Associations between omega-3 fatty acids and insulin resistance and body composition in women with polycystic ovary syndrome

Background Polycystic ovary syndrome (PCOS) is strongly associated with abdominal obesity and insulin resistance and effective approaches to nutrition (e.g., omega-3 fatty acids intake) might improve the cardiometabolic risk profile. This study aimed to examine the associations of dietary and serum omega-3 fatty acids with insulin resistance (IR) and body composition among PCOS patients. Methods A total of 185 patients with PCOS were included in our analysis. Dietary information was collected through face-to-face interviews using a 102-item food frequency questionnaire (FFQ). Serum omega-3 fatty acid levels were measured with the gas chromatography method. Body composition was measured by both dual-energy X-ray absorptiometry (DXA) and bioelectrical impedance (BIA) methods. The multivariable linear regression model was applied to analyze the associations of dietary and serum omega-3 fatty acids with the levels of Homeostasis Model Assessment for Insulin Resistance (HOMA-IR) and body composition parameters among PCOS patients. Results Our results indicated that the dietary long-chain omega-3 polyunsaturated fatty acids (PUFA) intakes were negatively associated with HOMA-IR (β = –0.089, P = 0.040), fat mass (β = –0.022, P = 0.047), and body fat percentage (β = –0.026, P = 0.032). For serum biomarkers, higher total omega-3 PUFAs levels (β = –0.158, P = 0.021) and long-chain omega-3 PUFAs levels (β = –0.187, P < 0.001), particularly eicosapentaenoic acid (EPA) (β = –164, P = 0.011) and docosahexaenoic acid (DHA) (β = –0.158, P = 0.001) were also associated with decreased HOMA-IR. In addition, generally, dietary and serum long-chain omega-3 PUFA levels, DPA, and DHA levels were both positively associated with muscle mass measured by DXA; whereas serum total, long-chain and individual omega-3 PUFA levels (e.g., DPA, EPA, and DHA) were all negatively associated with fat mass and body fat percentage. These findings were further confirmed by the findings for body composition measured by the BIA method. Conclusion Higher levels of dietary and serum omega-3 PUFAs, particularly long-chain omega PUFAs (DPA and DHA), might have beneficial effects on metabolic parameters and body composition among PCOS patients.

Methods: A total of patients with PCOS were included in our analysis.Dietary information was collected through face-to-face interviews using a -item food frequency questionnaire (FFQ).Serum omega-fatty acid levels were measured with the gas chromatography method.Body composition was measured by both dual-energy X-ray absorptiometry (DXA) and bioelectrical impedance (BIA) methods.The multivariable linear regression model was applied to analyze the associations of dietary and serum omega-fatty acids with the levels of Homeostasis Model Assessment for Insulin Resistance (HOMA-IR) and body composition parameters among PCOS patients.

Results:
Our results indicated that the dietary long-chain omegapolyunsaturated fatty acids (PUFA) intakes were negatively associated with HOMA-IR (β = -.
), particularly eicosapentaenoic acid (EPA) (β = -, P = . ) and docosahexaenoic acid (DHA) (β = -., P = .) were also associated with decreased HOMA-IR.In addition, generally, dietary and serum long-chain omega-PUFA levels, DPA, and DHA levels were both positively associated with muscle mass measured by DXA; whereas serum total, long-chain and individual omega-PUFA levels (e.g., DPA, EPA, and DHA) were all negatively associated with fat mass and body fat percentage.These findings were further confirmed by the findings for body composition measured by the BIA method.

Introduction
Characterized by a series of manifestations of hyperandrogenemia, persistent anovulation, and ovarian polycystic changes (1), polycystic ovary syndrome (PCOS) is one of the most common endocrine-metabolic disorders that affects 5 to 10% of women of reproductive age (2).PCOS entails diseases such as type 2 diabetes, hypertension, cardiovascular disease, and endometrial cancer (3,4), and insulin resistance (IR) and hyperandrogenemia might be closely associated with the pathogenesis of PCOS (5).Excessive insulin level is not only a potential cause of hyperandrogenemia but also one of the high-risk factors leading to metabolic syndromes among those with PCOS (6).PCOS can also be accompanied by obesity because central obesity could exaggerate IR and lead to hyperinsulinemia (7).Seventy percentage of overweight PCOS patients were accompanied by IR, with the majority of overweight PCOS patients also being accompanied by hyperinsulinemia (8), suggesting potential relationships between body fat with PCOS.
The etiology of PCOS is not yet clear, but genetic, environmental and lifestyle factors may contribute to its development (9,10).Lifestyle intervention can effectively alleviate the pathological manifestations of abnormal reproduction and metabolism in patients with PCOS, and dietary intervention may also exert important effects on the prevention and control of PCOS and its complications (11,12).Ensuring a certain intake level of dietary polyunsaturated fatty acids (PUFAs) might help improve the clinical manifestations of dyslipidemia, impaired vascular endothelial function, and IR in PCOS patients (11).The omega-3 fatty acids and omega-6 fatty acids are mainly dietary sources of the essential PUFAs (6), but the dietary sources of omega-3 fatty acids are relatively less abundant than omega-6 fatty acids, only found in plant foods (e.g., green leafy vegetables, vegetable oils, and seeds) and deep-sea fish (6).Particularly, the role of long-chain omega-3 PUFAs (eicosapentaenoic acid, EPA; docosahexaenoic acid, DHA) in the primary prevention of cardiovascular diseases (13), metabolic syndrome (adiposity, dyslipidemia, insulin resistance, and diabetes, etc.) have been reported before (14).
A growing number of evidence have shown that omega-3 PUFAs have been used to improve lipid and insulin profiles in inflammation and obesity (15)(16)(17).Omega-3 fatty acids are involved in the regulatory process of arachidonic acid derivatives in the human body, thereby down-regulating the expression levels of transcription factors related to inflammation in cells (18).An omega-3 fatty acids treatment could effectively enhance systemic insulin sensitivity in mice by up-regulating the function of G protein-coupled receptor 120 (GPR120) (19).A metaanalysis indicated that omega-3 fatty acids supplements have also been found to exert an anti-obesity effect on humans by reducing body fat (weight, waist circumference) (20).The chemical structures and effects of omega-3 PUFAs might vary between dietary and serum sources, but rare studies investigated the effects of dietary and serum omega-3 PUFA levels on IR and body composition in PCOS patients, respectively.
Hence, the purpose of this study is to evaluate whether dietary and serum of different types of omega-3 PUFAs are associated with IR and body composition among patients with PCOS.

Assessment of dietary omega-fatty acid intakes
The assessment of omega-3 fatty acids in diet was conducted using a quantitative food frequency questionnaire (FFQ), which was adapted from the 2002 China National Nutrition and Health Survey capturing the usual intake of nutrients and major foods from Chinese (21).Based on the dietary habits of the Sichuan population, a total of 102 items were included in our FFQ.The main food items included cereals and their products, beans, vegetables, fruits, fungi and algae, nuts, livestock meat, poultry, dairy products, eggs, fish and seafood, and condiments.The Spearman correlation coefficients between the FFQ and 3-d dietary records for different fatty acid intakes were calculated among 24 local participants, and the results ranged from moderate to good (r = 0.28-0.59).The participants were interviewed through a face-to-face interview by trained researchers to recall their food consumption over the past year before PCOS diagnosis.The participants were asked to report the frequency (none, daily, weekly, monthly, and yearly) and portion size of consumption for each food item.Dietary intakes of individual omega-3 fatty acids included ALA (18:3 omega-3), EPA (20:5 omega-3), docosapentaenoic acid (DPA, 22:5 omega-3), and DHA (22:6 omega-3) were measured in this study.

Measurement of serum omega-fatty acids
According to a previously published approach (22), blood fatty acid concentrations were determined using gas chromatography.Lipid extraction and subsequent gas chromatography analysis were used to assess the amounts of serum individual fatty acid levels.The Bligh et al. (23) method was used to extract and purify the total lipid from serum.An Agilent 6890A gas chromatograph (Agilent, Palo Alto, USA) and a capillary column (SP2380, 0.25 mm × 30 m, 0.25 µm film, Supelco, Bellefonte, PA) were used to separate fatty acid methyl esters.1, 2-dinonadecanoyl-sn -glycerol-3-phosphocholine (C19:0) was regarded as an internal standard.A total of 28 fatty acids were quantified and the percentage composition of fatty acid methyl esters was calculated and normalized to 100%.The intra-assay coefficients of variation (CV) for the total omega-3 PUFA and individual omega-3 PUFA were 7.8 and 5.2∼9.7%,respectively.All assays were performed at the laboratory of the First Affiliated Hospital of Chengdu Medical College.

Measurement of body composition
In this study, we have body composition measured by both dual-energy X-ray absorptiometry and a body composition analyzer.
Dual-energy X-ray absorptiometry (DXA; GE-lunar Prodigy, Wisconsin, USA) scans were used to measure body composition.Total lean mass, total fat mass (FM), and body fat percentage (FM%) were automatically analyzed using the software V.enCORE10.50.086.According to the manufacturer's instructions, daily quality assurance scans were conducted by scanning the spine phantom.During the test, the subjects should keep quiet and prohibit strenuous or confrontational movements.The coefficient of variation was <2% for LM, FM, and FM% by duplicate scans with 18 volunteer subjects.
A body composition analyzer (InBodyS10) was based on the principle of BIA.The analyzer is an eight-electrode, multifrequency (1, 50, 100, 250, and 500 kHz), 800 µA alternating current, which measures the human body in sections and measures the resistance and reactance at the same time to obtain the proportions of muscle and fat tissues of the limbs and trunk.Inter-individual coefficients of variations were 2.8% for muscle mass and 2.1% for fat mass among 35 randomly selected duplicates.

Participant recruitment
A cross-sectional study was conducted from July 15, 2016, to July 28, 2019.Potentially eligible patients who suffered PCOS were consecutively recruited from the Department of Gynecology, the First Affiliated Hospital of Chengdu Medical College in China.The detailed information on the study design has been described in a previous publication (24).
Based on the diagnostic criteria of the Rotterdam conference in 2003 (25), those would be diagnosed with PCOS if they meet at least two of the following items: (i) oligo menorrhea (interval between two menstrual periods more than 35 days) and/or chronic anovulation (no vaginal bleeding for at least 6 months); (ii) clinical signs (hirsutism by elevated Ferriman-Gallwey score) and/or biochemical (elevated testosterone or free androgen index) hyperandrogenism; (iii) and/or polycystic ovaries on an ultrasound exam (≥12 follicles measuring 2-9 mm in diameter, or ovarian volume >10 mL in at least one ovary).We excluded participants who (i) were pregnant; (ii) had other genetic or endocrine disorders; (iii) had significant changes in dietary habits in the past 1 year; (iv) had mental or other diseases that might affect the accurate answer of the questionnaire survey; (v) reported a daily energy intake <600 or >3,500 kcal/d (26).
A total of 351 eligible patients were identified and 325 (92.6% response rate) patients were successfully interviewed for dietary habits and biological sample collection.All subjects signed informed written consent forms for inclusion before they participated in the study.The protocol was approved by the ethical committee of the First Affiliated Hospital of Chengdu Medical College (2016CYFYHEC025).

Statistical analysis
Data were expressed as means and standard deviations (SDs) for normal disquieted continuous variables, or medians and interquartile ranges (IQRs) for skewed variables.To assess the associations of the omega-3 PUFAs with HOMI-IR and body composition (LM, FM, and FM%), the multivariable linear regression test was performed and the standardized beta-coefficients and corresponding standard errors (SEs) were computed.All statistical analyses were performed using SPSS22.0.All statistical tests were two-tailed, and statistical significance was considered at P-values <0.05.
As shown in Table 2, for dietary omega-3 PUFAs, the average daily dietary intake of total omega-3 PUFAs was 1.05 ± 0.42 g/d, and long-chain omega-3 PUFAs was 44.3 ± 17.1 g/d.Among them, the daily intake of EPA (19.8 ± 9.3 mg/d) was the highest, whereas the level of ALA was the lowest (1.02 ± 0.39).With regard to serum phospholipid omega-3 PUFAs, the total serumand the long-chain omega-3 PUFA levels were 4.84 ± 1.96 and 4.81 ± 1.93, respectively.Among the categories of omega-3 PUFAs, the level of DHA (3.45 ± 1.63) was the highest, while its serum level in ALA was the lowest (0.07 ± 0.04).
omega-3 PUFAs (total-and long-chain-omega-3 PUFAs, DPA, EPA, and DHA) were negatively associated with fat mass and body fat percentage, but positively correlated with muscle mass.
Prior evidence suggested that IR in ovarian tissue might appear to damage metabolic signaling but intact mitogenic and steroidogenic activity, which could further stimulate the production of ovarian androgens and aggravate hyperandrogenemia (28).Independent of BMI, hyperinsulinemia, which is closely related to personal lifestyle including inadequate nutrition accompanied by a lack of physical exercise, is a major component of PCOS pathogenesis (7).Diet is an effective, acceptable, and safe intervention for relieving IR (29).Regular consumption of fish and/or oral supplementation of omega-3 PUFAs are both key sources of omega-3 PUFAs in diet (30).
Prior studies showed a relationship between some omega-3 PUFAs and PCOS indices, though the effects of omega-3 PUFAs on insulin metabolism remain inconclusive.For instance, a double-blinded randomized placebo-controlled trial (RCT) among 30 pairs of PCOS patients and controls aged 18-40 years old suggested that a 12-week linseed oil omega-3 PUFAs supplementation (rich in DHA + EPA) significantly reduced insulin values and HOMA-IR, but increased insulin sensitivity check index (all P-values <0.05) (17).Another RCT suggested that a combination of caloric restriction (1,200-1,500 kcal/day) and long-chain omega-3 PUFAs supplementation (DHA + EPA) significantly decreased the levels of HOMA-IR, insulin, and glucose-dependent insulinotropic polypeptide (GIP) secretion during the oral glucose tolerance test (OGTT) in obese subjects, but these effects were not seen with caloric restriction alone (31).
Yet, prior studies have yielded inconsistent effects of omega-3 PUFAs on body composition.An animal model indicated that treatment with flaxseed oil (rich in EPA and DHA) significantly reduced blood glucose and the amount of liver fat in rats fed with a high-fat diet (32).Of note, the present study did not observe the beneficial effect of ALA, but a prior cross-sectional study (33) involving 554 women aged 65-72 years found significant associations between dietary ALA and lower fat mass (β = -0.081,P = 0.034).This inconsistent result might be partly ascribed to the different portion size estimation or other biases related to subjective consumption assessment via different food record methods (3-day food record vs. 102 items food frequency questionnaire).Furthermore, given that there is no consistent evidence that omega-3 PUFA supplementation could be related to weight loss or reduced body fat mass in adults, (34) further researches are necessary to demonstrate the beneficial effect of dietary omega-3 PUFAs on body composition.
Although the exact etiology remains elusive, the beneficial effects of omega-3 PUFAs on insulin could be partly attributable to its role in insulin signaling and gene expression.Omega-3 PUFAs could protect glucose tolerance and avoid the accumulation of bioactive lipid mediators by up-regulating the mRNA expression of insulin-stimulated glucose transporter-4 (GLUT4), insulin receptor substrate-1 (IRS1), and glycogen synthase-1 (GYS1) (35).Additionally, by reducing endoplasmic reticulum stress, increasing β-oxidation of mitochondrial fatty acids and mitochondrial uncoupling, as well as limiting lipid deposits and reactive oxygen species generation, omega-3 PUFAs could further improve insulin sensitivity (16).DHA has been found to alleviate IR by regulating the pathway of the silent information regulator 1 (SIRT 1), a member of a protein family that could play a role in glucose homeostasis and IR reduction through lowering mitochondrial dysfunction (36).Additionally, IR and obesity are characteristics of chronic macrophagemediated inflammation.Through signaling G-protein-coupled receptor 120 (GPR120), an omega-3 fatty acid receptor/sensor, omega-3 PUFAs could block both toll-like receptor (TLR) and TNF-inflammatory signaling pathways, then mediate M1-M2 macrophage polarization via down-regulating the expression of inflammatory genes (e.g., IL-6, TNF-a, MCP-1) but upregulating the expression of anti-inflammatory genes (e.g., IL-10, MGL1, YM-1) in adipose tissue (37).Particularly, DHA has been found to decrease macrophage-derived inflammation and angiogenesis in adipose tissues, then reduce adipocyte size and body fat composition in middle-aged rats fed a highfat diet (HFD) (36).Considering that most studies have been undertaken using animal and cellular models, whereas limited works have been conducted among humans, more high-quality studies are warranted to further explore the potential impact of omega-3 PUFAs on metabolic function in humans.
Concerning serum omega-3 PUFAs, we found that the HOMA-IR, fat mass and body fat percentages decreased as the circulating levels of total-, and long-chain omega-3 PUFAs (total-, DPA, EPA, and DHA) increased, but did not relate to ALA.However, the present study found a null association of total dietary omega-3 PUFA and its subtypes (EPA and DHA) with muscle mass, in accord with a study by Masoud Isanejad et al. (33).The inconsistent results between dietary and circulating omega-3 PUFAs may be attributable to the difference in concentration in omega-3 PUFA concentrations in the dietary and serum phospholipids (38).Because gastrointestinal digestion and the absorption process of foods could affect the bioavailability and clinical efficacy of dietary omega-3 PUFAs, leading to the underestimation of their beneficial effects (24).Furthermore, not only genetic factors could influence the molecular response to different sources of omega-3 PUFAs, but also epigenetic modifications could obscure the clinical benefits of omega-3 PUFAs (24).
The protective roles of omega-3 PUFAs on muscle cells are under-reported.A placebo-controlled, double-blind RCT among 28 boys with Duchenne muscular dystrophy (DMD) demonstrated that the blood levels of EPA and DHA in erythrocytes in patients who were supplemented with 2.9 g/d of long-chain omega-3 PUFAs for 6 months were both significantly higher than that of the placebo group (P < 0.05), and the supplementation could significantly slow the progression of muscle loss, lowered the fat mass and IR in DMD patients (39).An RCT conducted among 124 participants found that 16 weeks of omega-3 PUFA supplementation (containing 1,650 mg/day of DHA and 150 mg/day of EPA) significantly increased the quality of muscle in lower limbs in overweight/obese postmenopausal women (40).
The n-3 PUFAs could also counter pro-atrophic mediators in the muscle cells, such as maintaining the insulin receptor number, IRS-1 tyrosine phosphorylation, phosphatidylinositol (PI) 3 ′ -kinase activity, and GLUT-4 content in rat muscle (41).A vitro study by Vigdis Aas et al. (42) found that glucose transport and oxidation (CO 2 ) in muscle cells were increased by 2-folds and the glucose transporter-1 (GLUT1) expression was increased by 2.5-folds after preincubation of myotubes with 0.6 mM EPA for 24 h, indicating that exposure to EPA could stimulate the uptake and oxidation of glucose in skeletal muscle cells from young and healthy subjects.The supplementation of omega-3 PUFAs might improve muscle function, including increasing protein synthesis, improving metabolic function, and suppressing atrophy in muscle cells (43).

Limitations
Even though this is the first study to comprehensively show that dietary and serum n-3 PUFAs might exert a positive effect on metabolism indexes and body composition among PCOS patients, several limitations need to be taken into consideration when interpreting our study.Firstly, the nature of crosssectional epidemic research does not allow to draw of a causal relationship.Secondly, dietary intakes were self-reported, and therefore prone to some degree of misclassification.However, our trained investigators explained the dietary questions with common food pictures to help them estimate their dietary consumption.Thirdly, recall bias could not be ruled out as the patients were asked to recall their dietary patterns and other living habits over the past years.Nevertheless, to minimize the recall bias, face-to-face interviews were conducted with PCOS patients at the time when they were diagnosed.Fourthly, we cannot rule out selection bias because this is a hospital-based study.Finally, the conclusions were drawn from PCOS patients in Chengdu and Sichuan province in China, so our results may not be generalizable to other populations.

Conclusion
In summary, this study found that higher levels of both dietary and serum omega-3 PUFAs, particularly longchain omega PUFAs (DPA and DHA), might exert positive effects on metabolic parameters and body composition among PCOS patients.These findings may have important practical implications because they suggest that PCOS patients might gain health benefits from dietary supplementation of omega-3 PUFAs.
TABLE Demographic characteristics and biochemical parameters among PCOS cases.
a Data was expressed as median and interquartile ranges (IQRs).b Numbers (percentages) of participants who used specific drugs.
TABLE The multivariable analysis between Omega-fatty acids and the level of HOMA-IR among PCOS patients.
TABLE Associations between Omega-fatty acids and body composition measured by DXA method among PCOS patients.