A fish-based diet intervention improves endothelial function in postmenopausal women with type 2 diabetes mellitus: A randomized crossover trial

M ET ABOL I SM CL IN I CA L A N D E XP E RI ME N TAL XX ( 2 0 14 ) XXX– X XX

Available online at www.sciencedirect.com

Metabolism www.metabolismjournal.com

A fish-based diet intervention improves endothelial function in postmenopausal women with type 2 diabetes mellitus: A randomized crossover trial Keiko Kondo a , Katsutaro Morino a,⁎, Yoshihiko Nishio b , Motoyuki Kondo a , Keiko Nakao a , Fumiyuki Nakagawa a, c , Atsushi Ishikado a, d , Osamu Sekine a , Takeshi Yoshizaki a , Atsunori Kashiwagi a , Satoshi Ugi a , Hiroshi Maegawa a a

Department of Medicine, Division of Endocrinology and Metabolism, Shiga University of Medical Science, Shiga, Japan Department of Diabetes and Endocrine Medicine, Division of Human and Environmental Sciences, Kagoshima University, Graduate School of Medical and Dental Sciences, Kagoshima, Japan c Osaka Laboratory, JCL Bioassay Corporation, Osaka, Japan d R&D Department, Sunstar Inc, Osaka, Japan b

A R T I C LE I N FO Article history:

AB S T R A C T Objective. The beneficial effects of fish and n-3 polyunsaturated fatty acids (PUFAs)

Received 22 October 2013

consumption on atherosclerosis have been reported in numerous epidemiological studies.

Accepted 8 April 2014

However, to the best of our knowledge, the effects of a fish-based diet intervention on endothelial function have not been investigated. Therefore, we studied these effects in

Keywords: Diet

postmenopausal women with type 2 diabetes mellitus (T2DM). Materials/Methods. Twenty-three postmenopausal women with T2DM were assigned to

n-3 PUFA

two four-week periods of either a fish-based diet (n-3 PUFAs ≧ 3.0 g/day) or a control diet in

Intervention studies

a randomized crossover design. Endothelial function was measured with reactive

Endothelial function

hyperemia using strain-gauge plethysmography and compared with the serum levels of

Diabetes mellitus

fatty acids and their metabolites. Endothelial function was determined with peak forearm blood flow (Peak), duration of reactive hyperemia (Duration) and flow debt repayment (FDR). Results. A fish-based dietary intervention improved Peak by 63.7%, Duration by 27.9% and FDR by 70.7%, compared to the control diet. Serum n-3 PUFA levels increased after the fishbased diet period and decreased after the control diet, compared with the baseline (1.49 vs. 0.97 vs. 1.19 mmol/l, p < 0.0001). There was no correlation between serum n-3 PUFA levels and endothelial function. An increased ratio of epoxyeicosatrienoic acid/ dihydroxyeicosatrienoic acid was observed after a fish-based diet intervention, possibly due to the inhibition of the activity of soluble epoxide hydrolase.

Abbreviations: ADMA, asymmetric dimethylarginine; ANOVA, analyses of variance; DHA, docosahexaenoic acid; DHET, dihydroxyeicosatrienoic acid; EET, epoxyeicosatrienoic acid; eNOS, endothelial nitric oxide synthase; EPA, eicosapentaenoic acid; FBF, forearm blood flow; FDR, flow debt repayment; GPR, G-protein coupled receptor; 4-HHE, 4-hydroxy hexenal; 4-HNE, 4-hydroxy nonenal; HOMA, homeostatic model assessment; hs-CRP, high-sensitivity C-reactive protein; LC-MS/MS, liquid chromatography-tandem mass spectrometry; MCP-1, monocyte chemotactic protein-1; NOS, nitric oxide synthase; 8-OHdG, 8-hydroxydeoxyguanosine; PUFA, polyunsaturated fatty acid; RH, reactive hyperemia; SBP, systolic blood pressure; sEH, soluble epoxide hydrolase. ⁎ Corresponding author at: Department of Medicine, Division of Endocrinology and Metabolism, Shiga University of Medical Science, Tsukinowa, Seta, Otsu, Shiga 520-2192, Japan. Tel.: + 81 77 548 2222; fax: +81 77 543 3858. E-mail address: [email protected] (K. Morino). http://dx.doi.org/10.1016/j.metabol.2014.04.005 0026-0495/© 2014 Elsevier Inc. All rights reserved.

Please cite this article as: Kondo K, et al, A fish-based diet intervention improves endothelial function in postmenopausal women with type 2 diabetes mellitus: A randomized c..., Metabolism (2014), http://dx.doi.org/10.1016/j.metabol.2014.04.005

2

ME TA BOL I SM CL IN I CA L A N D E XP E RI ME N TAL XX ( 2 0 14 ) XXX– X XX

Conclusions. A fish-based dietary intervention improves endothelial function in postmenopausal women with T2DM. Dissociation between the serum n-3 PUFA concentration and endothelial function suggests that the other factors may contribute to this phenomenon. © 2014 Elsevier Inc. All rights reserved.

1.

Introduction

The beneficial effects of dietary fish intake on coronary heart disease, sudden cardiac death, and all-cause mortality in the general population have been discussed for many decades [1]. A negative correlation between the dietary intake of fish and cardiovascular risk has been noted [2–4]. In diabetic women, a higher consumption of fish has been associated with a lower incidence of coronary heart disease and total mortality [5]. Although the mechanisms underlying the beneficial effects of dietary fish intake remain unclear, n-3 polyunsatulated fatty acids (PUFAs), such as eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), might be important factors. Endothelial dysfunction is a major contributor to the pathogenesis of cardiovascular disease. Many studies have indicated that endothelial dysfunction is caused by smoking, obesity, dyslipidemia, hypertension, and type 2 diabetes mellitus [6–9]. Supplementation of n-3 PUFAs has been shown to improve endothelial function in patients with hypercholesterolemia [10], peripheral arterial disease [11], acute myocardial infarction [12], and type 2 diabetes mellitus [13,14]. Few reports have sought to evaluate if a fish-based diet intervention, instead of n-3 PUFA supplements, can improve endothelial function directly. Eicosanoids are generated from other n-3 or n-6 PUFAs and control multiple functions, including inflammation, thrombosis and endothelial function [15]. Increasing evidence has revealed that epoxyeicosatrienoic acids (EETs) contribute to endothelium-dependent vasodilatation [16]. Soluble epoxide hydrolase (sEH), which converts EETs to dihydroxyeicosatrienoic acid (DHET), is a potential pharmacological target of endothelial dysfunction [17]. It has been reported that plasma EETs levels and the EET/DHET ratio decreased in patients with hypertension [18]. In addition, one recent study showed that fish oil supplementation increased the levels of some eicosanoids in young, healthy volunteers [19]. However, the mechanism by which a fish-based dietary intervention changes the circulating eicosanoids in subjects with type 2 diabetes mellitus is not known. This study had two aims. First, we examined the effects of a fish-based dietary intervention on endothelial function in postmenopausal women with type 2 diabetes mellitus. Second, we sought to analyze the relationships among endothelial function, n-3 PUFAs, and the concentrations of their metabolites.

3.1 kg/m2) at Shiga University of Medical Science Hospital. Individuals taking pioglitazone, EPA, or fish oil supplements; those undergoing insulin treatment; and those with poor glycemic control (HbA1c ≧ 8.4% [68.3 mmol/mol]), high fish intake (frequency of fish intake ≧ seven times per week), fish allergy, smoking, or excessive alcohol ingestion were excluded. The nature and potential risks of the study were explained to all participants, and written informed consent was obtained. The study was performed in accordance with the principles of the Declaration of Helsinki. The protocol was approved by the ethics committee of Shiga University of Medical Science. The study was registered at UMIN Clinical Trials Registry (http://www.umin.ac.jp/ctr/index.htm) with the Identification No. UMIN000002277.

2.2.

Study design

2.

Materials and Methods

The study included two experimental periods in a randomized crossover design. Participants were randomly assigned to start with either the fish-based diet or the control diet for 4 weeks. Eleven participants (fish-first group) started with the fish-based diet, and 12 participants (control diet-first group) started with the control diet for 4 weeks. After the first experimental period, the participants crossed over to the other experimental diet for an additional 4 weeks. Participants were instructed to take more than 3.0 g/day n-3 PUFA derived from fish (e.g., Pacific saury, salmon, sardines, etc.) during the fish-based diet period. Conversely, during the control diet period, participants were instructed to avoid fish intake, particularly n-3 PUFA-rich fish. The intake of total energy was 126–135 kJ per kg ideal body weight throughout the study period. Participants were advised by a dietician to modify their habitual diet qualitatively before each diet period. At baseline and at the end of each experimental period, a dietician provided participants with written and verbal instructions regarding the completion of a 3-day dietary record with a digital photograph of each meal. Participants also kept a food diary to record their daily fish intake. Nutritional intake, including intake of n-3 PUFA, was calculated using food composition tables specific for the Japanese people (Eiyoukun, ver5.0, Kenpakusha, Tokyo, Japan) [20]. Participants were advised on maintaining a constant physical activity level throughout the study. They underwent a day of testing at baseline and after 4 weeks of each diet intervention. Tests included anthropometric measurements, blood samples, and body composition measurements after a 10–16 h overnight fast. No medications were allowed to be changed throughout the study period.

2.1.

Study population

2.3.

Twenty-three postmenopausal type 2 diabetic women were recruited (mean age, 69.7 ± 6.6 years; mean BMI, 22.5 ±

Weight and body composition

Body weight was recorded with an electronic scale without shoes and with light clothing weighing a maximum of 0.1 kg.

Please cite this article as: Kondo K, et al, A fish-based diet intervention improves endothelial function in postmenopausal women with type 2 diabetes mellitus: A randomized c..., Metabolism (2014), http://dx.doi.org/10.1016/j.metabol.2014.04.005

M ET ABOL I SM CL IN I CA L A N D E XP E RI ME N TAL XX ( 2 0 14 ) XXX– X XX

The percentage of body fat was determined by bioelectrical impedance analysis (Bio Electrical Impedance Analyzer System, Tanita, Tokyo, Japan).

2.4. Assessment of endothelial function by strain-gauge plethysmography The endothelial function of each participant was examined after an overnight fast. Each participant was kept in a supine position during this measurement. After 15 min in the supine position, the basal forearm blood flow (FBF) was measured using a mercury-filled silastic (Dow Corning, Midland, MI, USA) strain-gauge plethysmography (EC-5R, D.E. Hokanson, Issaquah, WA, USA), as described by Linder et al. [21]. The effect of reactive hyperemia (RH) on FBF was measured as described in other studies, with minor modifications [22]. To induce RH, FBF was occluded by inflating the cuff on the right upper arm to a pressure of 190 mmHg when the systolic blood pressure (SBP) was 140 mmHg and under, or 50 mmHg plus SBP when the SBP was over 140 mmHg, for 5 min. FBF was measured after release of the cuff until it returned to the basal level. The peak FBF response [21], duration of RH, and total reactive hyperemic flow (flow debt repayment [FDR]) [23] during RH were used to assess the endothelial function of resistance vessels. The percent peak FBF, as an index of the peak FBF response, was obtained by calculating the increment of the peak FBF divided by the mean basal FBF.

2.5.

Laboratory analyses

All qualifying participants underwent a complete medical history evaluation and a physical examination. Blood samples were taken after an overnight fast. Plasma glucose concentrations were measured using the hexokinase glucose 6-phosphate dehydrogenase ultraviolet method. Serum insulin concentrations were measured using a chemiluminescent enzyme immunoassay. Serum triglyceride concentrations were determined enzymatically. Serum total cholesterol concentrations were measured by the cholesterol dehydrogenase ultraviolet method. Serum HDL-cholesterol concentrations were determined by a direct method (Cholestest N HDL, Sekisui Medical, Tokyo, Japan). LDL-cholesterol concentrations were calculated using the Friedewald equation (total cholesterol − [HDL-cholesterol + triglyceride/5]). Insulin resistance was evaluated using homeostatic model assessment (HOMA) and calculated as the product of the fasting glucose and insulin concentrations. HbA1c levels were measured by high-performance liquid chromatography. Serum adiponectin (Otsuka Pharmaceutical, Tokyo, Japan), monocyte chemotactic protein-1 (MCP-1) (R&D Systems, Minneapolis, MN, USA), and asymmetric dimethylarginine (ADMA) (Immundiagnostik, Bensheim, Germany) concentrations were measured by ELISA. Serum high-sensitivity C-reactive protein (hs-CRP) levels were measured by nephelometry. Urine 8-Hydroxydeoxyguanosine (8-OHdG) concentrations were measured by enzyme immunoassays.

2.6.

Measurement of PUFA and metabolites

Serum fatty acid profiles were determined by gas chromatography–mass spectrometry. Serum 4-hydroxy hexenal (4-HHE)

3

and 4-hydroxy nonenal (4-HNE) levels were measured by liquid chromatography–mass spectrometry, as previously described [24] (1 out of 23 samples was removed because of an extremely high value). Eicosanoids in human serum samples were measured by liquid chromatography–tandem mass spectrometry (LC–MS/MS) using a modified procedure [25]. After blood sampling by vein puncture, the serum was separated by centrifugation at 1750 g for 10 min at room temperature, and 1000 μL of sample was transferred into polypropylene tubes and stored at –80 °C until measurement. Serum samples as an internal solution (50 ng/mL; PGE2-d4, 6-keto PGF1a-d4 20-HETE-d6 (Cayman Chemical, Ann Arbor, MI, USA) diluted with methanol) were spiked to all samples, including calibration standards. For the preparation of calibration standards, isotonic sodium chloride solution (Otsuka Pharmaceutical Factory, Tokushima, Japan) was used as the surrogate matrix. Into this surrogate matrix, a standard solution of 84 compounds of eicosanoid (Cayman Chemical), diluted with methanol in each concentration (0.05–100 ng/mL), was spiked. Calibration curve samples were extracted using the same procedure as for the serum samples. For the extraction procedure, 2 mL of water was added to all samples for dilution. After vortex mixing, the samples were added to solid-phase extraction cartridges (Empore C18-SD, 3 M, Maplewood, MN, USA). Ethyl acetate was used for elution, and this elute was evaporated under stream nitrogen gas at 36 °C. After being reconstituted with 40 μL of methanol, this aliquot was injected into an optimized liquid chromatography–tandem mass spectrometry system. Liquid chromatography was performed using an ACQUITY UPLC (Waters, Milford, MA, USA), and an API4000 triple quadrupole tandem mass spectrometer (AB Sciex, Foster City, CA, USA) was used as a detector. An analytical column (ACQUITY UPLC HSS T3, 1.8 μm, 2.1 mm × 150 mm, Waters) was used. Column temperature was maintained at 60 °C. Injected samples were eluted using water/acetonitrile/acetic acid (7:3:0.1, v/v/v) and acetonitrile/2-propanol (1:1, v/v), with a linear gradient at a total flow of 0.215 mL/min. To operate the API4000, electronic spray ionization in negative mode with selected reaction monitoring was used. The selected reaction monitoring transition parameter, the database of Lipid Map (http://www.lipidmaps.org/), was referred to for selected precursors to produce ions of m/z, respectively. Other parameters were adjusted to optimum values.

2.7.

Statistical analyses

We previously observed that fish-based dietary intervention increased the peak FBF by 47.0% in 10 healthy volunteers (unpublished data). In this study, we calculated a sample size using these data. As a result, 22 participants would have been significant to detect changes with a power of 80% and a 95% confidence level. Statistical analyses were performed with SPSS version 17.0 (SPSS, Tokyo, Japan, 2008). The distribution of variables was analyzed by checking histograms and normal plots of the data, and the normality was tested using the Kolmogorov– Smirnov and Shapiro–Wilk tests. Changes in nutritional intake were compared for all participants, and metabolic variables, serum fatty acid and endothelial function were compared in each group. For normally distributed data,

Please cite this article as: Kondo K, et al, A fish-based diet intervention improves endothelial function in postmenopausal women with type 2 diabetes mellitus: A randomized c..., Metabolism (2014), http://dx.doi.org/10.1016/j.metabol.2014.04.005

4

ME TA BOL I SM CL IN I CA L A N D E XP E RI ME N TAL XX ( 2 0 14 ) XXX– X XX

Table 1 – Physical and clinical characteristics of the subjects at baseline. Variable

All subjects (n = 23)

Fish-first group (n = 11)

Control diet-first group (n = 12)

Age (years) BMI (kg/m2) Body fat (%) SBP (mmHg) DBP (mmHg) Glucose (mmol/l) Insulin (pmol/l) HbA1c (%) HbA1c (mmol/mol) Total cholesterol (mmol/l) HDL-cholesterol (mmol/l) LDL-cholesterol (mmol/l) Triglyceride (mmol/l) Fish consumption (times/week) Use of glucose-lowering agents, n (%) Use of statin, n (%) Use of anti-hypertensive agents, n (%)

69.7 ± 6.6 22.5 ± 3.1 32.0 ± 5.3 131.3 ± 12.3 73.3 ± 8.2 6.88 ± 1.10 31.6 ± 15.0 6.8 ± 0.4 51.0 ± 4.7 5.71 ± 0.95 1.65 ± 0.31 3.53 ± 0.88 2.65 ± 1.39 5.6 ± 2.5 13 (56.5) 8 (34.8) 9 (39.1)

69.5 ± 6.0 22.0 ± 2.7 30.6 ± 4.3 133.9 ± 6.3 73.2 ± 5.2 7.06 ± 0.94 30.3 ± 14.3 6.9 ± 0.4 51.6 ± 4.5 5.16 ± 0.57 1.52 ± 0.28 3.09 ± 0.52 2.78 ± 1.64 5.1 ± 2.2 6 (54.5) 6 (54.5) 6 (54.5)

69.8 ± 7.3 23.0 ± 3.5 33.4 ± 6.0 129.0 ± 16.0 73.5 ± 10.5 6.71 ± 1.25 32.9 ± 16.1 6.8 ± 0.5 50.5 ± 5.0 6.21 ± 0.97 1.77 ± 0.30 3.93 ± 0.96 2.54 ± 1.18 6.1 ± 2.7 7 (58.3) 2 (16.7) 3 (25.0)

p value 0.919 a 0.447 a 0.209 a 0.341 a 0.938 a 0.457 a 0.684 a 0.563 a 0.563 a 0.019 a 0.053 a 0.037 a 0.686 a 0.330 a 1.000 b 0.089 b 0.214 b

Values are means ± SD for continuous variables. DBP, diastolic blood pressure; SBP, systolic blood pressure. a Unpaired t test. b Chi-square test.

repeated measures analysis of variance (RM-ANOVA) was performed for comparisons of different periods of determinations. In addition, post-hoc testing was performed using Tukey’s honest significance difference test. For variables with non-normal distributions, the non-parametric Friedman test with Wilcoxon test was used as a post-hoc test with Bonferroni correction. Analysis of covariance (ANCOVA) was performed to compare the effects of each dietary treatment. All data are expressed as the mean ± SD for continuous variables. P-values of < 0.05 were considered statistically significant, unless specifically described.

3.

Results

3.1.

Participants and diet diaries

The baseline characteristics of the study participants are shown in Table 1. The percentages of participants using oral glucose-lowering, lipid-lowering, or anti-hypertensive agents were 56.5%, 43.5%, and 39.1%, respectively. No participant stopped or started these medications during the study. There were no differences in potential confounding factors, such as age, BMI and medication, between the 2 groups, although the total and LDL-cholesterol levels were higher in the control diet-first group compared with fish-first group. Nutritional intake was estimated by analyzing diet diaries at baseline, during the fish-based diet period, and during the control diet period (Table 2). The intake of total energy was unchanged throughout the study period but decreased during the control diet period. The intake of protein significantly decreased, and that of carbohydrate significantly increased, during the control diet period compared with the baseline and fish-based diet periods. The intake of fat was unchanged throughout the study period. The intake of fish and n-3 PUFA showed a significant increase during the fish-based diet

period and a decrease during the control diet period, compared with that during the baseline period.

3.2.

Changes in metabolic variables in each group

The changes in metabolic variables during the study period are shown in Table 3 and were analyzed separately for each group. In both groups, the anthropometric and body composition values of the study participants did not change significantly during the study period. In the fish-first group, serum triglyceride concentrations decreased at week 4 of the fish-based diet period and increased at week 8 of the control diet period. Similarly, in the control diet-first group, serum triglyceride concentrations increased at week 4 of the control diet period and decreased at week 8 of the fish-based diet period. HDL-cholesterol concentrations tended to increase after the fish-based diet period and to decrease after the control diet period in both groups. Total cholesterol and LDLcholesterol concentrations did not change significantly during the study period. Fasting plasma glucose, serum insulin, HbA1c levels and HOMA-R were unchanged during the study period. In the control diet-first group, serum adiponectin levels decreased significantly after both the control and fishbased diet periods compared with the baseline period, but not in the fish-first group. Serum MCP-1, hs-CRP, ADMA and urine 8-OHdG levels were unchanged during the study period in both groups (Table 3).

3.3. Change in metabolic variables before and after each diet in all patients The changes in metabolic variables in all patients are shown in Table 4. The anthropometric and body composition values of the study subjects did not change significantly during the study period. The fish-based diet significantly decreased serum total cholesterol (p = 0.013) and triglyceride concentration (p = 0.008)

Please cite this article as: Kondo K, et al, A fish-based diet intervention improves endothelial function in postmenopausal women with type 2 diabetes mellitus: A randomized c..., Metabolism (2014), http://dx.doi.org/10.1016/j.metabol.2014.04.005

5

M ET ABOL I SM CL IN I CA L A N D E XP E RI ME N TAL XX ( 2 0 14 ) XXX– X XX

Table 2 – Nutritional intake at baseline and changes during the intervention. Variable Energy (kJ) Energy (kJ/kg) Protein (% energy) Carbohydrate (% energy) Fat (% energy) n-3 PUFA (g/day) n-3 PUFA (g/day) (from fish) n-6 PUFA (g/day) n-6/n-3 Fish intake (g/day)

Baseline 7508 143 16.9 55.6 25.7 3.2 1.9 9.3 3.6 123.5

± ± ± ± ± ± ± ± ± ±

1116 25 1.9 6.4 5.6 2.0 1.8 3.3 1.9 59.5

Fish-based diet period 7349 139 17.6 54.0 26.6 4.3 3.2 8.5 2.2 159.7

± ± ± ± ± ± ± ± ± ±

1046 22 1.6 5.8 5.5 1.4 c 1.1 2.4 1.0 31.9

Control diet period 7035 134 15.4 58.2 24.7 1.5 0.2 9.4 6.4 44.6

± ± ± ± ± ± ± ± ± ±

p value a

1060 23 1.2 c, f 5.3 e 5.0 0.5 d, f 0.3 2.4 1.2 30.8 d, f

p value b

0.186 0.164 <0.0001 0.005 0.268 <0.0001 <0.0001 0.333 <0.0001 <0.0001

Values are means ± SD for continuous variables. PUFA, polyunsaturated fatty acid. a Repeated-measures ANOVA. b Friedman test. c p < 0.01. d p < 0.0001 vs baseline. e p < 0.01. f p < 0.0001 vs fish-based diet period (analyzed by the post hoc test).

and tended to decrease LDL-cholesterol concentration (p = 0.057), whereas the control diet increased serum triglyceride (p = 0.015) but did not change total cholesterol and LDLcholesterol concentration. The HDL-cholesterol concentration increased significantly after the fish-based diet period (p = 0.028) but decreased after the control diet period (p = 0.036). Fasting plasma glucose and serum insulin concentrations were slightly but significantly elevated, and as a result, the HOMA-R was elevated at the end of the fish-based diet period. However, HbA1c was unchanged throughout the study period. Serum adiponectin, MCP-1, hs-CRP, ADMA and urine 8-OHdG concentrations were unchanged during the study period. These changes were also observed after adjusting for the use of glucose-lowering agents, statins or anti-hypertensive agents (data not shown).

3.4.

Changes in endothelial function

(control diet period). Serum n-6 PUFA concentrations decreased after 4 weeks (fish-based diet period) and significantly increased after 8 weeks (control diet period) (p < 0.0001). Serum 4-HHE levels increased after 4 weeks (fish-based diet period) and significantly decreased after 8 weeks (control diet period). Serum 4-HNE levels were unchanged throughout the study period. Conversely, in the control diet-first group, serum EPA (p < 0.0001), DHA (p < 0.0001) and total n-3 PUFA (p < 0.0001) concentrations significantly decreased after 4 weeks (control diet period) and significantly increased after 8 weeks (fish-based diet period). Serum n-6 PUFA concentrations increased after 4 weeks (control diet period) and significantly decreased after 8 weeks (fish-based diet period) (p < 0.0001). Serum 4-HHE levels significantly decreased after 4 weeks (control diet period) and significantly increased after 8 weeks (fish-based diet period). Serum 4-HNE levels were unchanged throughout the study period.

The changes in endothelial function evaluated by straingauge plethysmography for each group are shown in Fig. 1. In the fish-first group, the peak FBF response and FDR increased markedly after the fish-based diet period (4 weeks), and these effects were sustained after the control diet period (8 weeks). The duration of RH was unchanged during the study period. Conversely, in the control diet-first group, the peak FBF response, duration of RH, and FDR were unchanged after the control diet period (4 weeks). After the fish-based diet period, the durations of RH and FDR increased significantly, and the peak FBF showed improvement.

3.6.

3.5.

4.

Change in serum fatty acid profiles

Serum fatty acid profiles were determined for all participants. Changes in the serum fatty acid concentrations of each group are shown in Fig. 2. In the fish-first group, serum EPA (p < 0.0001), DHA (p < 0.0001) and total n-3 PUFA (p < 0.0001) concentrations significantly increased after 4 weeks (fishbased diet period) and significantly decreased after 8 weeks

Change in serum eicosanoids

Changes in the 86 species of serum eicosanoids evaluated by LC–MS/MS are shown in Fig. 3A and Supplemental Figs. 1–3. Several metabolites could not be detected, and so we focused on EET, a known vasodilator derived from arachidonic acid. The concentrations of 11,12-EET and 14,15-EET were not significantly different after a fish-based diet intervention (Supplemental Fig. 1), but the 11,12EET/11,12DHET ratio was significantly increased during the fish period (Fig. 3B).

Discussion

In the present study, we made two important clinical observations. First, a fish-based dietary intervention improved endothelial function in postmenopausal women with type 2 diabetes mellitus. Second, the improvement of endothelial function was not correlated with the serum concentrations of n-3 PUFAs.

Please cite this article as: Kondo K, et al, A fish-based diet intervention improves endothelial function in postmenopausal women with type 2 diabetes mellitus: A randomized c..., Metabolism (2014), http://dx.doi.org/10.1016/j.metabol.2014.04.005

6

ME TA BOL I SM CL IN I CA L A N D E XP E RI ME N TAL XX ( 2 0 14 ) XXX– X XX

Table 3 – Changes in the metabolic variables. a) Fish-first group (n = 11) Variable Body weight (kg) Body fat (%) Total cholesterol (mmol/l) Triglyceride (mmol/l) HDL-cholesterol (mmol/l) LDL-cholesterol (mmol/l) Glucose (mmol/l) Insulin (pmol/l) HbA1c (%) HbA1c (mmol/mol) HOMA-R Adiponectin (ng/ml) MCP-1 (pg/ml) hs-CRP (mg/l) ADMA (μmol/l) 4-HHE (nmol/l) c 4-HNE (nmol/l) c 8-OHdG (ng/mg · creatinine)

Baseline (0 week) 53.8 30.6 5.16 2.78 1.52 3.09 7.06 30.3 6.9 51.6 1.6 8.7 519.2 0.60 0.46 12.8 5.2 12.8

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

6.5 4.3 0.57 1.64 0.28 0.52 0.94 14.3 0.4 4.5 0.9 3.3 972.5 0.34 0.12 6.3 2.0 5.6

After the fish-based diet period (4 weeks) 53.9 30.9 4.97 2.88 1.55 2.84 7.37 33.2 6.9 52.2 1.8 8.7 550.8 0.51 0.47 17.7 6.3 12.2

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

6.9 5.2 0.49 1.60 0.29 0.51 1.09 17.5 0.4 4.2 1.0 2.7 1003.3 0.42 0.12 6.8 2.5 5.0

After the control diet period (8 weeks) 53.4 30.9 5.09 3.31 1.47 2.96 7.06 30.2 6.9 52.1 1.6 9.0 514.1 1.81 0.44 12.3 7.1 13.8

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

6.6 4.8 0.57 1.56 d 0.31 0.53 1.00 15.8 0.4 4.0 0.9 3.2 892.4 3.64 0.14 7.3 e 3.3 8.4

p value a

p value b

0.078 0.721 0.307 0.015 0.186 0.099 0.061 0.754 0.272 0.272 0.460 0.498 0.152 0.490 0.695 0.025 0.121 0.695

b) Control diet-first group (n = 12) Variable

Body weight (kg) Body fat (%) Total cholesterol (mmol/l) Triglyceride (mmol/l) HDL-cholesterol (mmol/l) LDL-cholesterol (mmol/l) Glucose (mmol/l) Insulin (pmol/l) HbA1c (%) HbA1c (mmol/mol) HOMA-R Adiponectin (ng/ml) MCP-1 (pg/ml) hs-CRP (mg/l) ADMA (μmol/l) 4-HHE (nmol/l) 4-HNE (nmol/l) 8-OHdG (ng/mg · creatinine)

Baseline (0 week) 54.1 33.4 6.21 2.54 1.77 3.93 6.71 32.9 6.8 50.5 1.6 11.4 244.9 0.49 0.42 16.7 6.7 9.8

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

7.8 6.0 0.97 1.18 0.30 0.96 1.25 16.1 0.5 5.0 0.8 3.8 21.8 0.25 0.07 9.0 3.6 3.0

After the control diet period (4 weeks) 54.2 33.1 6.26 3.44 1.71 3.87 6.46 29.7 6.8 51.0 1.4 10.7 231.7 0.80 0.43 9.4 5.4 11.1

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

7.8 5.4 0.79 1.90 0.34 0.85 1.25 17.2 0.5 5.4 0.7 3.8 d 36.8 0.72 0.09 6.1 3.3 5.1

After the fish-based diet period (8 weeks) 54.1 33.2 6.10 2.28 1.79 3.86 6.78 35.4 6.8 50.5 1.8 10.7 213.5 0.79 0.47 18.2 6.1 12.4

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

7.6 5.9 0.80 0.94 e 0.38 0.83 1.25 21.4 0.5 5.2 1.2 3.7 d 35.5 0.67 0.10 13.8 4.3 4.1

p value a

p value b

0.966 0.605 0.205 0.011 0.179 0.834 0.234 0.231 0.439 0.439 0.143 0.010 0.178 0.182 0.105 0.002 0.435 0.202

Values are means ± SD for continuous variables. HOMA, homeostatic model assessment; MCP-1, monocyte chemotactic protein-1; hs-CRP, high-sensitivity C-reactive protein; ADMA, asymmetric dimethylarginine; HHE, hydroxy hexenal; HNE, hydroxy nonenal; 8-OHdG, 8-Hydroxydeoxyguanosine. a Repeated measures ANOVA. b Friedman test. c n = 10 for 4-HHE and 4-HNE. d p < 0.05 vs baseline. e p < 0.05, vs 4 weeks (analyzed by the post hoc test).

The finding that a fish-based diet intervention improved endothelial function in postmenopausal women with type 2 diabetes mellitus is in agreement with a study showing that supplementation with fish oil improves endothelial function in the normoglycemic offspring of subjects with type 2 diabetes mellitus [26] and in patients with type 2 diabetes mellitus [13,14]. This is the first report to show that fish as a dietary intervention can improve endothelial function in humans, as far as we searched. This phenomenon may

contribute to the cardio-protective effects of fish, together with the beneficial effects of fish oil on dyslipidemia. Interestingly, we observed no association between improved endothelial function and serum concentrations of n-3 PUFAs. This is in agreement with more recent interventional studies showing no improvements in the endothelial function by fish oil supplementation, even though fish oil supplementation increased serum n-3 PUFA concentrations [27,28]. This discrepancy might explain the controversial results upon

Please cite this article as: Kondo K, et al, A fish-based diet intervention improves endothelial function in postmenopausal women with type 2 diabetes mellitus: A randomized c..., Metabolism (2014), http://dx.doi.org/10.1016/j.metabol.2014.04.005

7

M ET ABOL I SM CL IN I CA L A N D E XP E RI ME N TAL XX ( 2 0 14 ) XXX– X XX

Table 4 – Changes in the metabolic variables before and after each diet for 4 weeks in a crossover design. Variable

Fish-based diet period Before

Body weight (kg) Body fat (%) Total cholesterol (mmol/l) Triglyceride (mmol/l) HDL-cholesterol (mmol/l) LDL-cholesterol (mmol/l) Glucose (mmol/l) Insulin (pmol/l) HbA1c (%) HOMA-R Adiponectin (ng/ml) MCP-1 (pg/ml) hs-CRP (mg/l) ADMA (μmol/l) 4-HHE (mmol/l) b 4-HNE (mmol/l) b 8-OHdG (ng/mg · creatinine)

54.0 31.9 5.74 3.12 1.62 3.50 6.75 30.0 6.8 1.5 9.7 369.2 0.70 0.45 10.9 5.3 11.9

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

Control diet period

After

7.1 5.0 0.88 1.77 0.32 0.80 1.13 15.5 0.4 0.8 3.6 672.4 0.57 0.10 6.2 2.7 5.3

54.0 32.1 5.56 2.56 1.68 3.37 7.06 34.4 6.8 1.8 9.8 374.8 0.65 0.47 18.0 6.2 12.3

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

p value

7.1 5.6 0.87 1.30 0.36 0.86 1.19 19.2 0.4 1.1 3.4 698.5 0.57 0.11 10.9 3.5 4.4

a

0.701 0.464 0.013 0.008 0.028 0.057 0.013 0.033 1.000 0.013 0.769 0.602 0.664 0.148 0.002 0.256 0.673

Before 54.0 32.2 5.61 2.70 1.66 3.41 7.03 33.1 6.8 1.7 10.1 391.2 0.50 0.45 17.2 6.5 11.0

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

p value a

After

7.2 5.6 0.99 1.38 0.31 0.94 1.20 16.4 0.4 0.9 3.5 694.4 0.34 0.10 7.9 3.1 4.1

53.8 32.0 5.70 3.38 1.59 3.43 6.75 30.5 6.9 1.5 9.9 366.7 1.28 0.43 10.7 6.1 12.4

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

7.1 5.1 0.90 1.71 0.34 0.84 1.15 16.1 0.4 0.8 3.6 619.2 2.56 0.11 6.6 3.3 6.8

0.081 0.242 0.378 0.015 0.036 0.755 0.010 0.252 0.443 0.108 0.166 0.169 0.114 0.472 0.001 0.630 0.080

Values are means ± SD for continuous variables. a Analysis of covariance (adjusted for intervention allocation). b n = 22 for 4-HHE and 4-HNE.

A

B

Peak control diet

fish-based diet

1000

C

Duration control diet

fish-based diet

*

120

400

p=0.018

80

FDR (%)

600

Duration (s)

800

Peak (%)

control diet

fish-based diet

150

120

*

FDR

40

90 60

p=0.402

p=0.031

200

30

0

0 0

D

4

8 (week)

0 0

E

Peak

4

8 (week)

F

Duration

control diet fish-based diet

0

120

control diet fish-based diet

150

††

†

**

**

120

p=0.121

80

FDR (%)

400

Duration (s)

Peak (%)

800 600

40

p=0.001

200 0

0 0

4

8 (week)

8 (week)

FDR

control diet fish-based diet

1000

4

90 60 30

p=0.003

0 0

4

8 (week)

0

4

8 (week)

Fig. 1 – Changes in endothelial function, as evaluated by strain-gauge plethysmography. Endothelial function evaluated by strain-gauge plethysmography in the fish-first group (n = 11, black circle, A–C) and control diet-first group (n = 12, white circle, D–F). Data are shown as means ± SD. Peak forearm blood flow (Peak), duration of reactive hyperemia (Duration), or flow debt repayment (FDR) increased after the fish-based diet period. *p < 0.05, **p < 0.01 vs. 0 weeks, †p < 0.05, ††p < 0.01 vs. 4 weeks (analyzed by post hoc repeated-measures ANOVA). Please cite this article as: Kondo K, et al, A fish-based diet intervention improves endothelial function in postmenopausal women with type 2 diabetes mellitus: A randomized c..., Metabolism (2014), http://dx.doi.org/10.1016/j.metabol.2014.04.005

8

ME TA BOL I SM CL IN I CA L A N D E XP E RI ME N TAL XX ( 2 0 14 ) XXX– X XX

0.6 * †††

0.4

1.0

0.2

E

4

0.4

p<0.0001

F 1.2

control diet fish-based diet ***

0.6 **

0.4

0.0

4

8 (week)

1.5

†††

1.0 0.5

p<0.0001

††

**

p<0.0001

0

4

8 (week)

4

control diet

fish-based diet

6.0 5.0

††

4.0 3.0 2.0

p=0.004

1.0

8 (week)

n-3 PUFA control diet fish-based diet

2.5

** †††

2.0 **

1.5 1.0

p<0.0001

0.5 0.0

n-6 PUFA

0.0 0

3.0

control diet fish-based diet

0.4

0.0

**

G

0.6

0.2

2.0

8 (week)

DHA

0.8

0.2 0

4

1.0

DHA (mmol/l)

†††

p<0.0001

2.5

0.0 0

8 (week)

EPA

1.0

EPA (mmol/l)

0.6

0.0 0

0.8

††

0.2

0.0

1.2

*

0.8

7.0

control diet

fish-based diet

n-6 PUFA (mmol/l)

***

3.0

control diet

fish-based diet

D

n-3 PUFA

0

4

n-6 PUFA (mmol/l)

p<0.0001

0.8

C

DHA

n-3 PUFA (mmol/l l)

EPA (mmol/l)

1.0

1.2

control diet

fish-based diet

DHA (mmol/l)

1.2

B

EPA

n-3 PUFA (mmol/l)

A

8 (week)

0

4

8 (week)

H

n-6 PUFA

7.0

control diet fish-based diet

6.0 ***

5.0

†††

4.0 3.0

p<0.0001

2.0 1.0 0.0

0

4

8 (week)

Fig. 2 – Changes in serum fatty acid concentrations. Changes in serum fatty acid concentrations in the fish-first group (n = 11, black circle, A–D) and control diet-first group (n = 12, white circle, E–H). Data are shown as means ± SD. *p < 0.05, **p < 0.01, ***p < 0.0001 vs. 0 weeks, ††p < 0.01, †††p < 0.0001 vs. 4 weeks (analyzed by post hoc repeated measures ANOVA). EPA, eicosapentaenoic acid; DHA, docosahexaenoic acid; PUFA, polyunsaturated fatty acid.

which n-3 PUFA supplementation and serum n-3 PUFA concentration were not found to predict cardiovascular risk, in contrast to the fish consumption in a population-based study [29–31]. Therefore, we focused on the lipid components of cell membrane and lipid metabolites in this study. To investigate the membrane components, saline-washed red blood cells were also analyzed. The fish dietary intervention increased n-3 PUFA levels in the red blood cells, but a correlation was not observed between the change in the n-3 PUFA concentration and the change in endothelial function (data not shown). We measured the metabolites of PUFA and the eicosanoids to seek potential mediators of endothelial function. In agreement with our previous study in an animal model [24], fish-based dietary interventions increased the serum concentrations of 4-HHE only in the fish diet period (Table 3). However, the increments of 4-HHE levels were not correlated with endothelial function (data not shown). EETs are metabolites of arachidonic acid due to the action of cytochrome P450 (CYP) epoxigenase enzymes; they act to regulate the tone of blood vessels [16]. Plasma concentrations of EETs decreased in hypertensive patients and in those with hyperhomosystinemia [18]. In addition, many studies have reported that CYP inhibitors reduce the endothelium-dependent dilatation of human arteries, suggesting a physiological role of EETs [32]. Furthermore, studies on both the genetic and pharmacological inhibition of sEH, which converts EETs to DHET, have suggested the notion of sEH as a potential therapeutic target of cardiovascular disease and endothelial dysfunction [17,33]. We observed an increased EET/DHET ratio after a fish-based diet intervention, potentially because

the intervention decreased sEH activity (Fig. 3B). Our preliminary data are supported by a report in which dietary fish oil decreased hepatic sEH protein levels in apoE knockout mice [34]. Interestingly, in this study, we observed that the improvement in the endothelial function was sustained for at least 4 weeks after the fish-based diet intervention period in the fish-first group. We observed this persistent effect, even with drops in the n-3 PUFA component of the serum (Fig. 2) and RBC membrane (data not shown) after finishing the fish period. The possible mechanisms underlying the improvement in endothelial function and its persistent effect by a fish-based diet intervention might be the anti-inflammatory and antioxidative effects of n-3 PUFA. Recently, n-3 PUFA has been recognized as a ligand of G-protein coupled receptor 120 (GPR120) that is able to suppress the expression of MCP-1 and interleukin (IL)-6 in adipocytes [35]. In human studies, it has been reported that the supplementation of n-3 PUFA reduced serum levels of IL-18 in men with high cardiovascular risk. [36]. Oxidative stress a major mechanism of endothelial dysfunction caused by smoking and hypertension, and it is thought to be a potential target of n-3 PUFA [37]. However, we did not see changes in the serum levels of MCP-1 and hs-CRP, or urinary levels 8-OHdG and 4-HNE, suggesting that the antiinflammatory or anti-oxidative effects of n-3 PUFA might not be major causes of the improvement in the endothelial function. Instead, it has been suggested that lipid metabolites or the other nutrients from fish other than PUFA are fundamental players in improving endothelial function via dietary interventions in fish.

Please cite this article as: Kondo K, et al, A fish-based diet intervention improves endothelial function in postmenopausal women with type 2 diabetes mellitus: A randomized c..., Metabolism (2014), http://dx.doi.org/10.1016/j.metabol.2014.04.005

9

M ET ABOL I SM CL IN I CA L A N D E XP E RI ME N TAL XX ( 2 0 14 ) XXX– X XX

B

LOX

AA

CYP

COX HETEs

Leukotriens

HETEs

EETs

0.1

0.05

0

Prostanoids

14,15EET

sEH

vasodilatation

0.03

0.02

0.01

0

* 0

p=0.024

4

fish-based diet

8

control diet

(week)

11,12DHET 14,15DHET

0.25

1

14,15EET/14,15DHET ratio

11,12EET

Δ 11,12EET/11,12DHET ratio

PLA2

0.04

0.15

11,12EET/11,12DHET ratio

Plasma Membrane

Δ 14,15EET/14,15DHET ratio

A

0.8 0.6 0.4 0.2

* 0

p=0.065 0.2 0.15 0.1 0.05 0

0

4

8

fish-based diet

control diet

(week) Fig. 3 – Lipid metabolites. (A) Metabolism of arachidonic acid. AA, arachidonic acid; LOX, lipoxygenase; COX, cyclooxygenase; CYP, cytochrome P; HETE, hydroxyeicosatetraenoic acid; EET, epoxyeicosatrienoic acid; DHET, dihydroxyeicosatrienoic acid; sEH, soluble epoxide hydrolase. (B) Change in the EET/DHET ratio throughout the study period. The 11,12EET/11,12DHET and 14,15EET/14,15DHET ratios in the fish-first group (black circle) and control diet first group (white circle). Changes were calculated as Δ value (a difference between pre- and post-intervention). Data are shown as means ± SD. *p < 0.05 vs. 4 weeks (analyzed by post hoc repeated measures ANOVA). EET, epoxyeicosatrienoic acid; DHET, dihydroxyeicosatrienoic acid.

A source of significance of this paper is that the fish-based dietary intervention improved endothelial function and systemic EETs/DHETs ratio. Because drugs that raise EETs level are now in clinical trials for hypertension and other conditions [38], dietary treatments may have impact, not only pharmacologically but also health-economically. The present study had limitations. First, the number of participants was relatively small. The small study cohort could have led to an underestimation of the effect a fish-based diet intervention on multiple parameters. Second, there was no washout period between the two treatments. We observed a persistent effect on the primary outcome; in the future, further studies in different settings are necessary to clarify the mechanism of this persistent effect. In conclusion, a fish-based dietary intervention improved lipid metabolism and endothelial function in postmenopausal women with type 2 diabetes mellitus. The dissociation between n-3 PUFA and endothelial function suggests that other factors may contribute to this phenomenon.

Contribution statement KK wrote the manuscript. KK, MK, KN and FN collected data. KK, KM, YN and MK researched data. KM, YN, AI, OS, US, AK and HM

reviewed/edited the manuscript. KK, KM, YN and HM designed the study and obtained the grants. All authors approved the final version.

Funding This work was supported in part by a Grant-in-Aid from the Ministry of Education, Culture, Sports, Science and Technology of Japan (to K. Kondo, K. Morino and H. Maegawa), a Grant-in-Aid for cardiovascular disease from the Ministry of Health, Labour and Welfare of Japan (to K. Morino and K. Kondo), the Osaka Gas Group Welfare Foundation (to K. Morino) and the Mitsui Life Social Welfare Foundation (to K. Morino).

Conflicts of interest F. Nakagawa is an employee of JCL Bioassay and a graduate student of Shiga University of Medical Science. This fact does not alter the author’s adherence to all of the policies of Metabolism regarding the sharing of data and materials.

Please cite this article as: Kondo K, et al, A fish-based diet intervention improves endothelial function in postmenopausal women with type 2 diabetes mellitus: A randomized c..., Metabolism (2014), http://dx.doi.org/10.1016/j.metabol.2014.04.005

10

ME TA BOL I SM CL IN I CA L A N D E XP E RI ME N TAL XX ( 2 0 14 ) XXX– X XX

Acknowledgments We are indebted to Katsuyuki Miura, Yoshitaka Murakami, and Shinji Kume for their expert technical assistance with the studies and to the volunteers for participating in this study.

Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.metabol.2014.04.005.

REFERENCES

[1] Kris-Etherton PM, Harris WS, Appel LJ. Fish consumption, fish oil, omega-3 fatty acids, and cardiovascular disease. Arterioscler Thromb Vasc Biol 2003;23(2):e20–30. [2] Kromhout D, Bosschieter EB, de Lezenne Coulander C. The inverse relation between fish consumption and 20-year mortality from coronary heart disease. N Engl J Med 1985;312 (19):1205–9. [3] Iso H, Kobayashi M, Ishihara J, et al. Intake of fish and n3 fatty acids and risk of coronary heart disease among Japanese: the Japan public health center-based (JPHC) study cohort I. Circulation 2006;113(2):195–202. [4] Daviglus ML, Stamler J, Orencia AJ, et al. Fish consumption and the 30-year risk of fatal myocardial infarction. N Engl J Med 1997;336(15):1046–53. [5] Hu FB, Cho E, Rexrode KM, et al. Fish and long-chain omega-3 fatty acid intake and risk of coronary heart disease and total mortality in diabetic women. Circulation 2003;107(14):1852–7. [6] Steinberg HO, Chaker H, Leaming R, et al. Obesity/insulin resistance is associated with endothelial dysfunction. Implications for the syndrome of insulin resistance. J Clin Invest 1996;97(11):2601–10. [7] Chan NN, Vallance P, Colhoun HM. Endothelium-dependent and -independent vascular dysfunction in type 1 diabetes: role of conventional risk factors, sex, and glycemic control. Arterioscler Thromb Vasc Biol 2003;23(6):1048–54. [8] Mullen MJ, Kharbanda RK, Cross J, et al. Heterogenous nature of flow-mediated dilatation in human conduit arteries in vivo: relevance to endothelial dysfunction in hypercholesterolemia. Circ Res 2001;88(2):145–51. [9] Ceriello A, Taboga C, Tonutti L, et al. Evidence for an independent and cumulative effect of postprandial hypertriglyceridemia and hyperglycemia on endothelial dysfunction and oxidative stress generation: effects of shortand long-term simvastatin treatment. Circulation 2002;106 (10):1211–8. [10] Goodfellow J, Bellamy MF, Ramsey MW, et al. Dietary supplementation with marine omega-3 fatty acids improve systemic large artery endothelial function in subjects with hypercholesterolemia. J Am Coll Cardiol 2000;35(2):265–70. [11] Schiano V, Laurenzano E, Brevetti G, et al. Omega-3 polyunsaturated fatty acid in peripheral arterial disease: effect on lipid pattern, disease severity, inflammation profile, and endothelial function. Clin Nutr 2008;27(2):241–7. [12] Haberka M, Mizia-Stec K, Mizia M, et al. N-3 polyunsaturated fatty acids early supplementation improves ultrasound indices of endothelial function, but not through no inhibitors in patients with acute myocardial infarction: N-3 PUFA supplementation in acute myocardial infarction. Clin Nutr 2010;30(1):79–85.

[13] Sasaki J, Miwa T, Odawara M. Administration of highly purified eicosapentaenoic acid to statin-treated diabetic patients further improves vascular function. Endocr J 2012;59 (4):297–304. [14] McVeigh GE, Brennan GM, Cohn JN, et al. Fish oil improves arterial compliance in non-insulin-dependent diabetes mellitus. Arterioscler Thromb 1994;14(9):1425–9. [15] de Roos B, Mavrommatis Y, Brouwer IA. Long-chain n-3 polyunsaturated fatty acids: new insights into mechanisms relating to inflammation and coronary heart disease. Br J Pharmacol 2009;158(2):413–28. [16] Campbell WB, Fleming I. Epoxyeicosatrienoic acids and endothelium-dependent responses. Pflugers Arch 2010;459 (6):881–95. [17] Zhang LN, Vincelette J, Chen D, et al. Inhibition of soluble epoxide hydrolase attenuates endothelial dysfunction in animal models of diabetes, obesity and hypertension. Eur J Pharmacol 2011;654(1):68–74. [18] Minuz P, Jiang H, Fava C, et al. Altered release of cytochrome p450 metabolites of arachidonic acid in renovascular disease. Hypertension 2008;51(5):1379–85. [19] Zulyniak MA, Perreault M, Gerling C, et al. Fish oil supplementation alters circulating eicosanoid concentrations in young healthy men. Metabolism 2013;62 (8):1107–13. [20] Resources Council, Science and Technology Agency 2000, standard tables of food composition in Japan. 5th revised ed. Tokyo: National printing bureau; 2000 [in Japanese]. [21] Linder L, Kiowski W, Buhler FR, et al. Indirect evidence for release of endothelium-derived relaxing factor in human forearm circulation in vivo. Blunted response in essential hypertension. Circulation 1990;81(6):1762–7. [22] Takeshita A, Imaizumi T, Ashihara T, et al. Limited maximal vasodilator capacity of forearm resistance vessels in normotensive young men with a familial predisposition to hypertension. Circ Res 1982;50(5):671–7. [23] Tagawa T, Imaizumi T, Endo T, et al. Role of nitric oxide in reactive hyperemia in human forearm vessels. Circulation 1994;90(5):2285–90. [24] Ishikado A, Morino K, Nishio Y, et al. 4-hydroxy hexenal derived from docosahexaenoic acid protects endothelial cells via nrf2 activation. PLoS One 2013;8(7):e69415. [25] Blaho VA, Buczynski MW, Brown CR, et al. Lipidomic analysis of dynamic eicosanoid responses during the induction and resolution of lyme arthritis. J Biol Chem 2009;284 (32):21599–612. [26] Rizza S, Tesauro M, Cardillo C, et al. Fish oil supplementation improves endothelial function in normoglycemic offspring of patients with type 2 diabetes. Atherosclerosis 2009;206 (2):569–74. [27] Sanders TA, Hall WL, Maniou Z, et al. Effect of low doses of longchain n-3 PUFAs on endothelial function and arterial stiffness: a randomized controlled trial. Am J Clin Nutr 2011;94(4):973–80. [28] Skulas-Ray AC, Kris-Etherton PM, Harris WS, et al. Dose– response effects of omega-3 fatty acids on triglycerides, inflammation, and endothelial function in healthy persons with moderate hypertriglyceridemia. Am J Clin Nutr 2011;93 (2):243–52. [29] Chowdhury R, Stevens S, Gorman D, et al. Association between fish consumption, long chain omega 3 fatty acids, and risk of cerebrovascular disease: systematic review and meta-analysis. BMJ 2012;345:e6698. [30] Rizos EC, Ntzani EE, Bika E, et al. Association between omega3 fatty acid supplementation and risk of major cardiovascular disease events: a systematic review and meta-analysis. Jama 2012;308(10):1024–33. [31] Roncaglioni MC, Tombesi M, Avanzini F, et al. N-3 fatty acids in patients with multiple cardiovascular risk factors. N Engl J Med 2013;368(19):1800–8.

Please cite this article as: Kondo K, et al, A fish-based diet intervention improves endothelial function in postmenopausal women with type 2 diabetes mellitus: A randomized c..., Metabolism (2014), http://dx.doi.org/10.1016/j.metabol.2014.04.005

M ET ABOL I SM CL IN I CA L A N D E XP E RI ME N TAL XX ( 2 0 14 ) XXX– X XX

[32] Bellien J, Joannides R. Epoxyeicosatrienoic acid pathway in human health and diseases. J Cardiovasc Pharmacol 2013;61 (3):188–96. [33] Elmarakby AA, Faulkner J, Al-Shabrawey M, et al. Deletion of soluble epoxide hydrolase gene improves renal endothelial function and reduces renal inflammation and injury in streptozotocin-induced type 1 diabetes. Am J Physiol Regul Integr Comp Physiol 2011;301(5):R1307–17. [34] Mavrommatis Y, Ross K, Rucklidge G, et al. Intervention with fish oil, but not with docosahexaenoic acid, results in lower levels of hepatic soluble epoxide hydrolase with time in apoe knockout mice. Br J Nutr 2010;103(1):16–24.

11

[35] Oh DY, Talukdar S, Bae EJ, et al. Gpr120 is an omega-3 fatty acid receptor mediating potent anti-inflammatory and insulin-sensitizing effects. Cell 2010;142(5):687–98. [36] Troseid M, Arnesen H, Hjerkinn EM, et al. Serum levels of interleukin-18 are reduced by diet and n-3 fatty acid intervention in elderly high-risk men. Metabolism 2009;58(11):1543–9. [37] Casos K, Zaragoza MC, Zarkovic N, et al. A fish-oil-rich diet reduces vascular oxidative stress in ApoE(−/−) mice. Free Radic Res 2010;44(7):821–9. [38] Imig JD, Hammock BD. Soluble epoxide hydrolase as a therapeutic target for cardiovascular diseases. Nat Rev Drug Discov 2009;8(10):794–805.

Please cite this article as: Kondo K, et al, A fish-based diet intervention improves endothelial function in postmenopausal women with type 2 diabetes mellitus: A randomized c..., Metabolism (2014), http://dx.doi.org/10.1016/j.metabol.2014.04.005