Scholarly article on topic 'Erythrocyte omega-3 polyunsaturated fatty acid levels are associated with biomarkers of inflammation in older Australians'

Erythrocyte omega-3 polyunsaturated fatty acid levels are associated with biomarkers of inflammation in older Australians Academic research paper on "Clinical medicine"

CC BY-NC-ND
0
0
Share paper
OECD Field of science
Keywords
{"Omega-3 polyunsaturated fatty acids" / "Omega-3 index" / Inflammation / "C-reactive protein" / Fibrinogen / "White blood cells" / Monocytes / Neutrophils / Ageing}

Abstract of research paper on Clinical medicine, author of scientific article — Melinda Olliver, Martin Veysey, Mark Lucock, Suzanne Niblett, Katrina King, et al.

Abstract Background Elevated levels of pro-inflammatory mediators heighten the risk of developing or aggravating a spectrum of chronic diseases and are a strong predictor of mortality in elderly cohorts. Omega-3 polyunsaturated fatty acids (n-3PUFA), including eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), are known to possess anti-inflammatory properties. However, the relationship between erythrocyte membrane n-3PUFA and inflammation biomarkers has not been well established. Objective This study aimed to determine if n-3PUFA status, together with the omega-3 index (O3I, erythrocyte membrane % EPA plus DHA), is associated with pro-inflammatory mediators in older Australians. Methods The study was a cross-sectional analysis of randomly selected older men and women aged ≥65 years (n = 620) recruited from the Central Coast of NSW, Australia. Fasted blood samples were analysed for C-reactive protein (CRP), fibrinogen and full blood count using standardised laboratory methods. The fatty acid composition of erythrocyte membranes was analysed via gas chromatography to determine n-3PUFA levels. The relationships between n-3PUFA and inflammatory mediators were evaluated in multivariate regression models after adjusting for known inflammatory confounders. Results After excluding participants who had an inflammatory disease, CRP levels >10 mg/L, or who were taking anti-inflammatory medications or n-3PUFA supplements, 126 participants (age 77.6 ± 7.3 years; females, 46%) were included in the analysis. After multivariate adjustments, O3I was inversely associated with CRP (β = −0.209, p < 0.05) and monocyte cell counts (β = −0.205, p < 0.05), and total n-3PUFA was inversely related to WBC (β = −0.238, p < 0.05), neutrophils (β = −0.212, p < 0.05) and monocytes (β = −0.246, p < 0.05). However no association between fibrinogen and O3I or total n-3PUFA was detected. Conclusions This study demonstrated a negative association between O3I and biomarkers of inflammation in an older population. The findings support a potential role for n-3PUFA supplementation in the management of inflammatory diseases.

Academic research paper on topic "Erythrocyte omega-3 polyunsaturated fatty acid levels are associated with biomarkers of inflammation in older Australians"

Accepted Manuscript

Erythrocyte omega-3 polyunsaturated fatty acid levels are associated with biomarkers of inflammation in older Australians

Melinda Olliver, Martin Veysey, Mark Lucock, Suzanne Niblett, Katrina King, Lesley MacDonald-Wicks, Professor. Manohar L. Garg

PII: S2352-3859(15)30019-0

DOI: 10.1016/j.jnim.2016.03.002

Reference: JNIM 20

To appear in: Journal of Nutrition & Intermediary Metabolism

Received Date: 30 November 2015

Revised Date: 16 March 2016

Accepted Date: 18 March 2016

Please cite this article as: Olliver M, Veysey M, Lucock M, Niblett S, King K, MacDonald-Wicks L, Garg ML, Erythrocyte omega-3 polyunsaturated fatty acid levels are associated with biomarkers of inflammation in older Australians, Journal of Nutrition & Intermediary Metabolism (2016), doi: 10.1016/ j.jnim.2016.03.002.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Erythrocyte omega-3 polyunsaturated fatty acid levels are associated with biomarkers of inflammation in older Australians

Melinda Olliver1, Martin Veysey2, Mark Lucock3, Suzanne Niblett2, Katrina King2, Lesley MacDonald-Wicks4, Manohar L Garg1*

1Nutraceuticals Research Group, School of Biomedical Sciences & Pharmacy; 2School of Medicine & Public Health; 3School of Environmental & Life Sciences; 4School of Health Sciences, University of Newcastle, NSW, Australia

Correspondence to:

Professor Manohar Garg 305C Medical Science Building University of Newcastle Callaghan, NSW 2308 Australia

E-Mail: manohar. garg@newcastle. edu. au

Keywords: omega-3 polyunsaturated fatty acids; omega-3 index; inflammation; C-reactive protein; fibrinogen; white blood cells; monocytes; neutrophils; ageing

Running title: Erythrocyte omega-3PUFA and inflammation in older people

ABSTRACT

Background: Elevated levels of pro-inflammatory mediators heighten the risk of developing or aggravating a spectrum of chronic diseases and are a strong predictor of mortality in elderly cohorts. Omega-3 polyunsaturated fatty acids (n-3PUFA), including eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), are known to possess anti-inflammatory properties. However, the relationship between erythrocyte membrane n-3PUFA and inflammation biomarkers has not been well established.

Objective: This study aimed to determine if n-3PUFA status, together with the omega-3 index (O3I, erythrocyte membrane % EPA plus DHA), is associated with pro-inflammatory mediators in older Australians.

Methods: The study was a cross-sectional analysis of randomly selected older men and women aged >65 years (n=620) recruited from the Central Coast of NSW, Australia. Fasted blood samples were analysed for C-reactive protein (CRP), fibrinogen and full blood count using standardised laboratory methods. The fatty acid composition of erythrocyte membranes was analysed via gas chromatography to determine n-3PUFA levels. The relationships between n-3PUFA and inflammatory mediators were evaluated in multivariate regression models after adjusting for known inflammatory confounders.

Results: After excluding participants who had an inflammatory disease, CRP levels >10mg/L, or who were taking anti-inflammatory medications or n-3PUFA supplements, 126 participants (age 77.6 ± 7.3 years; females, 46%) were included in the analysis. After multivariate adjustments, O3I was inversely associated with CRP (P = -0.209, p<0.05) and monocyte cell counts (P = -0.205, p<0.05), and total n-3PUFA was inversely related to WBC (P = -0.238, p<0.05), neutrophils (P = -0.212, p<0.05) and monocytes (P = -0.246, p<0.05). However no association between fibrinogen and O3I or total n-3PUFA was detected.

25 Conclusions: This study demonstrated a negative association between O3I and biomarkers of

26 inflammation in an older population. The findings support a potential role for n-3PUFA

27 supplementation in the management of inflammatory diseases.

INTRODUCTION

Pro-inflammatory mediators (including cytokines and eicosanoids) have been shown to be elevated in elderly people by as much as two- to four-fold, when compared to a younger population (1). Sustained low grade inflammation indicated by increased circulating levels of C-reactive protein (CRP) is a strong predictor of mortality in elderly cohorts irrespective of concurrent morbidities or other known risk factors such as smoking or high blood pressure (1). The inflammatory process is designed to help fight infection, remove harmful substances and repair damaged tissue and organ systems (2). Although this process is generally protective, failure to resolve the inflammatory response can result in chronic inflammation, which may lead to the development and progression of a spectrum of chronic diseases including coronary vascular disease, Alzheimer's disease, depression, rheumatoid arthritis, cancer, sarcopenia and reduced physical functioning (1, 3-5).

Elevated levels of acute phase proteins, C-reactive protein (CRP) and fibrinogen are reliable markers of sustained low grade systemic inflammation (6-8). CRP and fibrinogen are released by hepatocytes in response to elevated cytokine levels during the acute inflammatory response. Fibrinogen is also a principal protein involved in blood clotting (7).

Omega-3 polyunsaturated fatty acids (n-3PUFA) of marine origin have attracted attention due to their anti-inflammatory effects (9). The proposed mechanisms by which n-3PUFA mediates these effects include preferential incorporation of n-3PUFA into phospholipids resulting in the production of 3-series eicosanoids with lower biological activity, production of inflammation resolving lipid mediators (resolvins and protectins) and altered cytokine gene expression (2).

Most epidemiological and experimental studies have shown an association between higher plasma n-3PUFA levels and reductions in circulating plasma inflammatory mediators; interleukin-6 (IL-6), tumour necrosis factor-alpha (TNF-a) and CRP (10-13). However, randomised controlled trials involving n-3PUFA supplementation have produced inconsistent results. Rangel-Heurta et al. were unable to draw a clear conclusion on the association between plasma n-3PUFA levels and

53 reductions in inflammatory mediators, due to heterogeneity in study methodologies that include

54 variations in study population, supplement dosage, study length, and sample size (14). The strongest

55 associations between plasma n-3PUFA levels and inflammatory mediators have been found in older

56 populations, and in populations with elevated inflammatory biomarkers (15-16). More recently a

57 meta-analysis of randomised controlled trials by Li et al. reported a reduction in inflammatory

58 mediators upon n-3PUFA supplementation in healthy and diseased populations (17).

59 The omega-3 index (O3I) is defined as the sum of EPA and DHA expressed as a percentage of total

60 erythrocyte fatty acids (18). O3I has been shown to correlate with n-3PUFA levels in body tissues

61 (18) and is considered a reliable biomarker of long term n-3PUFA intake. The determination of O3I

62 provides an opportunity to examine potential associations between n-3PUFA dietary intake and

63 reductions in inflammatory biomarkers such as CRP and fibrinogen. To date, the association of O3I

64 levels and inflammatory biomarkers has only been explored within unhealthy or moderately

65 hypertriglyceridemic populations and has produced inconsistent results (12, 19-20). The aim of this

66 project was to determine if n-3PUFA status, and in particular O3I, is associated with inflammatory

67 mediators in older Australians. Our recent studies have demonstrated that the association between

68 n-3PUFA status and chronic disease risk factors is sex-dependent (21, 22), therefore, we examined

69 the relationship between n-3PUFA and inflammation mediators separately in older men and

70 women.

MATERIALS AND METHODS Study population and design

This study is a cross-sectional analysis of data collected as part of the Retirement Health and Lifestyle Study (RHLS); a study of the health and lifestyle of older Australians (65 years and older) residing in retirement villages or within the community in the Central Coast region of NSW, Australia. Participants were eligible for the RHLS if they: were > 65 years of age; their primary residence was located within the Wyong or Gosford Local Government Areas; and they had been living at their current address for >12 months. Participants were not eligible if: they were not living independently or were residing in a communal setting other than a retirement village; another member of their household was taking part in the study; they had language and/or other communicative difficulties that limited participation; or they were cognitively impaired and/or were unable to provide informed consent. Those who were eligible and chose to participate in the study (n= 831) took part in an interviewer-administered questionnaire (IAQ) adapted from validated lifestyle surveys (21-24). A subset of eligible participants completed a clinical assessment (n=670) and provided a blood sample (n=649). Participants were included in the present study if they had erythrocyte samples available for fatty acid analyses (n=620) and valid anthropometric and inflammatory marker measurements. Participants with CRP levels > 10mg/L (indicative of acute inflammation), and/or who reported inflammatory disease (e.g. arthritis), and/or that they were taking anti-inflammatory medication or n-3PUFA supplements (25) were excluded from the analyses. All subjects provided written informed consent, and the study was approved by the University of Newcastle Human Research Ethics Committee (H-2008-0431) and the Northern Sydney Central Coast Health Human Research Ethics Committee (Reference No. 1001-031M). The participant recruitment process is summarised in Figure 1.

Biochemical analyses

Blood samples were collected from subjects in a fasting state (>10 hours) by a trained phlebotomist. Participants were asked to abstain from smoking on the morning of collection and to rest for 5

minutes prior to sample collection. CRP, fibrinogen, and white blood cell counts (WBC) including WBC differentials were analysed using standardised methodology by an accredited pathology service (Hunter New England Area Pathology Service).

For erythrocyte membrane fatty acid composition determination, blood samples were first centrifuged at 3000 x g for 10 minutes to separate the plasma and erythrocyte fractions. Fatty acid profiles were analysed using the method of Lepage and Roy via direct trans-esterification of the washed red blood cell (RBC) fraction followed by gas chromatography (26). Methanol :toluene (2 mL, 4:1 v/v, containing C19:0 as internal standard) was added to the erythrocyte samples. Acetyl chloride (200 ^L) was added while vortexing and the samples were then heated for 1 hour at 100°C. After the samples were cooled in water for 5 minutes, K2CO3 6% (5 mL) was added and the samples were centrifuged (3000 x g, 5 minutes, 4°C). The upper toluene phase was collected and stored at -20°C until GC analysis.

Methylated fatty acid samples were analysed using a Hewlett Packard 6890 Series GC equipped with a fixed carbon-silica column 30m x 0.25mm (DB-225) (J and W Scientific), flame ionization detector, autosampler and autodetector. Injector and detector ports were set at 250°C. The oven temperature was programmed at: 170°C for two minutes then increased 10°C per minute to 190°C where it remained stationary for one minute, then increased 3°C per minute to 220°C, and held for a total run time of 30 minutes per sample. A split ratio of 10:1 and an injection volume of 3 ^L were used. Sample peaks were identified by comparison of retention times to known fatty acid standards. Fatty acid profile data were stored and processed using Hewlett Packard Chemstations Version A. 04.02 software. The O3I was calculated by summing the peak area of erythrocyte membrane EPA and DHA and was expressed as a percentage of the total peak area for all erythrocyte membrane fatty acids measured and total n-3PUFA was calculated by summing a-linolenic acid (ALA), EPA, docosapentaenoic acid (DPA) and DHA.

Demographic, Physical Activity, Medical History and Dietary Intake

Demographic data (sex, age, marital status, income, employment, education and country of birth), and information about smoking and participation in physical activity, was collected via an IAQ. Information about diagnosed conditions, medical history and health status was collected via a self-administered medical history questionnaire. Food intake was assessed via a self-administered food frequency questionnaire (FFQ) adapted from the validated interviewer administered CSIRO FFQ (27). Dietary data was analysed using FoodWorks Professional Edition 2009, Version 6.0.2562 (Xyris software, Brisbane, QLD, Australia).

Anthropometrics

Anthropometric measures, including height, weight, waist and hip circumference, were taken by trained research officers according to World Health Organisation guidelines (28). Body weight in light clothing and without shoes was measured using a digital scale (Tanita HD316, Tanita Corporation, Tokyo, Japan; or Wedderburn UWPM150; Wedderburn Scales, Australia) to the nearest 100g, with subject's weight evenly distributed on the scales. Height was measured using a portable stadiometer (Design No. 1013522, Surgical & Medical Products, Australia) to the nearest 0.1cm. Waist circumference was measured at the mid-point of the waist between the lower coastal arch and the iliac crest using a soft measuring tape, to the nearest 0.1cm and upon normal expiration. Hip circumference was measured at the greatest posterior protuberance of the gluteal to the nearest 0.1cm. Measurement was taken upon normal expiration. All anthropometric measurements were taken twice. If the measurements disagreed by more than set tolerance limits a third measure was taken. Each measure was presented as the mean of the two observations or the mean of the two closest measurements if a third was taken. The mean height and weight were used to calculate body mass index (kg/m2).

Sitting Blood Pressure

Seated blood pressure was measured by trained staff using a calibrated Omron 1A2 digital automatic blood pressure monitor (Omron, Australia) as per the National Heart Foundation

147 Guidelines (29). Participants were asked to avoid strenuous exercise for 24 hours prior to

148 measurement, fast overnight, and abstain from smoking on the morning of measurement. Serial

149 readings were taken (minimum of three), and the average of the last two readings within 10mmHg

150 SBP and 6mmHg DBP were recorded.

151 Statistical Analysis

152 Data was analysed using Statistical Package for the Social Sciences (SPSS; Release 17.0, Chicago,

153 IL: SPSS Inc). Descriptive statistics were undertaken for total group, men and women. Continuous

154 data was reported as mean ± standard deviation (SD). Categorical data were reported as frequencies

155 and/or percentages. Group differences between male and female subgroups were assessed using

156 independent samples t-tests (two-tailed) and chi-square analyses. Pearson product-moment

157 correlation analyses were used to evaluate the association between erythrocyte membrane n-3PUFA

158 levels and plasma inflammatory mediators (CRP, fibrinogen, WBCs and WBC differentials).

159 Differences in the number of participants with pro-inflammatory mediator levels above and within

160 reference ranges in O3I quartiles was evaluated using a linear-by-linear association chi-square test

161 (30, 31).

162 Two-block hierarchical multiple regression analyses were used to examine the relationship between

163 O3I and total n-3PUFA and specific inflammatory mediators (32, 33). Anthropometric,

164 demographic and other potentially confounding variables (age, sex, BMI, systolic blood pressure,

165 smoking status, sedentary lifestyle, fasting blood glucose, total cholesterol and triglycerides) were

166 entered into the regression analyses at block 1. Either O3I or total n-3PUFA was entered at the

167 second block of the analysis. The multivariate relationships between select inflammatory mediators

168 and O3I or total n-3PUFA were analysed separately to avoid co-linearity. Pairwise exclusion for

169 missing data was employed in all analyses. Statistical significance was set at p <0.05.

RESULTS

After excluding those participants who reported a pre-existing inflammatory disease and/or had CRP levels >10mg/L (n=236), together with those who were taking anti-inflammatory medications (n=221) or n-3PUFA supplements (n=37), 126 participants (age 77.6 ± 7.3 years; females 46%) remained in the analysis (Table 1). As expected there was a significant difference between males and females for height (p<0.001), body weight (p<0.001), and waist circumference (p<0.001), being higher in men. Waist-to-hip ratio (WHR) was also significantly higher in men (p<0.001). The mean waist circumference for men and women were both above the National Heart Foundation defined cut-off points (men, <94cm; women, <80cm) for increased risk of heart disease (34). Mean total cholesterol (p<0.01), LDL-cholesterol (p<0.05) and HDL-cholesterol (p<0.01) was higher in women, while mean systolic and diastolic blood pressure was similar between males and females. The prevalence of smoking and physical inactivity was also similar across the two genders (Table 1). There were no significant differences in the dietary intake of macronutrients between male and female participants (Table 2). Compared to males, female participants had higher erythrocyte membrane DHA (p<0.05), O3I (p<0.05), and total n-3PUFA (p<0.05) (Table 3).

Females had significantly lower monocyte (p<0.001) and eosinophil (p<0.05) cell counts than males, while total WBC, neutrophil, lymphocyte and eosinophil cell counts, and plasma fibrinogen and CRP levels were not significantly different between the gender sub-groups (Table 4).

The results of bivariate correlation analyses of the relationships between erythrocyte membrane n-3PUFA and the various inflammatory markers are presented in Table 5. With the exception of the inverse relationship detected between fibrinogen and DPA in male participants, fibrinogen, lymphocytes and basophils were not correlated with any of the n-3PUFA reviewed. In contrast, CRP, WBC, neutrophils, monocytes and eosinophils displayed negative correlations with a number of n-3PUFA. CRP was inversely correlated to DHA, O3I and total n-3PUFA. These relationships were not gender specific, with similar but non-significant trends in association evident in both

gender subgroups. WBC and monocytes were negatively correlated with EPA, DPA, DHA, and consequently O3I and total n-3PUFA in the analysis of the overall group. Neutrophils were also negatively correlated with all n-3PUFA variables with the exception of DHA. Gender specificity was evident, with the relationships between WBC and n-3PUFA and neutrophils and n-3PUFA generally displaying specificity to males, and the relationship between monocytes and DPA displaying specificity to females. Eosinophils were also inversely correlated to n-3PUFA in women, but not men, with significant correlations detected between eosinophils and EPA, DHA, O3I and total n-3PUFA.

Hierarchical regression analyses were conducted to assess the association between key inflammatory markers (CRP, WBC, neutrophils, monocytes, eosinophils) and n-3PUFA variables (O3I, total n-3PUFA) after adjusting for potentially confounding variables (age, gender, BMI, systolic blood pressure, sedentary lifestyle, fasting glucose, total cholesterol and triglycerides). The multivariate relationships between each inflammatory marker and O3I, or total n-3PUFA, were assessed separately and the results are summarised in Table 6.

O3I, and total n-3PUFA levels, were significantly associated with monocyte cell counts in all participants, after correcting for the potentially confounding influences of age, gender, BMI, systolic blood pressure, sedentary lifestyle, fasting glucose, total cholesterol and triglycerides (O3I: R=0.433, p<0.05; total n-3PUFA: R=0.449, p<0.01). In both models, the addition of the n-3PUFA variable at the second block of the analyses resulted in significant increases in prediction (O3I: AR2 = 3.5%, p<0.05; total n-3PUFA: AR2 = 4.9%, p<0.05). O3I and total n-3PUFA were also independent negative predictors of monocyte counts in their respective models (O3I: p = -0.205, p<0.05; total n-3PUFA: p = -0.246, p<0.05); as was gender (O3I: p = -0.300, p<0.01; total n-3PUFA: p = -0.309, p<0.01), with female sex being associated with lower monocyte counts.

Hierarchical regression analyses of WBC counts against total n-3PUFA also resulted in a significant predictive model and multivariate R value, after adjustment for potentially confounding variables

(R=0.397, p<0.05). In this analysis, the addition of total n-3PUFA significantly increased the model's predictive power (AR2: 4.6%, p<0.05) and the negative correlation between total n-3PUFA and WBC counts was the only relationship that contributed independently and significantly to the multivariate association observed. A similar but non-significant trend in association was evident for the adjusted analyses of the relationship between WBC and O3I.

Separate hierarchical regression analyses of O3I and total n-3PUFA against CRP, neutrophils and eosinophils, adjusted for age, gender, BMI, systolic blood pressure, sedentary lifestyle, fasting glucose, total cholesterol and triglycerides, resulted in non-significant models with multivariate correlation coefficients between 0.311 and 0.349. Despite the lack of model significance, the addition of O3I or total n-3PUFA at the second block of the analyses contributed significantly to the prediction of CRP and neutrophil cell counts, respectively (AR2 for both models = 3.6%, p<0.05). O3I and total n-3PUFA were also the only independent variables to exert a significant independent influence on CRP levels (P = -0.215, p=0.04) and neutrophil cell counts (P = -0.215, p=0.04), with higher levels of n-3PUFA associated with lower levels of the inflammatory markers.

The percentage of participants with CRP levels within the reference range for each O3I quartile is presented in Figure 2. The proportion of participants with low to average CRP levels (<3.0mg/L) increased across increasing O3I quartiles (61.3%, 74.2%, 68.8%, 84.4%, ¿>-trend=0.04), with significant differences in the proportion of participants with CRP levels <3mg/L found between the first (61.3%) and fourth quartiles (84.4%; x2=4.26, p=0.04).

Similarly the proportion of participants with normal monocyte levels (<0.8x109/L) increased across increasing quartiles of O3I (80.6%, 93.5%, 93.8%, 100.0%, ¿>-trend=0.007; Figure 2), with significant differences in the proportion of participants with monocyte levels < 0.8x109/L found between the first (80.6%) and fourth quartiles (100%; x2=6.85, p=0.009). The proportion of participants with normal fibrinogen, WBC, neutrophil, lymphocyte, and eosinophil levels did not

244 differ between O3I quartiles (data not shown). Basophils were not analysed because the basophil

245 data for all participants were within the reference range.

DISCUSSION

The aim of this cross sectional study was to determine whether erythrocyte n-3PUFA status, and in particular O3I, is a predictor of inflammation in older Australians. This study revealed a significant inverse association between O3I and the inflammatory mediators, CRP, WBC, neutrophils and monocytes. To date literature on the association between erythrocyte n-3PUFA levels, in particular O3I, and inflammation has been limited, therefore the results of this current study represent novel findings. Previous observational studies have generated conflicting results regarding the effect of dietary n-3PUFA intake on inflammatory biomarkers (11, 13, 35-40). Prior studies have determined n-3PUFA levels using self-reported dietary intake, which is limited by recall bias and does not account for absorption or metabolism, or plasma n-3PUFA levels, which are a short term rather than long-term indicator of dietary intake (18, 41).

An inverse association between O3I with CRP and IL-6 has previously been reported in a study of 992 predominantly male patients with coronary artery disease, aged 64 years and over (12). Similarly, in a study of 357 healthy Yup'ik eskimos, there was an inverse association between erythrocyte DHA and EPA with CRP levels, however O3I was not reported in this study and several important confounders, such as physical activity and blood pressure, were not accounted for in the multivariate analysis (42). Conversely, a study of healthy Italian participants aged 20-98 years, did not find an association between erythrocyte n-3PUFA levels and CRP (10). However, the levels of CRP and erythrocyte EPA and DHA reported were low, which may have attenuated any potential relationship.

The O3I values in the current study were higher than those reported in other comparable literature (12, 19, 20). The age of the study population may have contributed to the higher overall O3I mean. A cross sectional study of 160,000 American participants found that O3I can increase by as much as 7% per decade up until 70 years (43). Variations in the methodologies used to analyse erythrocyte

membrane fatty acids may also result in different mean values between studies (44). Regardless, the relationships and trends found would be comparable.

The results of this study may have also been influenced by the BMI of participants (27.0 ± 4.0 kg/m2). BMI was adjusted for within the multivariate regression analyses, however, the percentage of obese participants within this study population was relatively low. Le et al. found in a systematic review and meta-analysis of RCTs that n-3PUFA reduce plasma CRP in subjects with BMI <30kg/m2 but not in subjects with BMI >30kg/m2 (17). Therefore the lower BMI of this study population may have contributed to the inverse correlation found. It has been proposed that the lack of association between CRP levels and n-3PUFA levels in obese participants is due to increased levels of the n-6PUFA, arachidonic acid, in adipose tissue. N-6PUFA are associated with the production of pro-inflammatory cytokines (45). Future studies assessing the association between n-3PUFA and CRP within obese populations are warranted (46).

WBC are widely recognised as reliable biomarkers of inflammation, however have not received the same attention as CRP with respect to their relationship with n-3PUFA or their role in the relationship between n-3PUFA levels and inflammation (47). To our knowledge, this is the first study to explore the relationship between n-3PUFA levels and WBC and the WBC subgroups. Both O3I and total n-3PUFA were inversely associated with WBC, monocyte and neutrophil cell counts independently, and with monocytes when corrected for potential confounders. Total n-3PUFA was also inversely related to WBC, neutrophils and monocytes following multivariate adjustment for confounding variables. In addition, we found inverse associations between DPA levels and WBC and neutrophil counts. Since dietary intake is poorly correlated with DPA levels, it is believed that DPA is predominately produced endogenously from the elongation of EPA or retroconversion of DHA (48). Most observational studies have explored alterations in WBC and WBC subgroup counts in relation to adherence to a Mediterranean diet, and have shown a reduction in WBC, monocyte and neutrophil count (49-51). However this association cannot be attributed to n-3PUFA

levels alone as many other factors may contribute to the health benefits of a Mediterranean diet including polyphenol, olive oil and legume intake (52, 53). Neutrophils and monocytes are the first WBC involved in the inflammatory response. Our finding that higher levels of O3I and total n-3PUFA are associated with reduced monocyte counts suggests that increased intake of n-3PUFA may reduce the proliferation of inflammatory WBC components (54).

This study did not detect an association between O3I and fibrinogen. Low fibrinogen levels (3.5g/L ± 0.6) within this study population may be a likely reason for this lack of association. Participants reported low levels of smoking and sedentary behaviour, two factors that when elevated are associated with increasing fibrinogen levels, and had high HDL levels which is associated with lower fibrinogen levels (7-8). Similarly Kalogeroulous et al. did not find an association between O3I and fibrinogen in healthy Greek male and female participants with fibrinogen levels of 3.18g/L and 3.14g/L, respectively (13). These findings support the postulate that overall healthy lifestyle and dietary behaviours may have attenuated the potential benefits from higher n-3PUFA levels. Conversely, in RCTs conducted in populations with elevated fibrinogen levels, supplementation has resulted in reductions in fibrinogen levels of 10.6% to 30.0% (55, 56). Similarly, an inverse association was found between n-3PUFA levels and fibrinogen in healthy cigarette smoking males (57). The results of the aforementioned studies together with those of the current study suggest that the benefits of n-3PUFA in regards to reducing fibrinogen levels may be most pronounced in those with elevated fibrinogen levels or in those with lifestyle factors known to increase levels.

The current study had several strengths including the use of an objective biomarker of long term dietary n-3PUFA intake rather than dietary n-3PUFA intake alone. Dietary intake assessment is known to have inherent measurement errors and bias (39). The relatively large sample and comprehensive data collection enabled the identification and analysis of potential confounding variables such as smoking status, sedentary lifestyle, blood pressure, BMI, fasting glucose levels, triglycerides and total cholesterol. Furthermore, as participants consuming n-3PUFA supplements

321 were excluded, findings were based on physiological erythrocyte concentrations; which provides a

322 more objective measure of long term PUFA exposure, taking into account intake and metabolism

323 (18). There are also some limitations, including the use of self-report medical history and food

324 frequency questionnaires which may be subject to recall bias. In addition, although the study

325 accounted for many potential variables, we cannot exclude entirely that other lifestyle or dietary

326 patterns have not contributed to the aforementioned associations. Finally, as this study was a cross-

327 sectional analysis, findings from this study cannot suggest causal inverse relationships between n-

328 3PUFA levels and inflammatory biomarkers. Moreover, plasma samples were not available to

329 measure additional markers of inflammation such as IL-6 and TNFa, this remains a weakness of the

330 current study.

331 In conclusion we found a significant independent inverse association between erythrocyte n-3PUFA

332 and circulating inflammatory biomarkers CRP, WBC, neutrophils and monocytes in a cross-

333 sectional analysis of older Australians. The inverse association between O3I and monocytes, and

334 between total n-3PUFA and WBC, neutrophils and monocytes, persisted after multivariate

335 adjustments for potential confounders. Future studies are required to explore the potential anti-

336 inflammatory effects of erythrocyte n-3PUFA levels in other populations, including younger and

337 overweight and obese adults.

338 ACKNOWLEDGMENTS

339 The research on which this paper is based was conducted as part of the Retirement Health and

340 Lifestyle Study, The University of Newcastle. We are grateful to the Australian Research Council

341 (ARC Linkage Proj ect Grant LP0883378), Central Coast Local Health District Public Health Unit,

342 UnitingCare Ageing NSW.ACT, Urbis Pty Ltd, Valhalla Village Pty Ltd and Hunter Valley

343 Research Foundation for funding the initial study and to the men and women of the Central Coast

344 region who provided the information recorded. The authors wish to acknowledge the involvement

345 of the members of the RHLS Group. The authors would also like to acknowledge Ms Melissa Fry

346 for her assistance with the fatty acid analysis.

347 Author contributions: MLG MV and ML designed research; SN KK and MO conducted research;

348 MO analysed data; MO, MLG, LMW, SN and KK wrote the paper; MO had primary responsibility

349 for final content; and all authors contributed to the critical revision of the manuscript and approved

350 the final manuscript.

351 Conflict of interest: None of the authors had a financial or personal conflict of interest.

REFERENCES

1. Krabbe KS, Pedersen M, Bruunsgaard H. Inflammatory mediators in the elderly. Exp Gerontol 2004; 39: 687-99.

2. Calder PC. Omega-3 polyunsaturated fatty acids and inflammatory processes: Nutrition of pharmacology? Br J Clin Pharmacol 2013; 75: 645-62.

3. McInnes IB, Schett G. The pathogenesis of rheumatoid arthritis. N Engl J Med 2011; 365: 2205-19.

4. Mantovani A, Allavena P, Sica A, Balkwill F. Cancer-related inflammation. Nature 2008; 454: 436-44.

5. Calle MC, Fernandez ML. Inflammation and type II diabetes. Diabetes Metab 2012; 38: 183-91.

6. Clearfield MB. C-reactive protein: A new risk factor assessment tool for cardiovascular disease. J Am Osteopath Assoc 2005; 105: 409-16.

7. Pulanic D, Rudan I. The past decade: Fibrinogen. Coll Antropol 2005; 29: 341-9.

8. Davalos D, Akassoglou X. Fibrinogen as a key regulator of inflammation in disease. Semin Immunopathol 2012; 34: 43-62.

9. Schmitz G, Ecker J. Review: The opposing effects of n-3 and n-6 fatty acids. Prog Lipid Res 2008; 47: 147-55.

10. Ferrucci L, Cherubini A, Bandinelli S, Bartali B, Corsi A, Lauretani F, Martin A, Andres-Lacueva C, Senin U, Guralnik JM. Relationship of plasma polyunsaturated fatty acids to circulating inflammatory markers. J Clinc Endocrinol Metab 2006; 91: 439-46.

11. Micallef MA, Munro IA, Garg ML. An inverse relationship between plasma n-3 fatty acids and C-reactive protein in healthy individuals. Eur J Clin Nutr 2009; 63: 1154-6.

12. Farzaneh-Fara R, Harris WS, Garg S, Na B, Whooley MA. Inverse association of erythrocyte n-3 fatty acid levels with inflammatory biomarkers in patients with stable coronary artery disease: The Heart and Soul Study. Atherosclerosis 2009; 205: 538-43.

13. N, Panagiotakos DB, Pitsavos C, Chrysohoou C, Rousinou G, Toutouza M, Stefanadis C. Unsaturated fatty acids are inversely associated and n-6/n-3 ratios are positively related to inflammation and coagulation markers in plasma of apparently healthy adults. Clin Chim Acta 2010; 411: 584-91.

14. Rangel-Huerta OD, Aguilera CM, Mesa MD, Gil A. Omega-3 long-chain polyunsaturated fatty acids supplementation on inflammatory biomarkers: a systematic review of randomised clinical trials. Br J Nutr 2012; 107: S159-70.

15. Kiecolt-Glaser JK, Belury MA, Andridge R, Malarkey WB, Seuk Hwang B, Glaser R. Omega-3 supplementation lowers inflammation in healthy middle-aged and older adults: A randomised controlled trial. Brain Behav Immun 2012; 26: 988-95.

16. Yusof HM, Miles EA, Calder P. Influence of very long-chain n-3 fatty acids on plasma markers of inflammation in middle-aged men. Prostaglandins Leukot Essent Fatty Acids 2008; 78: 219-28.

17. Li K, Huang T, Zheng J, Wu K, Li D. Effect of marine derived n-3 polyunsaturated fatty acids on C-reactive protein, interleukin-6 and tumor Necrosis Factor a: A Meta-Analysis. PLoS One 2014; 9: e88103.

18. Harris WS. The omega-3 index as a risk factor for coronary heart disease. Am J Clin Nutr 2008; 87: 1997S-2002S.

19. Grenon SM, Conte MS, Nosova E, Alley H, Chong K, Harris WS, Vittinghoff E, Owens CD. Association between n-3 polyunsaturated fatty acid content of red blood cells and inflammatory biomarkers in patients with peripheral artery disease. J Vas Surg 2013; 58: 1283-90.

20. Skulas-Ray AC, Kris- Etherton PM, Harris WS, Vanden Heuvel JP, Wagner PR, West SG. Dose-response effects of omega-3 fatty acids on triglycerides, inflammation, and endothelial function in healthy persons with moderate hypertriglyceridemia. Am J Clinc Nutr 2011; 93: 243-52.

21. Australian Institute of Health and Welfare. The Active Australia Survey: A Guide and Manual for Implementation, Analysis and Reporting. Canberra: AIHW, 2003. [cat. no. CVD 22].

22. Giles-Corti B, Timperio A, Cutt H, Pikora TJ, Bull FC, Knuiman M, Bulsara M, Van NK, Shilton. Development of a reliable measure of walking within and outside the local neighborhood: RESIDE's Neighborhood Physical Activity Questionnaire. Prevent Med 2006; 42: 455-59.

23. Craig CL, Marshall AL, Sjostrom M, Bauman AE, Booth ML, Ainsworth BE, Pratt M, Ekelund U, Yngve A, Sallis JF, Oja P. International physical activity questionnaire: 12-country reliability and validity. Med Sci Sports Exerc 2006 35: 1381-95.

24. Dunstan DW, Zimmet PZ, Welborn TA, Cameron AJ, Shaw J, de Courten M, Jolley D, McCarty DJ. The Australian diabetes, obesity and lifestyle study (AusDiab)--methods and response rates. Diabetes Res Clin Pract 2002; 57: 119-29.

25. Yeh EH, Willerson JT. Coming of age of c-reactive protein: using inflammation markers in cardiology. Circulation 2003; 170: 370-1.

26. Lepage G, Roy CC. Direct transesterification of all classes of lipids in a one-step reaction. J Lipid Res 1986; 27: 114-20.

27. Dufficy L, Naumovski N, Ng X, Blades B, Yates Z, Travers C, Lewis P, Sturm J, Veysey M, Roach PD, Lucock MD. G80A reduced folate carrier SNP influences the absorption and cellular translocation of dietary folate and its association with blood pressure in an elderly population. Life Sci 2006; 79: 957-66.

28. Luepker RV, Evans A, McKeigue P. Cardiovascular Survey Methods.3rd ed. Geneva, Switzerland: World Health Organisation (WHO), 2004.

29. National Heart Foundation of Australia, National Blood Pressure and Vascular Disease Advisory Committee. Guide to management of hypertension 2008. Updated December 2010. Internet:

http://www.heartfoundation.org.au/SiteCollectionDocuments/HypertensionGuidelines2008t o2010Update.pdf (accessed 13 th May 2014).

30. Pearson TA, Mensah GA, Alexander RW, Anderson JL, Cannon III RO, Criqui M. Markers of inflammation and cardiovascular disease: application to clinical and public health practice: A statement for healthcare professionals from the Centres for Disease Control and Prevention and the American Heart Association. Circulation 2003; 107: 499-511.

31. Skinner S. Monocytes. 2nd. Ed. In: Skinner S, ed. Understanding clinical inestigations: A quick reference maunual. London, UK: Nova Publishers, 2005:58-9.

32. White S, Greenland S. Simulation study of hierarchical regression. Stat Med 1996; 15: 1161- 70.

33. Fortmann SP, Ford E, Criqui MH, Folsom AR, Harris TB Hong Y, Pearson TA, Siscovick D, Vinicor F, Wilson PF. CDC/AHA workshop on markers of inflammation and cardiovascular disease: application to clinical and public health practice: report from the population science discussion group. Circulation 2004; e554-9.

34. Huxley R, Mendis S, Zhelezhyakov E, Reddy S, Chan J. Body mass index, waist circumference and waist:hip ratio as predictors of cardiovascular risk—a review of the literature. Euro J Clin Nutr 2010; 64: 16-22.

35. Lopez-Garcia E, Schulze MB, Manson JE, et al. Consumption of (n-3) fatty acids is related to plasma biomarkers of inflammation and endothelial activation in women. J Nutr 2004:1341806-11.

36. Ohsawa M, Itai K, Onoda T, et al. Dietary intake of n-3 polyunsaturated fatty acids is inversely associated with CRP levels, especially among male smokers. Atherosclerosis 2008;201(1):184-91.

37. Niu K, Hozawa A, Kuriyama S, et al. Dietary long-chain n-3 fatty acids of marine origin and serum C-reactive protein concentrations are associated in a population with a rich diet in marine products. Am J Clin Nutr 2006;84(1): 223-9.

38. Reinders I, Virtanen JK, Brouwer IA, Tuomainen TP. Association of serum n-3 polyunsaturated fatty acids with CRP in men. Eur J Clin Nutr 2012; 66: 736-41.

39. Sun Q, Ma J, Campos H, Rexrode KM, Albert CM, Mozaffarian D, Hu FB. Blood concentrations of individual long chain n-3 fatty acids and risk of non-fatal myocardial infarction. Am J Clin Nutr 2008; 88: 216:33.

40. Pischon T, Hankinson SE, Hotamisligil GS, Rifai N, WillettWC, TimmEB. Habitual dietary intake of n-3 and n-6 fatty acids in relation to inflammatory markers among US men and women. Circulation 2003;108:155-60.

41. Potischman N. Biologic and methodologic issues for nutritional biomarkers. J Nutr 2003; 133: 875S-80S.

42. Makhoul Z, Kristal AR, Gulati R, Luick B, Bersamin A, Boyer B, Mohatt GV. Associations of very high intakes of eicosapentaenoic acid and docosahexaenoic acid with biomarker of chronic disease risk among Yup'ik eskimos. Am J Clin Nutr 2010; 91: 777-85.

43. Harris WS, Pottala JV, Varvel SA, Borowski JJ, Ward JN, McConnell JP. Erythrocyte omega-3 fatty acids increase and linoleic acid decreases with age: observations from 160,000 patients. Prostaglandins Leukot Essent Fatty Acids. 2013 ;88: 257-63.

44. Harris WS, von Schacky C, Park Y. Standardising methods for assessing omega-3 biostatus. 1st. Ed. In: McNamara RK, ed. The omega-3 deficiency syndrome. Hauppauge, NY: Nova Publishers, 2013:385-98.

45. Calder PC. Long-chain polyunsaturated fatty acids and inflammation. Scandinavian Journal of Food and Nutrition 2006; 50(S2): 54-61.

46. Garaulet M, Pe'rez-Llamas F, Pe' rez-Ayala M, Marti nez P, de Medina FS, Tebar FJ, Zamora S. Site-specific differences in the fatty acid composition of abdominal adipose tissue in an obese population from a Mediterranean area: relation with dietary fatty acids, plasma lipid profile, serum insulin, and central obesity. Am J Clin Nutr 2001; 74: 585-91.

47. Danesh J, Whincup P, Walker M. Low grade inflammation and coronary heart disease: prospective study and updated meta-analysis. BMJ 2010; 321: 199-204.

48. Dalli J, Colas RA, Serhan CN. Novel n-3 immunoresolvents: Structures and actions. Scientific Report 2013; 3: 1-13.

49. Bonaccio M, Castelnuovo AD, Curtis AD, Costanoz S, Persichillo M, Donati MB, Cerletti C, Lacovielloc et al. Adherence to the Mediterranean diet is associated with lower platelet and leukocyte counts: results from the Moli-sani study. Blood 2014; 123: 3037-44.

50. Zampelas A, Panagiotakos DB, Pitsavos C, Das UN, Chrysohoou C, Skoumas Y, Stefanadis C. Fish consumption among healthy adults is associated with decreased levels of inflammatory markers related to cardiovascular disease The ATTICA Study. J Am Cardio 2005; 46: 120-4.

51. Chrysohoou C, Panagiotakos DB, Pitsavos C, Da UN, Stefanadis C. Adherence to the Mediterranean diet attenuates inflammation and coagulation process in healthy adults: The ATTICA Study. J Am Coll Cardio 2004; 44: 152-8.

52. Bonaccio M, Di Castelnuovo A, Bonanni A, et al. Adherence to a Mediterranean diet is associated with a better health-related quality of life: a possible role of high dietary antioxidant content. BMJ Open. 2013; 3 :e003003.

53. Scoditti E, Calabriso N, Massaro M, et al. Mediterranean diet polyphenols reduce inflammatory angiogenesis through MMP-9 and COX-2 inhibition in human vascular endothelial cells: a potentially protective mechanism in atherosclerotic vascular disease and cancer. Arch Biochem Biophys 2012 ;527: 81-9.

54. Swirski FK, Robbins CS. Neutrophils usher monocytes into sites of inflammation. Circ Res 2013; 112: 744-5.

55. Krishnan PV, Anuradha S, Bhattacharjee J, Gaiha M. Effects of low dose omega-3 fatty acids on platelet functions and coagulation profile in indian patients with type 2 diabetes

mellitus with vascular complications: A prospective, preliminary Study. JIACM 2007; 8: 45-52.

56. Seppänen-Laakso T, Laakso I, Lehtimäki T, Rontu, R, Moilanen E, Solakivi T, Seppo S, Vanhanen H, Kiviranta K, Hiltunen R. Elevated plasma fibrinogen caused by inadequate a-linolenic acid intake can be reduced by replacing fat with canola-type rapeseed oil. Prostaglandins, Leukot Essent Fatty Acids 2010; 83: 45-54.

57. Hassen LJ, Ueshima H, Curb JD, Choo J, Lee S, Masaki K, Kadwaki T, Shin C, Evans R, Seto TB, Significant inverse association of marine n-3 fatty acids with plasma fibrinogen levels in Japanese in Japan but not in whites or Japanese Americans. Eur J Clin Nutr 2012;66 :329-35.

TABLE 1

Characteristics of the study participants1

All Men Women P

(n=126) (n=68) (n=58)

Age (years) 77.6 ± 7.3 78.3 ± 7.2 76.9 ± 7.3 NS

Current Smoker (n, %) 3 (2.4%) 2 (2.9%) 1 (1.7%) NS

Physical inactivity (n, %) 8 (6.3%) 4 (5.9%) 4 (6.9%) NS

Height (m) 1.65 ± 0.09 1.70 ± 0.07 1.59 ± 0.07 <0.001

Weight (kg) 74.4 ± 14.6 79.8 ± 14.1 68.3 ± 12.6 <0.001

BMI (kg/m2) 27.0 ± 4.0 27.2 ± 3.9 26.9 ± 4.2 NS

Waist (cm) 96.8 ±12.1 101.8 ± 10.9 90.9 ± 10.9 <0.001

Hip (cm) 106.8 ± 8.6 106.3 ± 6.9 107.3 ± 10.3 NS

Waist: Hip ratio 0.91 ± 0.08 0.95 ± 0.06 0.85 ± 0.05 <0.001

SBP (mmHg) 147.3 ± 21.2 150.3 ± 19.4 143.9 ± 22.8 NS

DBP (mmHg) 75.1 ± 10.1 76.4 ± 9.7 73.5 ± 10.4 NS

Total Protein (g/L) 76.3 ± 4.3 76.7 ± 4.1 75.9 ± 4.6 NS

Glucose (mmol/L) 5.5 ± 1.0 5.6 ± 1.0 5.4 ± 0.9 NS

Total Cholesterol (mmol/L) 4.7 ± 1.0 4.5 ± 0.9 5.0 ± 1.0 <0.01

Triglycerides (mmol/L) 1.2 ± 0.5 1.2 ± 0.5 1.3 ± 0.6 NS

LDL (mmol/L) 2.6 ± 0.9 2.5 ± 0.8 2.8 ± 0.9 <0.05

HDL (mmol/L) 1.5 ± 0.5 1.4 ± 0.4 1.7 ± 0.5 <0.01

compared across gender subgroups using two-tailed independent samples student t-tests. Categorical variables were compared across gender subgroups using Pearson's chi-square tests. BMI = body mass index, DBP = diastolic blood pressure, SBP = systolic blood pressure, HDL = high-density lipoproteins, LDL = low-density lipoproteins, NS = not statistically significant at p<0.05.

TABLE 2

Nutrient intake of the study participants1

All Men Women P

(n=126) (n=68) (n=58)

Total energy (kJ/d) 7648 ± 2862 7984 ± 2930 7254 ± 2755 NS

Total lipid (g/d) 66.8 ± 29.5 67.4 ± 29.4 66.1 ± 29. 8 NS

MUFA (g/d) 25.6 ± 11.8 25.7 ± 12.1 25.4 ± 11.6 NS

PUFA (g/d) 10.9 ± 7.4 11.1 ± 7.1 10.7 ± 7.8 NS

SFA (g/d) 23.7 ± 11.8 23.9 ± 11.9 23.5 ± 11.8 NS

Carbohydrate (g/d) 193.6 ± 80.4 204.2 ± 83.8 181.2 ± 75.0 NS

Protein (g/d) 84.8 ± 34.3 87.1 ± 30.9 82.1 ± 37.9 NS

1 Data are presented as mean and standard deviation. Gender subgroups were compared using two-tailed independent samples student t-tests.

MUFA = monounsaturated fatty acids, PUFA = polyunsaturated fatty acids, SFA = saturated fatty acids, NS = not statistically significant at p<0.05.

TABLE 3

Erythrocyte fatty acids (%) of study participants1

All Men Women P

(n=126) (n=68) (n=58)

16:0 22.77 ± 1.44 22.81 ± 1.66 22.72 ± 1.16 y<NS

18:0 18.48 ± 1.44 18.46 ± 1.64 18.51 ± 1.17 NS

20:0 0.64 ± 0.23 0.62 ± 0.24 0.66 ± 0.21 Ö NS

Total SFA 41.89 ± 2.44 41.88 ± 2.80 41.89 ± 1.96 NS

16:1n-7 0.55 ± 0.28 0.54 ± 0.32 0.56 ± 0.24 NS

18:1n-7 1.75 ± 0.41 1.73 ± 0.44 1.76 ± 0.38 NS

Total MUFA 2.30 ± 0.60 2.28 ± 0.67 2.32 ± 0.52 NS

n-6PUFA

18:2n-6 9.11 ± 1.62 9.15 ± 1.79 9.06 ± 1.40 NS

18:3n-6 0.22 ± 0.30 0.23 ± 0.37 0.22 ± 0.20 NS

20:2n-6 0.34 ± 0.32 0.37 ± 0.42 0.30 ± 0.11 NS

20:3n-6 1.46 ±1.11 1.45 ± 1.12 1.47 ± 1.11 NS

20:4n-6 17.04 ± 2.33 17.14 ± 2.30 16.92 ± 2.38 NS

Total n-6PUFA 28.16 ± 2.44 28.34 ± 2.62 27.96 ± 2.23 NS

n-3PUFA

18:3n-3 (ALA) 0.28 ± 0.15 0.28 ± 0.16 0.28 ± 0.15 NS

20:5n-3 (EPA) 1.38 ± 0.65 ^ 1.28 ± 0.62 1.49 ± 0.67 NS

22:5n-3 (DPA) 3.46 ± 0.63 3.49 ± 0.65 3.42 ± 0.62 NS

22:6n-3 (DHA) 6.28 ± 1.44 5.98 ± 1.40 6.64 ± 1.41 <0.05

Total n-3PUFA 11.40 ± 2.14 11.03 ± 2.11 11.83 ± 2.11 <0.05

Omega-3 Index 7.66 ± 1.92 7.27 ± 1.83 8.13 ± 1.94 <0.05

1Data are presented as mean and standard deviation. Gender subgroups were compared using two-tailed independent samples student t-tests.

ALA = alpha-linolenic acid, DHA = docosahexaenoic acid, DPA = docosapentaenoic acid, EPA = eicosapentaenoic acid, MUFA = monounsaturated fatty acids, n-3PUFA = omega-3 polyunsaturated fatty acids, Omega-3 index= Erythrocyte EPA + DHA % total fatty acids, n-6PUFA= omega-6 polyunsaturated fatty acids, PUFA = polyunsaturated fatty acids, SFA = saturated fatty acids, NS = not statistically significant at p<0.05.

TABLE 4

Biomarkers of inflammation in study participants1

All Men Women P

(n=126) (n=68) (n=58)

Markers of Inflammation

Fibrinogen (g/L) 3.5 ± 0.6 3.4 ± 0.5 3.6 ± 0.7 y<NS

CRP (mg/L) 2.4 ± 2.0 2.3 ± 1.7 2.6 ± 2.3 NS

WBC & WBC differentials

WBC (109/L) 6.29 ± 1.64 6.55 ± 1.64 6.00 ± 1.61 NS

Neutrophils (109/L) 3.81 ± 1.28 3.98 ± 1.20 3.61 ± 1.34 NS

Lymphocyte (109/L) 1.71 ± 0.54 1.70 ± 0.53 1.73 ± 0.55 NS

Monocytes (109/L) 0.54 ± 0.21 0.61 ± 0.23 0.46 ± 0.15 <0.001

Eosinophils (109/L) 0.20 ± 0.15 0.23 ± 0.18 0.16 ± 0.09 <0.05

Basophils (109/L) 0.02 ± 0.04 0.02 ± 0.04 0.02 ± 0.04 NS

1Data are presented as mean and standard deviation. Gender subgroups were compared using two-tailed independent samples student t-tests.

CRP = C-reactive protein, WBC = white blood cell, NS = not statistically significant at p<0.05.

TABLE 5

Correlations between erythrocyte fatty acids and biomarkers of inflammation1

20:5n-3 22:5n-3 22:6n-3 Omega-3 Total

Biomarker EPA DPA DHA Index n-3PUFA

Fibrinogen All -0.139 -0.138 -0.050 -0.084 -0.112

Men -0.150 -0.310 * -0.057 -0.095 -0.178

Women -0.175 0.010 -0.100 -0.133 -0.110

All -0.160 -0.062 -0.182 * -0.190 * -0.193 *

Men -0.187 -0.170 -0.148 -0.177 -0.211

Women -0.172 0.042 -0.258 -0.246 -0.218

All -0.213 * -0.255 ** -0.177 * -0.204 * -0.257 **

Men -0.278 * -0.306 * -0.153 -0.212 -0.286 *

Women -0.095 -0.221 -0.133 -0.130 -0.169

Neutrophils All -0.227 * -0.227 * -0.172 -0.206 * -0.247 **

Men -0.272 * -0.351 ** -0.100 ^-0.169 -0.261 *

Women -0.146 -0.114 -0.190 -0.188 -0.188

Lymphocytes All 0.008 -0.148 0.007 0.008 -0.044

Men -0.157 -0.056 -0.156 -0.173 -0.175

Women 0.174 -0.256 0.180 0.191 0.093

Monocytes All -0.227 * -0.179 * -0.243 ** -0.258 ** -0.283 **

Men -0.179 -0.151 -0.166 -0.188 -0.213

Women -0.208 -0.330 * -0.214 -0.227 -0.293 *

Eosinophils All -0.124 -0.015 -0.178 * -0.175 -0.162

Men -0.014 -0.015 -0.047 -0.041 -0.048

Women -0.271 * -0.060 -0.354 ** -0.351 ** -0.323 *

Basophils All -0.075 -0.044 -0.029 -0.047 -0.040

Men -0.170 0.022 0.032 0.019 0.032

Women -0.130 -0.135 -0.086 -0.107 -0.114

1 Pearson bivariate two- tailed correlation analyses were used to test associations. Pearson's correlation coefficient (r) presented.*<0.05 **<0.01.

ALA = alpha-linolenic acid, CRP = C-reactive protein, DHA = docosahexaenoic acid, DPA = docosapentaenoic acid, EPA = eicosapentaenoic acid, MUFA = monounsaturated fatty acids, n-3PUFA = omega-3 polyunsaturated fatty acids, Omega-3 index= Erythrocyte EPA + DHA % total fatty acids, n-6PUFA= omega-6 polyunsaturated fatty acids, PUFA = polyunsaturated fatty acids, SFA = saturated fatty acids, WBC= white blood cell.

ACCEPTED MANUSCRIPT

TABLE 6

Hierarchical regression analysis of the relationship between inflammatory markers and (a) O3I and (b) total n-3PUFA 1_

CRP WBC Neutrophils Monocytes Eosinophils

(a) & (b) Model 1 R=0.257, R2= =0.066, p=ns R=0.334, R2 =0.112, p=ns R=0.292, R2= 0.085, p=ns R=0.390, R2= =0.152, p<0.05 R=0.281, R2= =0.079, p=ns

Variables ß P ß P ß P ß P ß P

Age (yrs) 0.075 ns -0.059 ns -0.021 ns -0.008 ns 0.060 ns

Gender 0.125 ns -0.142 ns -0.095 ns -0.336 0.001 -0.218 0.030

BMI -0.048 ns 0.015 ns -0.040 ns 0.042 ns 0.036 ns

SBP 0.067 ns 0.114 ns 0.114 ns 0.042 ns 0.026 ns

Smoking -0.025 ns 0.109 ns 0.097 ns 0.028 ns 0.135 ns

Sedentary lifestyle -0.108 ns -0.064 ns -0.066 ns -0.130 ns -0.051 ns

Glucose -0.039 ns 0.043 ns 0.106 ns 0.040 ns 0.000 ns

Total cholesterol -0.157 ns -0.044 ns -0.101 ns 0.001 ns 0.062 ns

Triglycerides 0.219 ns 0.189 ns 0.139 ns 0.105 ns 0.058 ns

(a) O3I: Model 2 R=0.320, R2= =0.102, p=ns R=0.373, R2 =0.139, p=ns R=0.330, R2= =0.109, p=ns R=0.433, R2= =0.187, p<0.05 R=0.324, R2= =0.105, p=ns

ar2= 0.036, p<0.05 ar2 =0.028, p=ns ar2 =0.023, p=ns AR2=0.035, p<0.05 ar2= 0.026, p=ns

Variables ß P ß P ß P ß P ß P

Age (yrs) 0.085 ns -0.051 ns -0.013 ns 0.002 ns 0.069 ns

Gender 0.162 ns -0.110 ns -0.066 ns -0.300 0.002 -0.187 ns

BMI -0.003 ns 0.054 ns -0.004 ns 0.086 ns 0.074 ns

SBP 0.084 ns 0.128 ns 0.128 ns 0.059 ns 0.040 ns

Smoking -0.003 ns 0.128 ns 0.115 ns 0.050 ns 0.154 ns

Sedentary lifestyle -0.094 ns -0.052 ns -0.055 ns -0.117 ns -0.039 ns

Glucose -0.071 ns 0.016 ns 0.081 ns 0.009 ns -0.027 ns

Total cholesterol -0.104 ns 0.002 ns -0.058 ns 0.053 ns 0.106 ns

Triglycerides 0.176 ns 0.152 ns 0.105 ns 0.064 ns 0.022 ns

O3I -0.209 0.045 -0.183 ns -0.168 ns -0.205 0.035 -0.177 ns

(b) Total n-3PUFA: Model 2 R=0.311, R2= =0.097, p=ns R=0.397, R2= 0.158, p<0.05 R=0.349, R2= =0.122, p=ns R=0.449, R2= =0.201, p<0.01 R=0.321, R2= =0.103, p=ns

ar2 =0.031, p=ns ar2= 0.046, p<0.05 ar2= 0.036, p<0.05 AR2=0.049, p<0.05 ar2= 0.024, p=ns

Variables ß P ß P ß P ß P ß P

Age (yrs) 0.092 ns -0.039 ns -0.003 ns 0.013 ns 0.075 ns

Gender 0.147 ns -0.116 ns -0.072 ns -0.309 0.001 -0.199 0.047

BMI -0.012 ns 0.058 ns -0.002 ns 0.086 ns 0.067 ns

SBP 0.077 ns 0.125 ns 0.125 ns 0.054 ns 0.035 ns

Smoking -0.005 ns 0.134 ns 0.120 ns 0.054 ns 0.154 ns

Sedentary lifestyle -0.096 ns -0.049 ns -0.053 ns -0.115 ns -0.040 ns

Glucose -0.058 ns 0.021 ns 0.087 ns 0.017 ns -0.016 ns

Total cholesterol -0.086 ns 0.042 ns -0.024 ns 0.091 ns 0.125 ns

Triglycerides 0.170 ns 0.130 ns 0.086 ns 0.044 ns 0.015 ns

Total n-3PUFA -0.195 ns -0.238 0.018 -0.212 0.038 -0.246 0.012 -0.173 ns

1 Two block hierarchical multiple regression analysis was conducted. O3I: Omega-3 index = erythrocyte %EPA+%DPA of total erythrocyte fatty acids (50). Total n-3PUFA = %EPA+%DPA+%DHA.

FIGURE LEGENDS:

Figure 1. Overview of participant recruitment.

Figure 2. The percentage of older adults (aged >65yrs) with CRP levels or monocyte counts within reference ranges for each O3I quartile. A. The percentage of participants within each O3I quartile with CRP levels <3.0mg/L or >3.1mg/L (30). B. The percentage of participants within each O3I quartile with monocyte count <0.8x109/L or >0.8x109/L (31).

12 3 4

O3I Quartile

B) 120

1234 O3I Quartile

DATE: 27th November 2015 The Editor in Chief

Journal of Nutrition & Intermediary Metabolism (JNIM) Dear Editor in Chief,

RE: Highlights of new submission titled 'Erythrocyte omega-3 polyunsaturated fatty acid levels are associated with biomarkers of inflammation in older Australians'

§ Study results suggest a negative association between the omega-3 index and inflammation in older people.

§ Results warrant future longer-term intervention trials to examine the role of increased intake of omega-3 fatty acids in reducing inflammation in older people for the prevention of non-communicable diseases.

I look forward to a favourable response to our submission. Yours sincerely,

Manohar Garg Corresponding author