A system model of the effects of exercise on plasma Interleukin-6 dynamics in healthy individuals: Role of skeletal muscle and adipose tissue

Abstract

Interleukin-6 (IL-6) has been recently shown to play a central role in glucose homeostasis, since it stimulates the production and secretion of Glucagon-like Peptide-1 (GLP-1) from intestinal L-cells and pancreas, leading to an enhanced insulin response. In resting conditions, IL-6 is mainly produced by the adipose tissue whereas, during exercise, skeletal muscle contractions stimulate a marked IL-6 secretion as well. Available mathematical models describing the effects of exercise on glucose homeostasis, however, do not account for this IL-6 contribution. This study aimed at developing and validating a system model of exercise's effects on plasma IL-6 dynamics in healthy humans, combining the contributions of both adipose tissue and skeletal muscle. A two-compartment description was adopted to model plasma IL-6 changes in response to oxygen uptake's variation during an exercise bout. The free parameters of the model were estimated by means of a cross-validation procedure performed on four different datasets. A low coefficient of variation (<10%) was found for each parameter and the physiologically meaningful parameters were all consistent with literature data. Moreover, plasma IL-6 dynamics during exercise and post-exercise were consistent with literature data from exercise protocols differing in intensity, duration and modality. The model successfully emulated the physiological effects of exercise on plasma IL-6 levels and provided a reliable description of the role of skeletal muscle and adipose tissue on the dynamics of plasma IL-6. The system model here proposed is suitable to simulate IL-6 response to different exercise modalities. Its future integration with existing models of GLP-1-induced insulin secretion might provide a more reliable description of exercise's effects on glucose homeostasis and hence support the definition of more tailored interventions for the treatment of type 2 diabetes.

Fig 1. Two-compartment description of the IL-6 dynamics during exercise.

Skeletal muscle secretes IL-6 in the local (muscle) blood flow (IL6_m(t)) in response to change in oxygen consumption (PVO_2max) with a secretion rate equal to SR_ex. Plasma IL-6 (IL6_p(t)) is the result of adipose tissue secretion (Ra_IL6), hepatosplanchnic viscera removal (k_e) and contribution coming from muscle compartment (through k_m).

Fig 2. Model fit results.

(A) PVO_2max model prediction (B) Mean model fit (solid line) for IL-6. Measured IL-6 concentrations (means ± SEM) from Ostrowski et al. [29] are shown, along with the model fit.

Fig 3. IL-6 weighted residuals.

Fig 4. Model validation results obtained using the conditions reported in the study by Fischer et al.

(A) PVO_2max model prediction (B) Measured IL-6 concentrations (means ± SEM) from Fischer et al. [30], shown along with the model prediction (solid line).

Fig 5. Model validation results obtained using the conditions reported in the study by Steensberg et al.

(A) PVO_2max model prediction (B) Measured IL-6 concentrations (medians and quartiles) from Steensberg et al. [8], shown together with the model prediction (solid line).

Fig 6. Model validation results obtained using the conditions reported in the study by Febbraio et al.

(A) PVO_2max model prediction (B) Measured IL-6 concentrations (means ± SEM) from Febbraio et al. [25], shown together with the model prediction (solid line).