SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. SUPPLEMENTARY MATERIAL
  8. Acknowledgment
  9. Disclosure
  10. REFERENCES
  11. Supporting Information

The n-3 polyunsaturated fatty acids, especially eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), exert hypolipidemic effects and prevent development of obesity and insulin resistance in animals fed high-fat diets. We sought to determine the efficacy of α-substituted DHA derivatives as lipid-lowering, antiobesity, and antidiabetic agents. C57BL/6 mice were given a corn oil-based high-fat (35% weight/weight) diet (cHF), or cHF with 1.5% of lipids replaced with α-methyl DHA ethyl ester (Substance 1), α-ethyl DHA ethyl ester (Substance 2), α,α-di-methyl DHA ethyl ester (Substance 3), or α-thioethyl DHA ethyl ester (Substance 4) for 4 months. Plasma markers of glucose and lipid metabolism, glucose tolerance, morphology, tissue lipid content, and gene regulation were characterized. The cHF induced obesity, hyperlipidemia, impairment of glucose homeostasis, and adipose tissue inflammation. Except for Substance 3, all other substances prevented weight gain and Substance 2 exerted the strongest effect (63% of cHF-controls). Glucose intolerance was significantly prevented (∼67% of cHF) by both Substance 1 and Substance 2. Moreover, Substance 2 lowered fasting glycemia, plasma insulin, triacylglycerols, and nonesterified fatty acids (73, 9, 47, and 81% of cHF-controls, respectively). Substance 2 reduced accumulation of lipids in liver and skeletal muscle, as well as adipose tissue inflammation associated with obesity. Substance 2 also induced weight loss in dietary obese mice. In contrast to DHA administered either alone or as a component of the EPA/DHA concentrate (replacing 15% of dietary lipids), Substance 2 also reversed established glucose intolerance in obese mice. Thus, Substance 2 represents a novel compound with a promising potential in the treatment of obesity and associated metabolic disturbances.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. SUPPLEMENTARY MATERIAL
  8. Acknowledgment
  9. Disclosure
  10. REFERENCES
  11. Supporting Information

Obesity, and especially abdominal obesity, is a risk factor for insulin resistance (IR), type 2 diabetes, and cardiovascular disease. “Dysfunctional” adipose tissue, which cannot properly handle the energy surplus resulting from a sedentary lifestyle combined with excessive calorie consumption, plays a central role in obesity-associated IR and type 2 diabetes (1). In general, IR develops as a consequence of exposure of insulin-responsive tissues to elevated dietary nutrients, resulting in the accumulation of lipid-derived metabolites and impairment of interorgan communication networks that are mediated by peptide hormones and inflammatory molecules (2). Lifestyle modification therapies, such as reduced caloric intake and increased physical activity, for the obesity-associated IR and metabolic abnormalities have proved to be difficult for the general population. However, current pharmacological interventions often require the use of multiple agents and are also associated with adverse side effects, as documented in the case of the thiazolidinediones (3). Therefore, new treatment strategies are sought.

At the same time, the quality of dietary lipids is important. The n-3 long-chain polyunsaturated fatty acids (n-3 LC-PUFA), namely, eicosapentaenoic acid (EPA; 20:5n-3) and docosahexaenoic acid (DHA; 22:6n-3), which are abundant in marine fish oils, act as potent hypolipidemics in both rodents (4,5,6) and humans (7). The n-3 LC-PUFA also prevented development of obesity and IR in rodents fed high-fat diets (8). However, their beneficial effect on body weight and IR in overweight patients was only apparent when n-3 LC-PUFA were combined with a weight-loss regimen (9). In general, the insulin-sensitizing effects of n-3 LC-PUFA in subjects with impaired glucose tolerance (IGT) are minimal (10).

The hypolipidemic and antiobesity actions of n-3 LC-PUFA depend on a suppression of lipogenesis combined with increased fatty acid oxidation in the liver (4,11). However, enhanced mitochondrial biogenesis and β-oxidation in white adipose tissue (WAT) may also contribute (12). Moreover, n-3 LC-PUFA reduce WAT inflammation associated with obesity (13), while stimulating secretion of the insulin-sensitizing hormone adiponectin (6,14). The effects of n-3 LC-PUFA are largely mediated by peroxisome proliferator-activated receptors (PPARs), with PPAR-α and PPAR-δ (-β) representing the main targets for n-3 LC-PUFA (15,16); however, other factors, such as liver X receptor, hepatocyte nuclear factor-4α, and sterol-regulatory element binding protein, are also involved (for review see refs. 17,18). Decreased binding of both sterol-regulatory element binding protein–1c and nuclear factor-Y to promoters of lipogenic genes (19), as well as activation of AMP-activated protein kinase (20), may be of great importance for the liver effects. In addition to being ligands themselves, n-3 LC-PUFA might also act indirectly through their metabolites, eicosanoids, and other lipid molecules (21).

Concerning new pharmacotherapies for obesity-associated IR and type 2 diabetes, mixed agonists of various PPARs (22) or higher-affinity analogues of currently available agonists are being developed. Based on a relatively low efficiency of EPA/DHA in the treatment of IR in humans, the aim of this study was to explore several chemical DHA derivatives, substituted at the C(2)-position, as potential antiobesity and antidiabetic agents. In a mouse model of high-fat feeding, the efficiency of DHA derivatives in the prevention and reversal of obesity, IGT, dyslipidemia, WAT inflammation, and lipid accumulation in nonadipose tissues was analyzed.

Methods and Procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. SUPPLEMENTARY MATERIAL
  8. Acknowledgment
  9. Disclosure
  10. REFERENCES
  11. Supporting Information

Animals and experimental design

Male mice of either C57BL/6N genetic background (Charles River, Sulzfeld, Germany) or C57BL/6J background were maintained on a 12:12-h light-dark cycle at 22 °C (2–4 animals/cage). Mice were allowed an unrestricted access to water and a standard chow (STD; extruded Ssniff R/M-H diet; Ssniff Spezialdieten GmbH, Soest, Germany). At 3 months of age, mice were randomly assigned to a composite high-fat diet (cHF; lipids ∼35% wt/wt, mostly corn oil), while some mice were maintained on STD diet. Energy content of STD and cHF diet was 13.0 and 22.8 kJ/g, respectively (see http:www.ssniff.com, and (5)). STD diet served as a control for the obesogenic effect of the HF diet. Two experimental approaches were used (see also Figure 1):

image

Figure 1. Timeline of experiments. In the “prevention study,” 3-month-old mice were switched from a low-fat chow (STD) diet to either composite high-fat (cHF) diet, or cHF diets containing 1.5% of their lipids as DHA derivatives (Substance 1–4). In the “reversal study,” 3-month-old mice were fed a cHF diet for a period of 4 months, followed by the treatment with Substance 2 for additional 2 months. A group of mice were also subjected to a 12% caloric restriction diet (cHF-CR; relative to food intake in the cHF group). In the “reversal study,” an intraperitoneal glucose tolerance test (IPGTT) was performed 1 week before the end as well as 1 week before the initiation of dietary treatments. In a separate “reversal study” experiment, the effect of EPA/DHA concentrate (EPAX1050 TG) and DHA on ad libitum-fed mice was also studied using cHF diets containing 15% of their lipids as the tested compounds (using cHF-fed mice as controls).

Download figure to PowerPoint

  • In the “prevention study,” various DHA derivatives (Substance 1–4; see below) were admixed to the cHF diet (1.5% of dietary lipids replaced by a DHA derivative) and administered to the 3-month-old C57BL/6N mice for a period of 4 months. The DHA derivatives were tested in three separate experiments A, B, and C (Experiment A: Substance 1; Experiment B: Substance 2 and 3; Experiment C: Substance 4).

  • In the “reversal study,” obesity and IGT were induced in C57BL/6J mice by feeding cHF diet for a period of 4 months prior to the subsequent 2-month-long administration of Substance 2 admixed to the cHF diet (1.5% of dietary lipids). To analyze potential contribution of a lower food intake in the beneficial effect of Substance 2 on glucose homeostasis (as observed in the “prevention study”), food intake in a subgroup of cHF-fed mice was reduced by 12% compared with mice fed cHF diet ad libitum during the final 2 months of the “reversal study.” A separate experiment was also performed to evaluate the effect of EPAX1050 TG (Pronova BioPharma AS, Lysaker, Norway), a triglyceride concentrate of EPA (∼15%) and DHA (∼45%; EPA/DHA), as well as of pure DHA (∼99%; ethyl ester; Pronova BioPharma AS) in ad libitum-fed mice; compared with all the experiments with DHA derivatives, a tenfold higher fraction of dietary lipids (i.e., 15% of lipids) was replaced by either EPA/DHA or DHA alone (see also our previous studies (5,6) and Figure 1).

The cHF-fed mice always served as controls. Body weight and food intake were monitored weekly, while a fresh ration was offered daily. Mice were killed by cervical dislocation under diethylether anesthesia between 9–11 am, subcutaneous dorsolumbar and epididymal WAT, interscapular brown adipose tissue, liver, and skeletal muscle (musculus quadriceps femoris) were dissected and snap-freezed in liquid nitrogen. Truncal blood was collected into the EDTA-containing tubes and plasma was isolated by centrifugation at 5,000g for 10 min at 4 °C. Tissues and plasma were stored at −70 °C for future analyses. The experiments were conducted under the guidelines for the use and care of laboratory animals of the Institute of Physiology.

Chemical α-substituted DHA derivatives

DHA derivatives (for chemical structures, see Supplementary Figure S1 online) were provided by Pronova BioPharma AS, including α-methyl DHA ethyl ester (Substance 1), α-ethyl DHA ethyl ester (Substance 2), α,α-di-methyl DHA ethyl ester (Substance 3), and α-thioethyl DHA ethyl ester (Substance 4).

Preparation of DHA derivatives

The NMR spectra were recorded in CDCl3, with a Bruker Avance DPX 200 instrument. Mass spectra were recorded with a LC/MS Agilent 1100 series, with a G 1956 A mass spectrometer (electrospray, 3000 V). (University of Oslo, Norway). All reactions were performed under nitrogen or argon atmosphere.

Ethyl (all-Z)-2-methyl-4,7,10,13,16,19-docosahexaenoate (Substance 1)

In this phase, n-Butyllithium (1.6 mol/l in hexane, 228 ml, 370 mmol) was added dropwise to a stirred solution of diisopropylamine (60 ml, 420 mmol) in dry tetrahydrofuran (THF; 800 ml) at 0 °C under N2. The resulting solution was stirred at 0 °C for 30 min, cooled to −78 °C and stirred an additional 30 min before dropwise addition of ethyl (all-Z)-4,7,10,13,16,19-docosahexaenoate (100 g, 280 mmol) in dry THF (500 ml) during 2 h. The resulting solution was stirred at −78 °C for 30 min before methyl iodide (28 ml, 450 mmol) was added. The solution was allowed to reach −20 °C during 1.5 h and then poured into water (1.5 l). The resulting mixture was extracted with heptane (2 × 800 ml). The combined organic extracts were washed with 1 mol/l HCl (1 l), dried (Na2SO4) and evaporated in vacuo. The residue was purified by flash chromatography (SiO2, heptane/ethyl acetate 99:1). Yield: 50 g (48%) as a slightly yellow oil. 1H-NMR (200 MHz, CDCl3): δ 1.02 (t, J = 7.5 Hz, 3H), 1.20 (d, J = 6.8 Hz, 3H), 1.29 (t, J = 7.1 Hz, 3H), 2.0–2.6 (m, 5H), 2.8–3.0 (m, 10H), 4.17 (t, J = 7.1 Hz, 2H), 5.3–5.5 (m, 12H), MS (ESI): 393 [M+Na+]+.

Ethyl (all-Z)-2-ethyl-4,7,10,13,16,19-docosahexaenoat (Substance 2)

n-Butyllithium (1.6 mol/l in hexane, 440 ml, 670 mmol) was added dropwise to a stirred solution of diisopropylamine (111 ml, 780 mmol) in dry THF (750 ml) at 0 °C under N2. The resulting solution was cooled to −78 °C and stirred an additional 45 min before dropwise addition of ethyl (all-Z)-4,7,10,13,16,19-docosahexaenoate (200 g, 560 mmol) in dry THF (1.6 l) during 4 h. The resulting solution was stirred at −78 °C for 30 min before ethyl iodode (65 ml, 810 mmol) was added. The solution was allowed to reach −40 °C before an additional amount of ethyl iodide (5 ml, 60 mmol) was added, and finally reach −15 °C (during 3 h from −78 °C) before the mixture was poured into water. The resulting mixture was extracted with hexane (2×). The combined organic extracts were washed with 1 mol/l HCl and water, dried (Na2SO4) and evaporated in vacuo. The residue was purified by flash chromatography (SiO2, heptane/ethyl acetate 99:1–50:1). Yield: 42 g (20%) as a yellow oil. 1H-NMR (200 MHz; CDCl3): δ 0.8–1.0 (m, 6H), 1.2–1.4 (m, 4H), 1.5–1.7 (m, 2H), 2.12 (m, 2H), 2.3–2.5 (m, 2H), 2.8–3.0 (m, 10H), 4.18 (t, J = 7.1 Hz, 2H), 5.3–5.6 (m, 12H), MS (ESI): 407 [M+Na+]+.

Ethyl (all-Z)-2,2-dimethyl-4,7,10,13,16,19-docosahexaenoate (Substance 3)

n-Butyllithium (1.6 mol/l in hexane, 100 ml, 170 mmol) was added dropwise to a stirred solution of diisopropylamine (28 ml, 200 mmol) in dry THF (100 ml) at 0 °C under N2. The resulting solution was stirred at 0 °C for 30 min, cooled to −78 °C before dropwise addition of (50 g, 140 mmol) in dry THF (200 ml). The resulting solution was stirred at −78 °C for 30 min before methyl iodide (17 ml, 280 mmol) was added. The solution was allowed to reach −10 °C and then poured into water. The resulting mixture was extracted with hexane (2×). The combined organic extracts were washed with 1 mol/l HCl, dried (Na2SO4) and evaporated in vacuo. This procedure was repeated, but the crude product was used instead of ethyl (all-Z)-4,7,10,13,16,19-docosahexaenoate. The residue was purified by flash chromatography (SiO2, heptane/ethyl acetate 99:1–98:2). Yield: 32 g (59%) as a slightly yellow oil.1H-NMR (200 MHz; CDCl3): δ 1.01 (t, J = 7.5 Hz, 3H), 1.21 (s, 6H), 1.28 (t, J = 7.1 Hz, 3H), 2.08 (m, 2H), 2.34 (d, J = 6.8 Hz, 2H), 2.8–3.0 (m, 10H), 4.15 (q, J = 7.5 Hz, 2H), 5.3–5.6 (m, 12H) 13C-NMR (50 MHz; CDCl3): δ 14.7, 21.0, 25.3, 26.0, 26.1, 38.3, 42.8, 60.7, 125.8, 127.4, 128.3, 128.5, 128.6, 128.7, 129.0, 130.7, 132.4, 177.9, MS (ESI): 385 [M+H+]+.

Ethyl (all-Z)-2-thioethyl-4,7,10,13,16,19-docosahexaenoate (Substance 4)

Step 1: Synthesis of ethyl (all-Z)-2-iodo-4,7,10,13,16,19-docosahexaenoate. n-Butyllithium (1.6 mol/l in hexane, 158 ml, 253 mmol,) was added dropwise to a stirred solution of diisopropylamine (42 ml, 298 mmol) in dry THF (150 ml) at 0 °C under N2. The resulting solution was cooled to −78 °C and stirred for 35 min before dropwise addition of ethyl (all-Z)-4,7,10,13,16,19-docosahexaenoate (75 g, 210 mmol) in dry THF (300 ml). The resulting solution was stirred at −78 °C for 30 min before iodine (91 g, 359 mmol) in THF (200 ml) was added dropwise. The solution was stirred at −78 °C for 20 min, then poured into water (200 ml). The resulting mixture was extracted with heptane (300 ml). The organic extract was washed with 1 mol/l HCl (150 ml) and water (200 ml), dried (Na2SO4) and evaporated in vacuo. The residue was purified by flash chromatography (SiO2, heptane/ethyl acetate 100:1). Yield: 26 g (26%) as a yellow oil. 1H-NMR (200 MHz, CDCl3): δ 0.94 (t, J = 7.5 Hz, 3H), 1.24 (t, J = 7.1 Hz, 3H), 2.04 (quint, J = 7.1 Hz, 2H), 2.69–2.84 (m, 12 H), 4.17 (q, J = 7.1 Hz, 2H), 4.22 (t, J = 7.9 Hz, 1H), 5.24–5.49 (m, 12 H), 13C NMR (50 MHz, CDCl3): δ 13.7, 14.2, 25.5, 26.0 (2 signals), 25.8, 34.0, 61.7, 126.1, 127.0, 127.4, 127.8, 127.9, 128.0, 128.2, 128.5, 128.5, 131.6, 131.9, 170.9 (4 signals hidden), MS (ESI): 505 [M+Na+]+.

Step 2: Synthesis of ethyl (all-Z)-2-thioethyl-4,7,10,13,16,19-docosahexaenoate. Sodium ethyl thiolate (2.1 g, 25 mmol) was added to a solution of ethyl (all-Z)-2-iodo-4,7,10,13,16,19-docosahexaenoate (11.0 g, 23 mmol) in THF (100 ml) at 0 °C under N2. The resulting mixture was stirred at 0 °C for 1 h. 1 mol/l HCl was added followed by heptane. The phases were separated and the organic phase was washed with water (2×), dried (Na2SO4) and evaporated in vacuo. The residue was purified by flash chromatography (SiO2, heptane/ethyl acetate 30:1. Yield: 7.3 g (76 %) as a pale yellow oil. 1H-NMR (200 MHz, CDCl3): δ 1.1–1.3 (m, 9H), 2.05 (m, 2H), 2.3–2.7 (m, 4H), 2.7–2.9 (m, 10H), 3.25 (m, 1H), 4.17 (q, J = 7.1 Hz, 2H), 5.3–5.5 (m, 12H), MS (ESI): 439 [M+Na+]+.

Plasma metabolites, hormones, and enzymes

Blood glucose was measured using calibrated glucometers OneTouch Ultra (Life Scan, Milpitas, CA). Plasma triglycerides, cholesterol, aspartate aminotransferase, alanine aminotransferase, and creatine kinase were measured using a clinical analyzer and enzymatic kits from Roche Diagnostics (Mannheim, Germany). Nonesterified fatty acids (NEFA) were measured by a kit from Waco Chemicals (Neuss, Germany). Plasma insulin levels were determined by the Sensitive Rat Insulin RIA Kit (LINCO Research, St Charles, MO), and total immunoreactive adiponectin and leptin were measured by a 2-site ELISA (R&D Systems, Minneapolis, MN; (6)).

Glucose tolerance test

An intraperitoneal glucose tolerance test was performed after an overnight fasting (15–16 h). Blood glucose was assessed by tail bleeds at the baseline (fasting blood glucose; FBG) and after the injection of d-glucose (1 g/kg body weight). In the “reversal study,” intraperitoneal glucose tolerance test was performed 1 week before the start (baseline) and 1 week before the end (final) of the 2-month-long treatment period. Results were expressed either as area under the curve (AUC) for glucose or as a change in blood glucose levels (Δ Blood glucose; final-baseline), derived from the glycemic curves of intraperitoneal glucose tolerance test measured before and after the treatment. In the latter case, the greater was the beneficial effect on glucose tolerance after the treatment, the greater was the negative deviation of a glycemic curve from the baseline.

RNA extraction and real-time quantitative PCR analysis

Total RNA was isolated from samples of WAT, liver, and skeletal muscle stored in RNAlater Solution (Ambion, Austin, TX) by using TRIzol Reagent (Invitrogen, Carlsbad, CA). Muscle samples were grinded under liquid nitrogen prior to homogenization. A quantity of 0.5 µg of total RNA was reverse transcribed to cDNA, and gene expression was analyzed by real-time PCR, using the LightCycler Instrument (Roche Diagnostics, Mannheim, Germany) as before (6). Oligonucleotide primers, described in Supplementary Table S1 online, were designed using Lasergene software (DNAStar, Madison, WI). Gene expression data were expressed as a percentage of the cHF-fed controls.

Tissue triglycerides

Tissue fragments (∼50 mg) were digested in 0.15 ml of 3 mol/l alcoholic KOH and the resulting homogenates were diluted tenfold with H2O. After neutralization by 2.5 N HClO4, deliberated glycerol was assayed in supernatants (4 µl) by Free Glycerol Reagent (Sigma-Aldrich, Prague, Czech Republic). Tissue triglyceride concentration was calculated relative to a glycerol standard (1 mg/ml; Sigma-Aldrich, Prague, Czech Republic) using a ratio of 1:10 for molecular weights of glycerol:triglyceride.

Light microscopy and immunohistochemical analysis

Samples of epididymal WAT and liver were fixed in 4% formaldehyde, embedded in paraffin and cut into 5 µm-sections. The liver sections were stained by hematoxylin-eosine, while the sections of epididymal fat were processed to detect a macrophage marker, MAC-2/galectin-3, by the use of specific antibodies (23). Digital images were captured using Olympus AX70 light microscope and a DP 70 camera (Olympus, Tokyo, Japan). Adipocyte morphometry was performed using a Lucia IMAGE version 4.81 (Laboratory Imaging, Prague, Czech Republic).

Statistical analysis

Data are presented as means ± s.e. Data were analyzed by a one-way ANOVA or two-way Repeated Measures ANOVA (only the analysis of glucose tolerance in the “reversal study”) using SigmaStat statistical software. Logarithmic transformation was used to stabilize variance in cells when necessary. The Holm-Sidak test for multiple comparisons was used. Threshold of significance was defined at P < 0.05.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. SUPPLEMENTARY MATERIAL
  8. Acknowledgment
  9. Disclosure
  10. REFERENCES
  11. Supporting Information

Reduction of body weight and adiposity by DHA derivatives

The effect of four different DHA derivatives (Substance 1–4) on body weight, food intake, and adiposity in the “prevention study” is summarized in Table 1. Compared to the cHF-fed mice, all DHA derivatives except Substance 3 reduced weight gain and Substance 2 exerted the strongest effects. Mice fed Substance 2 had a reduced food intake, and the feeding efficiency was decreased by ∼70% in these animals. Substance 2 reduced the weight of subcutaneous and epididymal WAT by 73 and 42%, respectively, while the remaining DHA derivatives had less effect on adiposity. The dramatic effect of Substance 2 could be partly explained by a reduction of cellularity of WAT, as reflected by a 66% decrease in the DNA content of epididymal WAT. Furthermore, Substance 2 and 3 also decreased the weight of interscapular brown adipose tissue.

Table 1.  The effect of DHA derivatives on energy balance and tissue parameters in the “prevention study”
inline image

The effect of the most potent DHA derivative, Substance 2, was also examined in dietary obese mice in the “reversal study” (Table 2). Administration of Substance 2 for a period of 2 months resulted in a net weight loss in obese mice. This effect could be explained, at least in part, by a decreased weight of fat depots, which was accompanied by a reduction of cellularity (DNA data in Table 2). In line with the “prevention study,” the average food intake in Substance 2-fed mice was decreased by 9%. A separate experiment revealed that the reduction in body weight gain induced by either DHA alone or EPA/DHA concentrate (admixed at a tenfold higher dose to cHF diet as compared to Substance 2; that is, replacing 15% vs. 1.5% of dietary lipids) was relatively mild, while no effect on food intake was observed (Supplementary Table S2 online).

Table 2.  The effect of Substance 2 in the “reversal study
inline image

Prevention of adipocyte hypertrophy and macrophage infiltration of WAT by Substance 2

In the “prevention study,” histological analysis of epididymal WAT (Figure 2a, b, c) revealed adipocyte hypertrophy in the cHF-fed mice, resulting in a approximately twofold increase in the mean cell size. This effect was completely prevented by Substance 2 (Figure 2d). Moreover, Substance 2 also completely prevented obesity-associated macrophage infiltration of WAT, as revealed by immunohistochemical detection of Mac-2 (Figure 2; white arrows). Macrophages aggregate in crown-like structures surrounding individual adipocytes (23). While the density of crown-like structures was ∼77-fold higher in cHF-fed compared with STD-fed mice, Substance 2 completely prevented this effect (Figure 2e). Moreover, in epididymal WAT Substance 2 reduced mRNA levels of CD68 and monocyte chemoattractant protein-1 (MCP-1), two factors that are closely linked to macrophage function, by 91 and 56%, respectively (Supplementary Table S4 online). In the “reversal study,” Substance 2 reduced the accumulation of macrophages in epididymal WAT by 65% (not shown) and expression of CD68 and MCP-1 by 32 and 50%, respectively (Supplementary Table S4 online). Thus, Substance 2 completely prevents and even partially reverses adipocyte hypertrophy and macrophage infiltration of WAT, induced by the obesogenic cHF diet.

image

Figure 2. The effect of Substance 2 on adipose tissue morphology and macrophage infiltration in the “prevention study.” The amount of MAC-2 immunoreactive macrophages (brownish color) was analyzed in epididymal fat. Sections were counterstained with hematoxylin-eosin. (a) Mice fed a low-fat chow (STD) diet. (b) Composite high-fat (cHF) diet. (c) Substance 2. Arrows indicate crown-like structures (CLS) surrounding individual adipocytes, where the majority of macrophages are lozalized. Bar = 50 µm. (d) Size of adipocytes. (e) CLS density. The morphometry data are based on >1,000 cells taken randomly from 5 different areas per animal (n = 3). *P < 0.05 vs. cHF diet (ANOVA).

Download figure to PowerPoint

Given the strong effect of Substance 2 on adiposity and obesity-associated inflammation of WAT, two major adipokines, leptin and adiponectin, were evaluated after 2 months of treatment in the “prevention study.” Compared with cHF-fed mice, plasma leptin levels were strongly reduced by Substance 2 (4.4 ± 0.3 vs. 86.0 ± 6.2 ng/ml; P < 0.00001) and reached the levels observed in the STD-fed mice (6.9 ± 0.9 ng/ml). Plasma adiponectin levels were also slightly reduced (Substance 2, 7.1 ± 0.6 vs. cHF, 9.3 ± 0.5; P < 0.05) and were similar to those observed in STD-fed mice (6.9 ± 0.4 ng/ml).

The effects on liver and muscle

In both “prevention” (Table 1) and “reversal” (Table 2) study, cHF diet significantly increased liver weight. Compared to cHF-fed mice, Substance 2 increased the liver weight in the “reversal study.” Importantly, in both studies, Substance 2 reduced the accumulation of triglycerides in liver and skeletal muscle, normally induced after cHF feeding. The induction of liver steatosis by cHF diet and a decrease of triglyceride accumulation by Substance 2 in the “reversal study” were also confirmed by light microscopy (Figure 3). Importantly, livers from mice treated with Substance 2 contained small parenchymal cells of normal morphology, a picture compatible with active regeneration and extensive remodelling of the tissue.

image

Figure 3. The effect of Substance 2 on liver histology and gene expression in the “reversal study.” Liver morphology assessed by hematoxyline-eosin staining of liver sections from mice fed (a) low-fat chow (STD) diet, (b) composite high-fat (cHF) diet, and (c) Substance 2. Visualization of neutral lipids (in red) by Sudan III staining in the liver of mice fed (d) cHF diet and (e) Substance 2. Nuclei (in blue) were counterstained by hematoxyline. Bar = 50 µm. (f) The expression of genes involved in fatty acid oxidation in the liver. Values represent means ± s.e. (n = 4–8). AOX, acyl-CoA oxidase; CPT-1α, carnitine palmitoyltransferase-1α. *P < 0.05 vs. cHF diet (ANOVA).

Download figure to PowerPoint

While aspartate aminotransferase levels in plasma of Substance-2-treated mice were unchanged (not shown), alanine aminotransferase levels were increased (Substance 2, 11.33 ± 2.61 vs. cHF, 1.04 ± 0.11 µkat/l; P < 0.01) in the “prevention study.” Plasma levels of creatine kinase, a marker of muscle cell integrity, were unchanged (not shown). In the “reversal study,” Substance 2 increased both aspartate aminotransferase (Substance 2, 12.07 ± 2.30 vs. cHF, 4.98 ± 0.55 µkat/l; P < 0.05) and alanine aminotransferase (Substance 2, 11.19 ± 2.45 vs. cHF, 0.97 ± 0.17 µkat/l; P < 0.05). The elevated plasma levels of hepatic markers in Substance-2-treated mice, especially in the “reversal study,” can be explained by intense regeneration of liver parenchyma accompanying removal of lipid-engorged hepatocytes.

The expression of genes involved in fatty acid oxidation in the liver, such as acyl-CoA oxidase-1 (Aox-1) and carnitine palmitoyltransferase-1α (Cpt-1α), was higher in animals fed cHF compared with STD diet. This was evident in both the “prevention” and the “reversal” study. These genes were even more upregulated by Substance 2. There was a stronger induction of the peroxisomal (Aox-1) than the mitochondrial (Cpt-1α) pathway. In both cases, the induction was stronger in the “reversal” than in the “prevention” study (Supplementary Table S4 online and Figure 3f). In a separate “reversal study” experiment, EPA/DHA concentrate and to a lesser extent also DHA alone upregulated Cpt-1α (but not Aox-1) mRNA levels. However, these changes were relatively small compared to the effects of Substance 2 admixed at a tenfold lower dose to cHF diet. Strong induction of Aox-1 and Cpt-1α expression by Substance 2 correlated well with a marked increase in the expression of their regulatory transcription factor PPAR-α (Supplementary Table S4 online).

n-3 LC-PUFA are known to decrease expression of lipogenic genes like stearoyl-coenzyme A desaturase-1 (Scd-1), as well as other genes (Spot 14 and farnesyl diphosphate synthase (Fdps); (19)). Expression of Scd-1, Spot 14, and Fdps was downregulated by cHF when compared with STD diet. In the “reversal study,” expression of Scd-1 and Spot 14 was decreased by both EPA/DHA and by DHA alone. In contrast, expression of Scd-1 and Fdps was markedly induced by Substance 2, namely in the “reversal study” (Supplementary Table S4 online).

In the skeletal muscle, Substance 2 exerted negligible effects on gene expression except for a downregulation of Scd-1 (Supplementary Table S4 online).

Beneficial effects of DHA derivatives on systemic markers of lipid metabolism

In the “prevention study” study, Substance 2 reduced plasma levels of total triglycerides, NEFA, and total cholesterol by 53, 19, and 20%, respectively (Table 3). The other substances exerted less pronounced lipid-lowering effects, while Substance 1 even increased plasma NEFA by 34%. In the “reversal study,” Substance 2 lowered plasma triglycerides, NEFA, and cholesterol levels by 55, 24, and 25%, respectively (Table 2). A separate experiment revealed that DHA alone admixed at a tenfold higher dose to cHF diet compared to Substance 2 had no effect on plasma triglycerides, while EPA/DHA concentrate reduced plasma triglycerides by 40% (Supplementary Table S2 online), that is, similarly to Substance 2 admixed at a tenfold lower dose (see above).

Table 3.  The effect of DHA derivatives on plasma markers of lipid metabolism in the “prevention study
inline image

Beneficial effects of DHA derivatives on glucose homeostasis

In the “prevention study,” FBG and glucose tolerance were markedly impaired by cHF feeding, while Substance 2 improved FBG and both Substance 1 and Substance 2 improved glucose tolerance (Figure 4). cHF-fed mice also demonstrated hyperinsulinemia, while Substance 2 and Substance 3 exerted protective effects (Figure 4d).

image

Figure 4. The effect of DHA derivatives on glucose homeostasis in the “prevention study.” Mice were fed either a low-fat chow (STD), composite high-fat (cHF) diet, or cHF diet in which 1.5% of lipids was replaced by various DHA derivatives (S1, Substance 1; S2, Substance 2; S3, Substance 3; S4, Substance 4). (a) Intraperitoneal glucose tolerance test (IPGTT; glycemic curves for mice fed the STD, cHF, and Substance 2 diets are shown). *P < 0.05 vs. other groups (ANOVA). (b) Fasting blood glucose (FBG) corresponding to baseline blood glucose levels from IPGTT. (c) Total area under the curve for glucose (AUCglucose) derived from IPGTT data. (d) Plasma insulin in ad libitum-fed mice at the time of killing, that is, 1 week after IPGTT. Data are expressed as percentages of the control cHF diet and represent means ± s.e. (STD, n = 11; cHF, n = 7–13; Substances, n = 8). *P < 0.05 vs. cHF diet (ANOVA).

Download figure to PowerPoint

Substance 2 also lowered FBG and plasma insulin in dietary obese mice in the “reversal study” (Table 2). Importantly, glucose tolerance was also improved by Substance 2 (Figure 5). In contrast, neither EPA/DHA nor DHA alone (both admixed at a tenfold higher dose to cHF diet as compared to Substance 2) exerted significant effects on FBG and glucose tolerance (total AUC glucose) in dietary obese mice in the “reversal study,” although there was a trend for EPA/DHA to improve glucose tolerance (Supplementary Table S2 online).

image

Figure 5. Reversal of impaired glucose tolerance in dietary obese mice by Substance 2 in the “reversal study.” Glucose tolerance was assessed by intraperitoneal glucose tolerance test (IPGTT) performed 1 week before the start and 1 week before the end of 2-month-long treatment in mice fed ad libitum either a composite high-fat (cHF) diet or cHF diet, containig DHA derivative (Substance 2). Calorie-restricted cHF-fed mice (cHF-CR; 12% restriction) were also analyzed. (a) Glycemic curves, representing a change in blood glucose levels during IPGTT measured before and after the treatment. (b) Total area under the curve for glucose (AUCglucose) derived from IPGTT data obtained before and after the treatment. Values represent means ± s.e. (cHF, n = 20; Substance 2, n = 21; cHF-CR, n = 12). *P < 0.05 before vs. after the treatment (two-way repeated measures ANOVA).

Download figure to PowerPoint

Because Substance 2 reduced food intake by 12% during the “prevention” study, the effect of a 12% calorie restriction on glucose tolerance was analyzed in the “reversal study.” Except for increasing FBG, calorie restriction did not have any significant effect on glycemia during intraperitoneal glucose tolerance test (Figure 5a) or on total AUC (Figure 5b) in obese cHF diet-fed mice. In contrast, calorie restriction decreased body weight by 1.25 ± 0.62 g as compared to a weight loss of 6.89 ± 1.23 g induced by Substance 2 (P < 0.001).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. SUPPLEMENTARY MATERIAL
  8. Acknowledgment
  9. Disclosure
  10. REFERENCES
  11. Supporting Information

We report for the first time the metabolic effects of DHA derivatives substituted at the C(2)-position of the molecule. We show that replacement of 1.5% of dietary lipids by various DHA derivatives affected the development of diet-induced obesity and associated metabolic traits in C57BL/6 mice fed a high-fat diet. Substance 2 (α-ethyl DHA ethyl ester) completely prevented and even partially reversed the development of obesity, fat accumulation, IGT, dyslipidemia, and WAT inflammation. Therefore, besides the general characterization of various DHA derivatives, our study was largely focused on a detailed description of the action of Substance 2.

In agreement with its effect on body weight and adiposity, Substance 2 profoundly affected WAT properties. Similar to the effect of EPA/DHA (5,13), Substance 2 reduced (i) tissue cellularity, (ii) the size of adipocytes, and (iii) macrophage infiltration of WAT. Small adipocytes are more insulin sensitive and less lipolytic, while releasing less inflammatory cytokines, including MCP-1 (24). The reduction of macrophage infiltration should have beneficial systemic effects, as macrophages represent an additional source of proinflammatory cytokines, which induce IR and contribute to a state of chronic low-level inflammation in obesity (25). Similar to the effects of n-3 LC-PUFA (5), Substance 2 also partially prevented downregulation of Glut4 in WAT, otherwise induced by high-fat diet. WAT is also a source of antiinflammatory and insulin-sensitizing adipokines, leptin and adiponectin. Circulating leptin levels reflect adiposity and obesity is associated with leptin resistance (26,27). Substance 2 markedly decreased plasma leptin levels, reflecting the reduction of adiposity. However, downregulation of Scd-1 in skeletal muscle of Substance-2-treated mice could also imply improved muscle leptin sensitivity and elevated fatty acid oxidation (28). In contrast to the induction of adiponectin by EPA/DHA (6), Substance 2 decreased plasma adiponectin by ∼25% despite dramatically improving glucose tolerance. However, similar plasma adiponectin levels in Substance-2-treated mice and mice fed a low-fat STD diet suggest that plasma adiponectin might not be the best predictor of the metabolic state in this mouse model.

A relatively strong suppression of insulin levels by Substance 2 could hardly represent an indirect effect secondary to improvements in insulin sensitivity. It has been shown that n-3 LC-PUFA reversed glucose-stimulated insulin hypersecretion, normally induced by obesogenic diet, in rat islets (29,30). Therefore, in addition to its effects in other tissues, Substance 2 might act directly on pancreatic β-cells via reduction of insulin secretion. However, further studies are required to clarify this issue.

As published by others (31), DHA derivatives with a hydrophilic substituent at the C(4)-position could lower glucose levels in animal models of diabetes. However, they did not lower blood triglycerides. In contrast, Substance 2 not only prevented and even partially reversed IGT, but it also lowered plasma triglycerides, NEFA, and cholesterol levels. Substance 2 strongly reduced the accumulation of triglycerides in both liver as well as skeletal muscle, resembling the effects of n-3 LC-PUFA (30). The reduced lipid accumulation in the liver and muscle might be a major mechanism, by which Substance 2 counteracted development of IGT.

Substance 2, and to a lesser extent also EPA/DHA concentrate and DHA alone, upregulated Ppar-α and its target genes Aox-1 and Cpt-1α in the liver, documenting induction of lipid catabolism and suggesting that Substance 2 acted as a potent PPAR-α agonist (18). In agreement with the known induction of lipogenic genes by pharmacological stimulation of PPAR-α in mouse liver (32), lipogenic Scd-1 and Fdps were strongly induced by Substance 2. Importantly, hepatic Scd-1 gene expression has been shown to be upregulated by PPARs directly, through a mechanism distinct from the regulation of this gene by polyunsaturated fatty acids (33). A set of sterol-regulatory element binding protein–1c target genes (Scd-1, Spot 14, and Fdps) was downregulated by EPA/DHA concentrate or DHA alone as expected (19).

The induction of Scd-1 and other lipogenic genes by Substance 2 seems to be liver specific because it did not occur in skeletal muscle, where Substance 2 even downregulated Scd-1 expression. The simultaneous stimulation of in situ lipogenesis and lipid oxidation by Substance 2 in the liver suggests induction of futile substrate cycling, which may be responsible for the reduced accumulation of triglycerides in the tissue and possibly also for decreased feeding efficiency of Substance-2-treated mice.

Besides increasing the expression of fatty acid oxidation genes, Substance 2 also lowered FBG in both “prevention” and “reversal” study, suggesting a reduction in hepatic glucose production and gluconeogenesis. This finding is in disagreement with previously published reports, linking increased oxidation of fatty acids to the activation of gluconeogenesis in the liver (34,35). However, the coordinated regulation of these metabolic pathways by physiological stimuli such as fasting (35) might be dramatically different from the situation, when hepatic fatty acid oxidation is stimulated by pharmacological activation of PPAR-α. In fact, improved liver insulin sensitivity in response to PPAR-α agonist treatment, as evidenced by lower endogenous glucose production, has been already observed before (36).

The efficacy of Substance 2 is striking because the dose used in our experiments (1.5% of dietary lipids replaced by the DHA derivative) was approximately six- to tenfold lower compared either with the dose of EPA/DHA concentrate (or DHA alone) also used in this study or with other animal studies in which significant effects of EPA and DHA on body weight, adiposity, and plasma lipids (5,6), or IR (8) were observed. Substance 2 reduced body weight gain in association with a reduced feeding efficiency; however, food intake was also slightly reduced. Calorie restriction itself exerts beneficial effects on lipid and glucose metabolism (6,37). However, the “reversal study,” which also included the calorie-restricted cHF diet-fed mice, indicated that Substance 2 reversed glucose intolerance independently of the reduction in food intake. Nevertheless, the contribution of body weight change to improved glucose tolerance in Substance-2-treated mice could not be directly estimated. In contrast to Substance 2 and in agreement with human studies (10), EPA/DHA or DHA alone could not reverse established glucose intolerance.

In summary, among the four DHA derivatives tested, Substance 2 (α-ethyl DHA ethyl ester) appeared to exhibit a similar range of beneficial effects on obesity and associated metabolic traits as naturally occurring n-3 LC-PUFA, but with a higher efficacy. Therefore, this compound could qualify as a novel drug for the treatment of obesity, dyslipidemia, and insulin resistance.

Acknowledgment

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. SUPPLEMENTARY MATERIAL
  8. Acknowledgment
  9. Disclosure
  10. REFERENCES
  11. Supporting Information

This study was supported by the grants from the Czech Science Foundation (303/08/0664 and 303/07/0708), Pronova BioPharma AS (Lysaker, Norway), and MITOFOOD (COST Action FA0602). We acknowledge Synthetica AS (Forskningsparken, Oslo) for the synthesis of the DHA derivatives. We also thank Saverio Cinti (University of Ancona, Italy) for advice concerning histological analysis.

Disclosure

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. SUPPLEMENTARY MATERIAL
  8. Acknowledgment
  9. Disclosure
  10. REFERENCES
  11. Supporting Information

Dr Bryhn, Dr Berge, and Dr Holmeide were employees of Pronova BioPharma AS at the time of this study.

REFERENCES

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. SUPPLEMENTARY MATERIAL
  8. Acknowledgment
  9. Disclosure
  10. REFERENCES
  11. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. SUPPLEMENTARY MATERIAL
  8. Acknowledgment
  9. Disclosure
  10. REFERENCES
  11. Supporting Information

Supporting Information

FilenameFormatSizeDescription
oby_1023_sm_oby2008602x1.doc117KSupporting info item
oby_1023_sm_oby2008602x2.doc47KSupporting info item
oby_1023_sm_oby2008602x3.doc51KSupporting info item
oby_1023_sm_oby2008602x4.doc50KSupporting info item
oby_1023_sm_oby2008602x5.doc81KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.