Effect of food intake on 92 biomarkers for cardiovascular disease

Objective The present study evaluates the effect of food intake on 92 biomarkers for cardiovascular disease (CVD). Methods Twenty two healthy subjects (11 male and 11 female aged 25.9±4.2 years) were investigated. A total of 92 biomarkers were measured before a standardized meal as well as 30 and 120 minutes afterwards with the Proseek Multiplex CVD III kit. Results The levels for eight biomarkers decreased significantly (P<0.05) 30 minutes after food intake. The levels for seven biomarkers remained significantly decreased 120 minutes after food intake. Nine biomarker decreased significantly at 120 minutes after food intake. The changes were between 4–30%, most commonly around 5%. Only six biomarkers showed a difference of 10% or more due to food intake. The biggest differences were observed for Insulin-like growth factor-binding protein 1 (30%); Azurocidin, Cystatin-B, and Myeloperoxidase (13%); Monocyte chemotactic protein 1 (11%); and Myeloblastin (10%), all 120 minutes after food intake. Conclusions This study shows that food intake affects several different CVD biomarkers, but the effect is predominantly modest. Timing of blood sampling in relation to food intake, therefore, appears not to be a major concern. Further studies are warranted in older healthy subjects and in patients with various cardiac diseases to determine whether the findings are reproducible.


Results
The levels for eight biomarkers decreased significantly (P<0.05) 30 minutes after food intake. The levels for seven biomarkers remained significantly decreased 120 minutes after food intake. Nine biomarker decreased significantly at 120 minutes after food intake. The changes were between 4-30%, most commonly around 5%. Only six biomarkers showed a difference of 10% or more due to food intake. The biggest differences were observed for Insulin-like growth factor-binding protein 1 (30%); Azurocidin, Cystatin-B, and Myeloperoxidase (13%); Monocyte chemotactic protein 1 (11%); and Myeloblastin (10%), all 120 minutes after food intake.

Conclusions
This study shows that food intake affects several different CVD biomarkers, but the effect is predominantly modest. Timing of blood sampling in relation to food intake, therefore, appears not to be a major concern. Further studies are warranted in older healthy subjects and in patients with various cardiac diseases to determine whether the findings are reproducible. PLOS

Introduction
Cardiovascular disease (CVD) is the leading cause of death in the world. Current risk assessment systems are far from perfect [1,2]. There is a continuous search for novel biomarkers that could improve the risk assessment in CVD [3][4][5][6][7][8][9]. Digestion of food is known to have significant hemodynamic and metabolic effects [10][11][12][13][14][15], and may, therefore, affect different biomarkers. It is relevant from a practical point of view to investigate whether these biomarkers are affected by food intake, as the timing of blood sampling would then become relevant. Moreover, it could be of physiological interest to investigate biological responses to food intake. The present study evaluates the effect of food intake in healthy volunteers on 92 different emerging biomarkers for CVD. To the best our knowledge, this has not been done before.

Study population
The study investigated 22 healthy Caucasians (11 male and 11 female, aged 25.9±4.2 years). None of the subjects had a prior history or showed any symptoms of cardiovascular disease or any other chronic diseases. None of the subjects were taking any cardiovascular medication. Height and weight were measured and BMI and body surface area [16] were calculated. The subjects ingested a standardized meal consisting of 300 g rice pudding (AXA Goda Gröten Risgrynsgröt; Lantmännen AXA, Järna, Sweden). The total caloric value of the meal was 330 kcal: 10% from protein (9 g), 58% from carbohydrates (48 g), and 32% from fat (12 g). Our ambition with the meal was to mimic a breakfast. Written informed consent was obtained from all participants. The trial is registered in the US National Library of Medicine with the trial registration number NCT01027507. The study was approved by the regional ethical review board in Lund, Sweden.

Statistical analysis
Data are presented as mean ± standard deviation (SD), and the difference ± SD (for the difference) at 30 and 120 minutes after food intake. Statistical analyses were carried out using Statistica 12 (StatSoft Inc, Tulsa, OK, USA). Comparison between fasting values versus 30 and 120 minutes after food intake for any given biomarker was analyzed for significance with repeated measures ANOVA with Post hoc analysis by Tukey's test. Statistical significance was set at P<0.05. Table 1 displays descriptive statistics of the study population. The levels for eight biomarkers decreased significantly 30 minutes after food intake. The levels for seven biomarkers remained significantly decreased 120 minutes after food intake. Nine biomarker decreased significantly at 120 minutes after food intake. The changes were between 4-30%, most commonly around 5%. Only six biomarkers showed a difference of 10% or more due to food intake. The biggest differences were observed for Insulin-like growth factor-binding protein 1 (30%); Azurocidin, Cystatin-B, and Myeloperoxidase (13%); Monocyte chemotactic protein 1 (11%); and Myeloblastin (10%), all 120 minutes after food intake. A summary of the findings is in Table 2.

Results and discussion
Statistical significance is reported in three different levels; Ã (P<0.05), ÃÃ (P<0.01), and ÃÃÃ (P<0.001). The values between P<0.05 and P<0.01 should be treated with caution because of the possibility of mass significance. To our knowledge, this is the first study to evaluate the effect of food intake on plasma proteins measured by the Proseek Multiplex CVD III kit. This investigation showed that 17 of the 92 investigated biomarkers were affected by food intake. The changes were, however, modest. Only six biomarkers showed a difference of 10% or more due to food intake. The changes were, for the most part, around 5%. The observed effects of food intake are approximately half of the intra-and interassay variation which are about 10% for most of the biomarkers. This suggests that the need to standardize food intake is not generally necessary when using this kit (Proseek Multiplex CVD III). There were, however, some exceptions. The biggest differences were observed for Insulin-like growth factor-binding protein 1 (30%); Azurocidin, Cystatin-B, and Myeloperoxidase (13%); Monocyte chemotactic protein 1 (11%); and Myeloblastin (10%), all 120 minutes after food intake. Insulin-like growth factor-binding protein 1 has been shown to predict cardiovascular mortality and morbidity in patients with acute myocardial infarction [19,20]. The level of insulin-like growth factor-binding protein has been shown to regulate insulin-like growth factor-I bioactivity, glucose homeostasis, and tissue regeneration, and increases during inflammation [21]. It is known that insulin suppresses the insulin-like growth factor-binding protein 1 expression [22]. This may explain why we observed a significantly lower level of insulin-like growth factor-binding protein 1 in the postprandial period in our study. It could, however, be difficult in the acute setting to standardize food intake, but our findings clearly suggest that this should be done when investigating Insulin-like growth factorbinding protein 1. Inflammation has been suggested to be the underlying pathogenesis of a range of the most common cardiovascular diseases [23,24]. Myeloblastin is a protease that degrades proteins such as elastin, fibronectin, laminin and collagen and has mainly been investigated in vasculitis [25]. Azurocidin, also known as cationic antimicrobial protein CAP37 or heparin-binding protein is an important multifunctional inflammatory mediator [26]. Myeloperoxidase has also been shown to be an inflammatory marker in coronary heart disease [24]. A postprandial increase in white blood cell and platelet counts and a decrease in the release of myeloperoxidase from polymorphonuclear leukocyte has previously been observed [27]. In our study, we also observed a decreased Myeloperoxidase level in the postprandial period. Both Cystatin-B, and Monocyte chemotactic protein 1 have both recently been investigated in  Food intake and biomarkers for CVD large population studies and have been associated with an increased incidence of CVD [5,9]. The difference of 10-13% for Myeloblastin, Cystatin-B, Monocyte chemotactic protein 1, Azurocidin and Myeloperoxidase suggests that not standardizing food intake prior to sampling may introduce additional error. Recent technological advances makes it possible to measure multiple plasma proteins simultaneously. In the current study the Proseek Multiplex CVD III kit was used to simultaneously measure 92 plasma proteins. The proteins have been selected because of their importance in cardiovascular disease. According to the manufacturer, 28 of the biomarkers are inflammatory/immune system, 25 exploratory, 21 CVD, 11 metabolic and 6 CVD/Inflammation. This technology has recently been used in a multitude of studies [3][4][5][6][7][8][9], and our investigation adds methodological information in this developing field.
We are aware of only one study that has evaluated the impact of food intake on the biomarkers in the present study. Jahn and co-workers evaluated the effect of food intake on 26 of the 91 biomarkers in the present study [28]. This study evaluated the biomarker levels before and 240 minutes after a high-fat breakfast in 16 healthy and 18 metabolic syndrome subjects. Their findings were partly at odds with the present investigation. The findings in the study by Jahn et al. and in our study were concordant for 11 biomarkers (Growth differentiation factor-15, TNF receptor 1, Interleukin-6 receptor subunit alpha, TNF-receptor 2, FAS receptor, ST2 protein, Cathepsin D, Integrin beta binding protein 2, Spondin-1, Epidermal growth factor recptor, N-terminal prohormone-BNP), whereas it was not for 15 biomarkers. It is interesting to observe that Jahn et al. found no change in 7 biomarkers where we did (Cystatin B, Monocyte chemotactic protein 1, TNF-receptor superfamily member 14, Myloperoxidase, Resistin, Chitinase-3-like protein 1, Urokinase-type plasminogen activator receptor), and the opposite was also found for 8 biomarkers (Matrix metalloproteinase-3, Galectin-3, E-selectin, Osteoprotegerin, Kallikrein-6, Myoglobin, Tissue plasminogen activator, Fatty acid binding protein-4). The differences can most probably be explained by the differences in statistical analysis, observation time, or the differences in the meal composition between the studies.
There are some limitations that should be addressed. The present investigation studied the effect of food intake only in young healthy Caucasian subjects. Studies are warranted in older healthy subjects, different ethnic groups, and in patients with various cardiac diseases to determine whether the findings in the present investigation are reproducible in such populations. Future investigations should also address the effect of meal size and different diets such as high or low fat [29].
In conclusion, the present investigation shows that several biomarkers that have been used to investigate CVD are affected by food intake, but the effect is predominantly modest. Timing of blood sampling in relation to food intake, therefore, appears not to be a major concern. There are, however, some exceptions. Caution concerning food intake should be taken when investigating Insulin-like growth factor-binding protein 1, Azurocidin, Cystatin-B, Myeloperoxidase, Monocyte chemotactic protein 1, or Myeloblastin.