Enhanced interleukin‐6 in human adipose tissue vein after sprint exercise: Results from a pilot study

Low‐volume sprint exercise is likely to reduce body fat. Interleukin (IL‐6) may mediate this by increasing adipose tissue (AT) lipolysis. Therefore, the exchange of AT IL‐6 and glycerol, a marker of lipolysis, was examined in 10 healthy subjects performing three 30‐s all‐out sprints.

. Such changes may increase lipolysis and thermogenesis in white adipose tissue (Brownstein et al., 2022;Frayn, 2010;Laurens et al., 2020;Mu et al., 2021;Nelson et al., 2018;Rognstad & Katz, 1966;Thompson et al., 2012).The stimulatory effect of the sympatho-adrenergic system on adipose tissue lipolysis is one of the most well-known lipolytic stimulus (Lange, 2004;Thompson et al., 2012).Of particular interest, beta-adrenergic stimulation appears to enhance the release of the cytokine interleukin (IL-6) from adipose tissue (Mohamed-Ali et al., 2001), which is an important contributor to systemic IL-6 (Mohamed-Ali et al., 1997, 1999, 2001).In fact, IL-6 has been shown to increase adipose tissue lipolysis in vivo in humans (Lyngsø et al., 2002b;van Hall et al., 2003) and may, locally in adipose tissue, enhance the effect of sympatho-adrenergic stimulation on lipolysis and thermogenesis.Aerobic exercise has previously been found to induce release of IL-6 from subcutaneous abdominal adipose tissue (Lyngsø et al., 2002a) and to precede postexercise adipose tissue lipolysis (Lyngsø et al., 2002a).It has been suggested that IL-6 promotes postexercise lipolysis and fatty acid mobilisation from adipose tissue (Lyngsø et al., 2002a).To our knowledge, there is no previous study of IL-6 release from adipose tissue after sprint exercise.Such studies may improve the understanding of how lipolysis is regulated in adipocyte tissue, which may contribute to insights into how fat mass can be reduced in humans.
The main aim of the present study was, therefore, to examine the acute effects of sprint exercise on IL-6 exchange between subcutaneous adipose tissue and blood.In addition, it was of interest to relate the exercise-induced changes in IL-6 to corresponding changes in plasma glycerol, since glycerol is a marker of lipolysis (Coppack et al., 1990;van Hall et al., 2002).It was hypothesised that both IL-6 and glycerol concentrations in the abdominal subcutaneous vein would increase in adipose tissue after sprint exercise.

| Subjects
Ten healthy voluntary subjects (six females and four males) participated in the study.Mean ± standard deviation (SD) age, height, body weight and body mass index were 27 ± 7 years, 171 ± 7 cm, 69 ± 10 kg and 23 ± 2 kg × m −2 , respectively.The subjects were recruited through advertisement and most of them were students at a college for sports and recreation instructors.The inclusion criteria were healthy, nonsmoking young adults of both sexes, having regular physical activity, but not at an elite or competitive athletic level.An activity index (Jansson & Hedberg, 1991) (values ranging between 5.5 and 20.5) was calculated to estimate the physical activity during leisure time and was 16 ± 2. Body fat percentage, fatfree body mass and body fat mass were estimated by skinfold measurements from triceps, biceps and sub scapula (Durnin & Womersley, 1974) and were 21 ± 7%, 55 ± 10 and 14 ± 5 kg, respectively.Smokers, individuals on medication and women in menstrual phase were not included in the study.The subjects were fully informed about the procedures and potential risks of the experiment before giving their consent to participate in the study, which was approved by the Ethics Committee of Karolinska Institutet (DR 235-00).

| Experimental protocol
The subjects were asked to refrain from any heavy exercise during the 24-h period preceding the experiment and were familiarised with the Wingate test (Bar-Or, 1987) at least 24 h before the experiment.The subjects reported to the laboratory in the morning after an overnight fast except for a piece of white bread and a glass of water one to 2 h before.A Teflon catheter was inserted percutaneously into the brachial artery and into a superficial subcutaneous vein on the anterior abdominal wall, providing access to the venous drainage from the subcutaneous abdominal adipose tissue (Frayn et al., 1989;Simonsen et al., 1994;Summers et al., 1998).Five mL of blood was sampled from each of the artery and the abdominal vein, 2 min before and nine after each sprint, with the subject in supine position.

| Sprint exercise protocol
After 60-s warm up on a load of 25% of the individual braking load, the subjects performed three 'all-out' cycle sprints of 30-s duration (Bar-Or, 1987) on a mechanically braked cycle ergometer (Cardionics) with 20 min rest between the sprints.They were instructed to pedal as fast as possible at an individual braking load set at 0.075 kp × kg body weight −1 .Peak power (i.e., the highest 5-s power) and mean power (the average power during the 30-s cycle) were calculated for each of the three 30-s cycle sprints.The rational for choosing this protocol with relatively long rest in between the bouts was that it is a well-characterised protocol of metabolic and hormonal response to sprint exercise for both sexes (Esbjörnsson-Liljedahl et al., 2002, 2009, 2012a, 2012b;Rundqvist et al., 2017Rundqvist et al., , 2019) ) originally designed to reach a close-to-full recovery of power output between the sprints (Esbjörnsson et al., 1993).

| Blood/plasma preparation and analyses
For separation of plasma, blood samples were centrifuged at 4°C after collection.The plasma supernatant was frozen and stored at −70°C until performing enzyme immunoassay analysis of IL-6 (Quantikine HS, R&D systems) in duplicates, according to the manufacturer's manual, a lower detection limit of 0.09 pg × mL −1 , an intraassay coefficient of variation (CV) in the order of 4% and an intraassay CV up to approximately 10%.In our hands, the intraassay CV was 8%.
Plasma glycerol was analysed by a colorimetric enzymatic method (Sigma-Aldrich, Glycerol Assay Kit) with a linear detection range of 10-1000 µM.The colourimetric detection was chosen, as plasma values up to approximately 200 µmol × L −1 were expected.The intraassay CV was, in our hands, approximately 8%.To save plasma material, analyses in duplicates for glycerol, were focused on samples after sprints 2 and 3, were increased levels of adipose venous IL-6 were expected.Lactate was determined in duplicates in neutralised perchloric acid extract of whole blood by a fluorometric enzymatic method (Lowry & Passonneau, 1972) with an, in our hands, intraassay CV of approximately 2%.Individual average concentrations of lactate (average exercise lactate) were calculated from the samples obtained after sprints 1, 2 and 3.

| Statistics
Values in the text are means and SD unless otherwise stated.The p-values were accepted as statistically significant at p < 0.05.For the plasma and power variables, a one-way analysis of variance with repeated measures was applied to test the response over time.Twotailed Student's t test for paired observations was applied when appropriate.The statistical analysis of the relationship between arterio-subcutaneous abdominal venous IL-6 plasma concentration and the mean arterial lactate concentration was performed using Pearson's correlation analysis.A multiple correlation analysis was performed with adipose tissue venous concentration of glycerol at 9 min after the last sprint as dependent variable and adipose tissue venous concentration of IL-6 at 9 min after the last sprint and body fat percentage as independent variables.No corrections for multiple comparisons were made.In the present study, parametric statistics was chosen.However, it is difficult to make an accurate assessment of normality with only 10 subjects, even though most of the skewness and kurtosis values were within the range of ±1.This means that data followed an approximate normal distribution (Mishra et al., 2019).To minimise the risk of type 2 error, we also run all analyses with nonparametric statistics and the outcome was consistent with the parametric statistics (data not shown).

| RESULTS
Peak power and mean power were 707 ± 161 and 554 ± 115 Watt, respectively during the first sprint and decreased by 7% and 4%, respectively from the first to the last sprint (p < 0.03 and p < 0.05).
Arterial (a) IL-6 increased over the study period up to approximately 2-fold from rest to 9 min postsprint 3, p < 0.03 (Figure 1a).Adipose tissue venous (v) IL-6 increased approximately 15-fold from rest to 9 min postsprint 3, p < 0.0001 (Figure 1b).Adipose tissue v-a concentration difference increased up to 45-fold over the same period, p < 0.0001 (Figure 1c).An accelerating increase was observed during the exercise period with a peak value at the end, particularly for the adipose tissue venous concentration and the v-a difference.Arterial (a) glycerol increased 2.5-fold from rest to 9 min postsprint 1 (p < 0.0001) and stayed at this level over the three bouts of sprints exercise (Figure 2a).Adipose tissue venous (v) and v-a concentration difference of glycerol increased approximately 2-fold from rest to 9 min postsprint 1 (p < 0.0001 and p = 0.01), decreased until 18 min postsprint 2 (p < 0.001 and p < 0.0001) and increased again to 9 min postsprint 3 (p < 0.01 and p < 0.01) (Figure 2b,c).The multiple correlation analysis demonstrated that approximately 60% of the variation in adipose tissue venous glycerol concentration at 9 min after the last sprint was explained by venous IL-6 at 9 min after the last sprint (p < 0.05) and body fat percentage (p < 0.05).
Average exercise arterial blood lactate (see methods) was 12 ± 2 mmol × L −1 and a positive correlation was observed between adipose tissue v-a concentration difference of IL-6 at 9 min after the | 173 last sprint exercise and the average exercise arterial blood lactate concentration (Figure 3).

| Major findings
To our knowledge, this is the first study of IL-6 exchange over adipose tissue after sprint exercise.The study confirmed the hypothesis of an increase in IL-6 in the adipose tissue vein after sprint exercise, in parallel with an increase in the lipolysis marker glycerol.This was observed after the last bout of sprint exercise.In contrast, a marked peak in glycerol in the adipose tissue vein was observed after the first sprint, without a corresponding increase in IL-6.
4.2 | Why is there an increase of IL-6 in adipose tissue venous concentration?
Previously, a large increase in systemic catecholamines has been shown during sprint exercise with the same protocol as the current one (Esbjörnsson-Liljedahl et al., 2002).This could stimulate the release of IL-6 from the adipose tissue (Keller et al., 2003;Mohamed-Ali et al., 2001).Catecholamines were not measured in the present study, but the correlation between systemic lactate and adipose tissue venous IL-6 in the present study supports such a mechanism in that the increase in lactate reflects the increase in catecholamines during sprint exercise (Esbjörnsson-Liljedahl et al., 2002).However, why the increase in IL-6 in adipose tissue venous blood was delayed until the last and third bout is difficult to explain in light of a very large increase in systemic catecholamines already after the first sprint (Esbjörnsson-Liljedahl et al., 2002).It is possible that adipose tissue venous IL-6 could be entrapped in adipose tissue by reduced adipose tissue blood flow during and closely after the sprint, similar to what has been proposed for free fatty acids (FFA) (Romijn, Coyle, et al., 1993).Blood flow was not measured in the present study.
However, preliminary data in three subjects, using the 133 Xe-wash out method to study adipose tissue blood flow supported this.All subjects had a markedly reduced adipose tissue blood flow, especially early after the first 30-s sprint, but less so after the second and third F I G U R E 3 Relationship between adipose tissue venous interleukin-6 plasma concentration 9 min after last sprint and average arterial plasma lactate concentration during the entire exercise protocol in 10 subjects.The correlation coefficient was 0.82, p < 0.01.sprint (M.Esbjörnsson, J. Bülow, B. Norman and Jansson, unpublished data).The suggested entrapped IL-6 may be released after subsequent sprints when flow is normalised or increased above the basal level (Romijn, Klein, et al., 1993).In addition, the increase in adipose tissue venous IL-6 could be due to increased production or release of IL-6 by activated leucocytes in the adipose tissue through, for example, catecholamines (Freire et al., 2022).Speculatively, this could be reinforced by entrapped FFA having a toxic effect in adipose tissue (Chitraju et al., 2017).

| Marginal increase in arterial IL-6
Similarly to the present study, only a small increase in arterial IL-6 level was found after 60 min of aerobic exercise (Lyngsø et al., 2002a).
Production of IL-6 in contracting human skeletal muscles was suggested to account for the exercise-induced increase in arterial IL-6 (Steensberg et al., 2000).The low arterial IL-6, but high venous level in adipose tissue suggests that a possible IL-6-mediated increase adipose tissue lipolysis after sprint exercise was a paracrine or autocrine local effect of IL-6 rather than an endocrine systemic effect.

| IL-6, lipolysis, fat mass and energy expenditure
Two waves of lipolysis were indicated by the measurement of glycerol in adipose tissue venous blood in the present study.The early wave of increase in adipose tissue venous glycerol during repeated bouts of sprint exercise was not related to IL-6.From the literature, the most likely explanation is that the early activation of sympathoadrenergic activity at the onset of intense exercise stimulated lipolysis.However, the second wave of increase in AT venous glycerol was concurrent with the increase in adipose tissue venous IL-6.
FFA levels in adipose tissue venous blood were not presented in this study.FFA may, however, be a less reliable marker for lipolysis than glycerol after sprint exercise because the possible entrapment of FFA in adipose tissue during sprint exercise (Hodgetts et al., 1991;Romijn, Coyle, et al., 1993).This is evident from the systemically measured FFA in an antecubital vein in two earlier studies using the same exercise protocol.A decrease in FFA was seen already after the first sprint that continued over the three bouts of exercise, followed by a marked increase during the postexercise period (Rundqvist et al., 2017).In a subset of samples from four subjects from the present study, FFA was analysed in arterial and adipose tissue venous blood.A marked decrease of both arterial and venous FFA was found in all subjects when comparing basal values with values 9 min after the last sprint (M.Esbjörnsson, J. Bülow, B. Norman and E. Jansson, unpublished data).This confirms the discrepancy between the changes in glycerol and FFA during sprint exercise.
Energy expenditure has been shown to increase after sprint exercise (Hazell et al., 2012).IL-6 has been shown to increase energy expenditure when infused into humans (Lyngsø et al., 2002b;van Hall et al., 2003), and could, therefore, contribute to the sprint exerciseinduced increase in energy expenditure.Whether this is causally related to IL-6 is not known.Other agents that increase lipolysis such as catecholamines and growth hormone may also contribute to increase energy expenditure in response to sprint exercise.Indeed, lipolysis is followed by re-esterification and thus increased triglyceride-FFA cycling and energy expenditure (Townsend et al., 2017).Interestingly, such an increase in triglyceride-FFA cycling rate has been shown to be particularly elevated during the postexercise period (Wolfe et al., 1990) and increased by exercise intensity (Pritzlaff et al., 2000).The lipolytic hormones growth hormone and catecholamines are important candidates for high ATP consumption in adipose tissue (Rognstad & Katz, 1966).Support for regulation of body mass and fat mass by IL-6 has been demonstrated both in mice and humans.For instance, in mice, deficiency of IL-6 has been shown to lead to the development of obesity (Wallenius et al., 2002) and IL-6 has been shown to mediate the exercise-induced protective effect on adiposity (Li et al., 2021).In humans, IL-6 blockade was also found to impair mobilisation of FFA at rest and during exercise (Trinh et al., 2021) and the exerciseinduced reduction in visceral adipose tissue mass was abolished by IL-6 blockade (Wedell-Neergaard et al., 2019).Regulation of fat mass by IL-6 may also be mediated by a reduced appetite, especially after sprint exercise (Islam et al., 2017).

| IL-6 and lactate
The observed increase in adipose tissue venous IL-6 concentration shortly after three bouts of 30-s sprint exercise was related to arterial blood lactate.Whether this is a causal relationship is not known, although both IL-6 and lactate may be related to exercise-induced sympathoadrenergic stress (Mohamed-Ali et al., 2001).Regardless of the cause and effect of the relationship between adipose tissue venous IL-6 and arterial lactate, it is suggested that a high IL-6 in adipose tissue vein is associated with high systemic lactate after sprint exercise and possibly elevated energy expenditure through the Cori cycle (Brownstein et al., 2022).This could further contribute to body fat mass regulation associated with sprint exercise.

| Strength and limitations
A major strength of the study is that we have canulated a vein that provides access to the venous drainage of subcutaneous adipose tissue, allowing studies of adipose tissue metabolism in human subjects (Frayn et al., 1989).
The external validity is limited by the low number of subjects and that they all were young and fit.No control group was included in this study.However, Lyngsø et al. included a nonexercise control in their study of IL-6 exchange over subcutaneous adipose tissue after aerobic exercise (Lyngsø et al., 2002a).They found no significant increase in arterial concentration or release of IL-6 from adipose tissue during the first 60 min, which corresponds to the duration of the exercise protocol in the present study.The correlation between venous adipose tissue IL-6 and arterial lactate concentrations supports that the increased IL-6 after sprint exercise was a consequence of exercise.
Since we did not measure adipose tissue blood flow, it was not possible to quantify the release of IL-6 from adipose tissue.However, the net release was estimated to be 20 pg × 100 g tissue −1 × min −1 9 min after the third sprint by using the preliminary blood flow data (see above) together with adipose tissue v-a concentration difference of IL-6 from the current study.This release was of the same magnitude as after 1-2 h postaerobic exercise (Lyngsø et al., 2002a).
In addition, it would have been interesting to measure the IL-6 release for a longer period of time postsprint exercise for comparison with aerobic exercise (Lyngsø et al., 2002).In fact, we measured net release of IL-6 from the adipose tissue 2 h postexercise in two subjects and it was found to be 24 and 30 pg × 100 g tissue −1 × min −1 , respectively (M.Esbjörnsson, J. Bülow, B. Norman and E. Jansson, unpublished data).Thus, a continued release of IL-6 for at least 2 h can be expected after sprint exercise, as earlier observed after aerobic exercise (Lyngsø et al., 2002a).Finally, only the release of adipose tissue IL-6 was analysed in the current study.In future studies, it would be interesting use an unbiased 'omic' approach to identify new lipolytic agents stimulated by exercise.

| CONCLUSIONS
In healthy young subjects, a concurrent increase of IL-6 and the lipolysis marker glycerol was observed in the adipose tissue vein after repeated bouts of sprint exercise.However, the glycerol data also indicated an early wave of lipolysis during sprint exercise unrelated to IL-6.Thus, IL-6 may complement other sprint exercise-induced lipolytic agents such as catecholamines and growth hormone.
U R E 1 (a) Interleukin (IL-6) arterial (a) and (b) adipose tissue venous (v) plasma concentration and (c) adipose tissue v-a IL-6 concentration difference during sprint exercise in 10 subjects.Main effect of time over all points of time was tested by repeatedmeasures analysis of variance.Values are expressed as mean ± standard error.ESBJÖRNSSON ET AL.
U R E 2 (a) Glycerol arterial (a) and (b) adipose tissue venous (v) plasma concentration and (c) adipose tissue glycerol v-a concentration difference during sprint exercise in 10 subjects.Main effect of time over all points of time was tested by repeatedmeasures analysis of variance and changes induced by sprint 1 and sprint 3 by paired t-test.Values are expressed as mean ± standard error.
Increased lipolysis and thereby an increased energy expenditure is hypothesised to contribute to sprint exercise-induced reduction in body fat.AUTHOR CONTRIBUTIONSMonaEsbjörnsson and Eva Jansson conceived and designed the research.Mona Esbjörnsson, Eva Jansson, Jens Bülow and Barbara Norman performed experiments.Moa Persson, Amarjit Saini and Barbara Norman performed analyses.Mona Esbjörnsson and Eva Jansson analysed data.Eva Jansson, Mona Esbjörnsson, Jens Bülow and Barbara Norman interpreted the results of experiments.Mona Esbjörnsson prepared figures.Eva Jansson and Mona Esbjörnsson drafted manuscript.All authors edited and revised the manuscript as well as approved final version of the manuscript.