Metabolic Cytokines at Fasting and During Macronutrient Challenges: Influence of Obesity, Female Androgen Excess and Sex

Abstract

Scope: Cytokines have pleiotropic functions within the organism and their levels may be influenced by obesity, visceral adiposity and sex hormones. Diet composition may also affect their systemic concentrations during fasting and in the postprandial period. Hence, we studied the influence of sex steroids and obesity on the circulating levels of a panel of metabolic cytokines in the fasting state and after single macronutrient challenges.

Methods: On alternate days we submitted 17 women with polycystic ovary syndrome (PCOS) (9 non-obese, 8 obese), 17 non-hyperandrogenic control women (9 non-obese, 8 obese) and 19 control men (10 non-obese, 9 obese) to isocaloric oral glucose, lipid and protein loads. Serum levels of omentin-1, vaspin, lipocalin-2, adipsin, PAI-1, chemerin, FGF-21 and FGF-23 were determined by Luminex multiplex technology.

Results: During fasting, obese patients presented higher levels of PAI-1, chemerin and adipsin but decreased FGF-23 and omentin-1 compared with non-obese subjects. Vaspin showed sexual dimorphism with lower levels in men than women with PCOS and female controls. Following macronutrient ingestion, most metabolic cytokines presented a similar physiological response consisting of a decrease in circulating concentrations, which was inversely associated with the fasting levels of these molecules. Protein intake caused the major postprandial decrease whereas glucose did not significantly reduce PAI-1, FGF-23 and vaspin, and even increased FGF-21. Regardless of the macronutrient administered, vaspin levels showed a larger reduction in non-obese individuals while the decrease in PAI-1 was particularly noticeable in the obese subgroup. The postprandial reductions of omentin-1 and FGF-23 after glucose and protein loads were influenced by obesity. No major differences were found between patients with PCOS and male and female controls.

Conclusions: Obesity, but not PCOS or sex, markedly influences metabolic cytokine levels at fasting and after macronutrient ingestion. The observed postprandial decrease in their circulating concentrations might represent a physiological compensatory mechanism against food-induced inflammation and oxidative stress. This mechanism is altered by obesity and is differently modulated by macronutrients, suggesting a larger contribution of glucose to stressful postprandial responses.

Keywords: PCOS; adipokines; adiposity; glucose; lipids; oral loads; proteins; sex hormones.

Figure 1

Areas under the curve (AUC) of circulating Omentin-1, Vaspin, Lipocalin-2 and Adipsin during oral macronutrient challenges considering all subjects as a whole, and according to obesity and group of subjects. Data are means ± SEM. P values represent the statistical significances of the differences within (introducing the three values along the curve as within-subjects factor) and between (introducing the AUC of each load as within-subjects factor and group and obesity as between-subjects factors) macronutrient loads analyzed by repeated-measures general linear models. *p < 0.05, **p < 0.01 and ***p < 0.001 for differences from fasting levels within macronutrient load. †p < 0.05 for the differences among the glucose, lipid, and protein loads, regardless of obesity and group of subjects. §p < 0.05 for the global effect of obesity, irrespective of macronutrient and group of subjects. ‖p < 0.05 for the interaction between macronutrient and obesity, regardless of group of subjects.

Figure 2

Areas under the curve (AUC) of circulating PAI-1, Chemerin, FGF-21 and FGF-23 during oral macronutrient challenges considering all subjects as a whole, and according to obesity and group of subjects. Data are means ± SEM. P values represent the statistical significances of the differences within (introducing the three values along the curve as within-subjects factor) and between (introducing the AUC of each load as within-subjects factor and group and obesity as between-subjects factors) macronutrient loads analyzed by repeated-measures general linear models. ^**p < 0.01 and ^***p < 0.001 for differences from fasting levels within macronutrient load. ^†p < 0.05 and ^‡p < 0.01 for the differences among the glucose, lipid, and protein loads, regardless of obesity and group of subjects. ^§p < 0.05 for the global effect of obesity, irrespective of macronutrient and group of subjects. ^‖p < 0.05 for the interaction between macronutrient and obesity, regardless of group of subjects.