26). Erythrocyte fatty acid composition was determined using GC-MS as previously described (27). Briefly, a 600 l aliquot of methanol containing butylated hydroxytoluene (BHT; 20 mg/100 ml) as an antioxidant was added to erythrocyte samples (300 l). The cells were homogenized with glass rods (1 min), flushed with nitrogen gas, and incubated on ice for 30 min. A 600 l aliquot of chloroform was added to the suspension and cells were homogenized, flushed, and incubated as before. The preparation was centrifuged (3,000 g, 4°C, 5 min), and the supernatant was withdrawn, flushed with nitrogen gas, and stored on ice. This procedure was repeated twice with 300 l volumes of methanol/BHT and chloroform and incubation periods of 10 min. In a final extraction step, 1 ml of pooled lipid supernatant was combined with chloroform (800 l) and KCl (0.05 M; 460 l), the solution was mixed by vortex, flushed with nitrogen gas, and centrifuged (3,000 g, 4°C, 10 min). The supernatant was discarded and the fraction containing lipids was dried under nitrogen gas. The extracted lipids were hydrolyzed in the presence of 500 l of 9 M HCl:water:acetonitrile (1:1:18 containing 25 mg/50 ml BHT) during overnight incubation at 65°C. The hydrolyzed samples were dried under nitrogen gas and, following a 10 min incubation at 80°C, were freeze-dried (Thermo Savant Modulyo freeze dryer system, Thermo Fisher Scientific) for 30 min. Samples were derivatized in 250 l of hexane containing 10 l 1-tert-butyldimethylsilylimidazole during a 2 h incubation at 37°C. Fatty acids were separated and analyzed with a Perkin Elmer Clarus 580 gas chromatograph coupled to a Perkin Elmer Clarus SQ85 mass spectrometer using a Perkin Elmer Elite 5MS column (30 m × 0.25 mm internal diameter × 0.25 mm film). The GC method included a split ratio of 30:1 with injector at 300°C and temperature programming of 170°C (initial) ramped to 310°C at a rate of 6°C/ min followed by a 5 min hold period. The MS method included ionization at 70 eV with a scan range of m/z 45-450 for 4.0-28.3 min. Peaks were identified by comparison with known standards or the National Institute of Standards and Technology Library (NIST 2008 Library). Omega-3 indices were expressed as the combined percentage of DHA and EPA integrated peak areas divided by the total fatty acid integrated peak areas.

Enzyme activity
Product to precursor erythrocyte phospholipid fatty acid ratios served as indices of activity levels of enzymes involved in fatty acid metabolism. The index of -9 desaturase activity was calculated as the ratio of oleic acid to stearic acid (28), and the index of -4 desaturase activity was calculated as the ratio of DHA to docosapentaenoic acid (DPA) (29). The index of -6 desaturase was calculated in two ways: 1) as a ratio of dihomo--linolenic acid to linoleic acid (LA) (29) (this calculation includes the activities of the enzymes -6 desaturase and elongase 5); and 2) as a ratio of -linolenic acid to LA (28) (this calculation includes the activity of -6 desaturase only). The ratio of arachidonic acid (AA) to LA served as an index of desaturase/elongase-mediated LA → AA biosynthesis (30). The index of stearoyl-CoA desaturase activity was calculated as the ratio of palmitoleic acid to palmitic acid (29), and the index of elongase 2/5 activity was calculated as the ratio of adrenic acid to AA (31).

Full blood count analysis
Whole blood from fasted participants was collected into EDTA tubes, and full blood count analyses were conducted within 10 min of the collection using a Coulter Ac·T diff™. Data regarding red blood cell indices [hemoglobin, hematocrit, mean corpuscular volume, mean corpuscular hemoglobin, mean corpuscular hemoglobin concentration, and red blood cell distribution width (RDW)], white blood cell counts (lymphocytes, monocytes, and granulocytes), and platelet counts and indices (mean platelet volume, plateletcrit, and platelet distribution width) were collected. Instrument calibration was verified prior to measurement of each sample using appropriate quality controls.

Data analysis
Observational (case-control) study. Erythrocyte fatty acid content and full blood count analysis have not been investigated previously in small AAA and healthy control participants. Group size estimates for the observational study were based on RDW values reported in healthy controls (RDW, 13.1 ± 1.3, n = 40) and patients with coronary artery ectasia (RDW, 14.8 ± 1.6, n = 54) (32). A group size estimate of 14 was calculated with 85% power ( level of 0.05) using power/sample size (University British Columbia) and pooled variance (Solvers statistics) calculators.
Continuous demographic data for AAA patients and control participants were compared using a Student's t-test and are presented as mean ± SD. Categorical demographic variables were compared using a Fisher's exact test. Experimental data are presented as mean ± SEM, and between group differences were examined by Student's t-test analysis. The association between independent variables identified as being significantly different (see Table 1) and AAA was assessed using linear regression analysis with adjustment for covariates shown to be imbalanced between groups (hypertension, diabetes mellitus, coronary heart disease (CHD), dyslipidemia, use of statins or low-dose aspirin, current and previous smoking history, and age). All variables were introduced in one step in order of decreasing tolerance. Data were analyzed with Prism (GraphPad Software, La Jolla, CA). IBM SPSS Statistics Version 24 was used for multi-variable regression analysis and statistical significance was set at P < 0.05.

Observational (case-control) study
Baseline characteristics. Baseline characteristics of the case-control study are shown in Table 2. AAA patients were older and were characterized by a higher prevalence of Omega-3 fatty acid supplementation in small AAA 1157 hypertension, diabetes mellitus, dyslipidemia, CHD, and smoking history. AAA patients were more commonly prescribed statins and anti-platelet drugs.
Fatty acid methyl ester analysis. Twenty-three fatty acid methyl esters were consistently identified in erythrocyte membranes obtained from AAA patients and healthy control participants (see supplemental Table S1).
Erythrocyte fatty acid profiles. Mean omega-3 index values were similar in cases and controls, while the mean omega-6:omega-3 (n-6/n-3) ratio was significantly lower than the corresponding value obtained for the control cohort (P = 0.047; Table 3). Levels of the saturated fatty acid, margaric acid (C17:0), were significantly higher in AAA patients compared with healthy control participants (P = 0.007), while levels of the n-6 fatty acid, LA (C18:2), were significantly lower (P = 0.007). The n-3 PUFA, DPA (C22:5), was significantly higher in erythrocytes from AAA patients compared with control participants (P < 0.001). Data was adjusted for hypertension, diabetes mellitus, CHD, low-dose aspirin and statin use, active or previous smoking history, and age (Table 1).

Red blood cell indices, white blood cell counts, and platelet counts and indices. RDW and other red blood cell indices
were similar in cases and controls, as were white blood cell counts and platelet counts and indices following adjustment for covariates shown to be imbalanced between groups ( Table 5).

Omega-3 clinical trial
Baseline characteristics. Baseline characteristics were similar between groups with the exception of statin use (Table 2).
Participant compliance and tolerability. A high level of adherence to supplement intake was identified by return capsule counts (placebo: 96.4 ± 1.2%, 96.8 ± 2.1%, and 95.0 ± 2.9% at weeks 3, 8, and 12, respectively; omega-3 fatty acid group: 97.2 ± 0.9%, 94.0 ± 1.7%, and 95.0 ± 1.6% at weeks 3, 8, and 12, respectively). This was supported by GC-MS analysis of fatty acid incorporation in red blood cell membrane phospholipids. The omega-3 index was markedly increased in all participants in the omega-3 fatty acid group over the 12 week trial period ( Table 6). No change in omega-3 index was observed in participants who were randomized to receive placebo capsules. Two participants in the placebo group withdrew from the trial citing gastrointestinal disturbances (Fig. 1). Among participants who completed the 12 week trial, three in the placebo group reported burping, with one of them also experiencing nausea. In the omega-3 fatty acid group, three participants reported burping or reflux, and one participant reported flatulence.
Enzyme activity. Twelve week n-3 PUFA supplementation significantly increased the index of -4 desaturase (DPA → DHA) activity in erythrocytes from AAA patients (P < 0.001), while the index of -6 desaturase (LA → DGLA and LA → GLA) activity, the index of -9 desaturase (stearic → oleic) activity, and the index of stearoyl-CoA desaturase (palmitic → palmitoleic acid) activity remained unchanged ( Table 7). Omega-3 PUFA supplementation of AAA patients lowered the index of desaturase/elongasemediated LA → AA biosynthesis and the index of elongase 2/5 (AA → adrenic acid) activity to levels that were compa-rable to the control cohort (P = 0.031 and P = 0.002, respectively). No changes in any enzyme activity levels were observed for the placebo cohort.

Observational (case-control) study
Analysis of erythrocyte fatty acid profiles among AAA patients highlighted a mean omega-3 index value that was comparable to the control cohort in conjunction with an n-6/n-3 fatty acid ratio that was significantly lower. Among individual fatty acids measured, levels of the saturated fatty acid, margaric acid, and the omega-3 fatty acid, DPA, were significantly higher in erythrocytes from AAA patients,   and levels of the omega-6 fatty acid, LA, were significantly lower compared with control. The lower levels of LA in conjunction with higher levels of adrenic acid in the AAA cohort are of note. A meta-analysis of prospective cohort studies (n = 310,602 participants, n = 12,479 cases) indicated that the highest levels of dietary LA intake were associated with a 15% lower risk of CHD events and a 21% lower risk of CHD deaths when compared with the lowest levels of LA intake (34). In addition, a higher level of circulating LA in a community-based US cohort (n = 2,792) was associated with lower total mortality risk (extremequintile hazard ratio = 0.87; P = 0.005) and a lower risk of cardiovascular disease mortality (22% lower risk in the highest versus the lowest quintile; P = 0.02) (35). It has been suggested that the observed cardio-protective effects of LA are related to its competition with pro-inflammatory AA for reacylation into membrane phospholipids (36). In line with this, a strong inverse association has been reported between serum LA levels and high-sensitivity C-reactive protein (hsCRP), a key marker of inflammation (37). These findings suggest that the altered n-6 fatty acid profile in AAA skews cellular responses toward pro-inflammatory eicosanoid production and upregulated inflammatory pathways.
Desaturase enzymes catalyze the rate-limiting steps in long chain fatty acid synthetic pathways and their activities influence erythrocyte phospholipid fatty acid composition (28). In this study, -4 and -9 desaturase activity levels in erythrocytes from AAA patients were comparable to those of a healthy control cohort. In contrast, the index of desaturase/ elongase-mediated LA → AA biosynthesis was significantly higher in erythrocytes from AAA patients compared with control participants, as was the index of -6 desaturase activity. The higher index of -6 desaturase activity suggests that anti-inflammatory n-3 PUFA formation is repressed in favor of pro-inflammatory n-6 PUFA biosynthesis. It is of note that the higher index of desaturase/elongase-mediated LA → AA biosynthesis did not translate into higher levels of AA in this cohort. AA is a known substrate for inflammatory eicosanoid synthesis and, while it is possible that the absence of an increase in AA levels in AAA patients may be due to excessive shunting of this fatty acid to an alternative eicosanoid biosynthetic pathway, we have previously reported lower levels of prostaglandin E 2 (a product of AA metabolism) in this AAA cohort compared with healthy control participants. It is thus possible that the higher conversion of LA to AA in AAA patients led to increased flux through the n-6 biosynthetic pathway beyond AA to adrenic acid resulting in the observed higher levels of the latter. The anomalies observed in enzyme activity indices among AAA patients suggest a predisposition toward reactions that favor increased production of n-6 PUFAs with high proinflammatory potential and repressed production of their anti-inflammatory n-3 PUFA counterparts.

Omega-3 clinical trial
Twelve week n-3 PUFA supplementation decreased levels of the n-6 fatty acids, AA and adrenic acid, and increased levels of the n-3 fatty acids, DHA and EPA, in AAA patient erythrocytes, while concomitantly raising the mean omega-3 index to a value almost double that at study entry (8.03%). Importantly, the postsupplementation omega-3 index value was within the 8-12% range that affords cardioprotection (26). The large reduction in AA and adrenic acid levels with n-3 PUFA supplementation likely reflects competition of DHA and EPA for incorporation into existing erythrocyte membranes and/or greater availability of n-3 PUFAs for integration into newly synthesized membranes (38). The anti-inflammatory and immunomodulatory properties of increased cellular phospholipid EPA and DHA levels are likely to favorably impact AAA disease. Higher n-3 PUFA intake results in partial substitution of AA for EPA and DHA, a net decrease in pro-inflammatory eicosanoid production, and favorable impacts on inflammatory responses (39). The latter is due to competition of n-3 PUFAs with AA for the same metabolic enzymes, resulting in production of an alternate series of less biologically potent eicosanoids with weaker pro-inflammatory, platelet-aggregating, and vasoconstrictive activities (39)(40)(41). Omega-3 PUFAs are, in addition, substrates of cytochrome P450 enzymes, and increasing the levels of these fatty acids results in enhanced production of DHA-and EPA-derived metabolites at the expense of AA-derived metabolite production (42). The lower n-6/n-3 ratio, a value reflecting the balance between precursor PUFAs giving rise to downstream pro-and antiinflammatory eicosanoids, respectively, supports an n-3 PUFA-driven switch toward a more favorable eicosanoid profile.
Twelve week n-3 PUFA supplementation significantly increased the index of -4 desaturase activity in erythrocytes from AAA patients and lowered the index of desaturase/ elongase-mediated LA → AA biosynthesis and the index of elongase 2/5 activity to levels that were comparable to the control cohort (see supplemental Fig. S1 for a summary). The index of -6 desaturase activity was unaffected by n-3 PUFA supplementation. -4 desaturase and -6 desaturase each form part of a distinct biochemical pathway that yields DHA as a biosynthetic product. -6 desaturase forms part of a coupled microsomal-peroxisomal pathway that produces DHA through sequential desaturations and elongations, while -4 desaturase forms part of an alternative pathway that yields DHA through a single 4 desaturation step (43). As expected with DHA supplementation, the ratio of DHA:DPA increased. Determination of the contribution of -4 desaturase, if any, to this result would require direct measurement of the activity of this enzyme. The decreases in enzyme activities leading to pro-inflammatory n-6 PUFA production suggest that n-3 PUFA supplementation alters fatty acid metabolism in a manner that is likely to improve the inflammatory status of AAA patients.
Twelve week supplementation with n-3 PUFAs (1.8 g/day) significantly decreased RDW, red blood cell counts, hematocrit percentage, and platelet count and platelet percentage in AAA patients. RDW is a numerical measure reflecting size variability or heterogeneity of volume among circulating erythrocytes (44). It is well-documented that RDW correlates with inflammatory biomarkers in a multitude of clinical settings (45)(46)(47), and a large cohort study has supported the existence of a strong graded relationship between RDW and high-sensitivity C-reactive protein and RDW and erythrocyte sedimentation rate, independent of confounding factors (48). In light of this and evidence suggesting the existence of extensive cross-talk between the pathways of inflammation and coagulation (49,50), it is likely that the observed changes reflect n-3 PUFA-mediated improvements in inflammatory status in AAA patients who received this supplement.
The study, while limited by small sample size, was characterized by multiple strengths that included a placebocontrolled double-blind study design and use of a validated biomarker of erythrocyte membrane fatty acid content. Although it is clear that n-3 PUFA supplementation improves fatty acid status among AAA patients, further studies will be required to determine whether this improvement translates into positive alterations in the histopathologic appearance of AAA disease at the level of the aorta. In addition, while an aberrant fatty acid profile was observed in AAA patients compared with healthy control participants, it is not yet known why this occurs. It is possible that expression of fatty acid metabolizing enzymes is dysregulated in AAA patients, resulting in the observed aberrant profile of enzyme activity and erythrocyte fatty acid proportions. It is noteworthy that participants were screened at entry for dietary intake of fish and seafood, with all participants consuming no more than two oily fish meals per week. It is therefore unlikely that the observed differences in fatty acid profile are attributed to differences in diet between the two cohorts.
Taken together, the results presented here indicate that erythrocytes from AAA patients demonstrate a distinct fatty acid profile that is characterized by aberrant proportions of n-6 fatty acids with high inflammatory potential, while full blood count parameters among these patients are characterized by alterations that reflect the influence of systemic inflammation and oxidative stress. Improvements in inflammatory parameters and n-6 fatty acid status following n-3 PUFA supplementation suggest that dietary fatty acids represent a viable therapeutic intervention in AAA.