HIIT has been confirmed to reduce VAT and relieve metabolic syndrome effectively2, 5–7. However, individual differences between subjects are large2. A previous meta-analysis found that the effect of HIIT on reducing VAT is evident in subjects with obesity or overweight but weak in normal weight subjects5. This phenomenon may be related closely to the difference between obesity phenotypes in adipose catabolism. Previous experiments have found that the fat loss effect of HIIT is related to the activation lipolysis26–28, 31. Because gene polymorphisms underlie individual differences in food intake, PA and adipose metabolism, we hypothesised that OP and OR individuals show different adaptive changes to HIIT, especially in lipolysis regulated by the SNS.
We compared how 12 weeks of HIIT affected VAT loss in OP and OR rats, as well as how this training affected catecholamine-regulated lipolysis pathways. We found: (1) HIIT could reduce HFD-induced body weight gain, liver fat and VAT mass in both OP/OR rats, and the influence in OP rats was even stronger. (2) After training, VAT of OP rats was more sensitive to catecholamines and showed stronger lipolysis compared with OR rats. (3) Increased β3-AR expression, rather than increased SNS activity, plays a key role in mediating these adaptive changes.
HIIT Had a Stronger Influence on Body Weight, VAT mass and Liver Lipids in OP Compared with OR Rats
Similarly to the heterogeneity of human obesity, rodents also show two different obesity phenotypes, namely OP and OR14, 15. Although the baseline weights of these rats are the same, marked differences in food intake, PA, central regulation and peripheral tissue metabolism lead OP rats to become obese after several weeks of receiving a HFD, while OR rats tend to remain at a normal weight32–34. We established a model of OP and OR rats by using a classic programme involving a HFD14, 15, 34. While establishing the model, the food intake and weight gain of OP rats were significantly higher than for OR rats, indicating that the model was established successfully.
Studies have shown that when the environment changes, OR individuals often exhibit a steady state of energy metabolism, while OP individuals respond more strongly to caloric restriction or food surplus35, 36, suggesting that the OP phenotype has a survival advantage when energy is insufficient but is harmful during periods of excess food. Previous comparative studies of OP and OR have often focussed on aerobic exercise rather than HIIT20–22, 35, 36. Levin 18 found that OP rats lose more body weight and fat during aerobic exercise than OR rats, and Jen et al. 23 and Zachwieja et al.24 showed that aerobic exercise has similar effects on the weight loss of OP and OR rats, but a stronger effect on reducing VAT mass in OP subjects. However, differently from aerobic exercise, HIIT does not consume fat during exercise, and whether it has different effects on different metabolic phenotypes requires new evidence. To our knowledge, this study compared the fat-loss effects of HIIT on OP and OR subjects for the first time. We found that HIIT had a stronger effect on reducing VAT in OP than OR rats, indicating that the differential adaptation of OP and OR subjects may be the reason for the strong heterogeneity of the HIIT fat-loss effect.
Because the 12 weeks of training had no effect on the food intake, the heterogeneity of the VAT decrease between OP and OR rats may be more related to altered peripheral metabolism than feeding regulation by the central nervous system. The VAT mass in adulthood is more related to the volume rather than the number of adipocytes, and the main contributor (> 95%) of the cell volume comes from TG accumulation in lipid droplets37, 38. We found that after 6 weeks of a HFD, the adipocyte area of OP rats was larger than that of OR rats, a finding that is characteristic of the classic OP animal model39, 40. After 12 weeks of HIIT, the cell area had decreased in both OP and OR rats, although the reduction was greater in OP rats, accompanied by an increase in the number of mitochondria (mean higher local lipid oxidation), suggesting that HIIT reduces the TG content of VAT in OP rats more strongly than OR.
It is well known that VAT accumulation is associated with metabolic syndrome, non-alcoholic fatty liver disease and cardiovascular disease41. Weight reduction by energy restriction or exercise is still the preferred method to alleviate liver steatosis rather than drugs42. A previous study confirmed the close relationship between VAT mass and hepatic lipid deposition43, and the exercise-induced decrease in VAT is often accompanied by decreased hepatic TG content44. Our results are consistent with previous findings: very similarly to VAT, the decrease in the liver lipid level in OP rats was larger than in OR rats, indicating greater health benefits in the former group. In summary, HIIT showed a greater influence on the VAT mass, cell volume and liver lipid deposition of OP compared with OR rats, findings that imply a stronger catabolic change in OP individuals.
VAT of OP Rats Showed Greater Lipolytic Potential than OR Rats after HIIT
Based on the data that the VAT of OP rats decreased more than OR rats after training, we hypothesised that heterogeneity in VAT changes between OP and OR rats is related to the different adaptations of adipocyte lipolytic pathways. Lipolysis is the primary step of TG decomposition, which refers to the process by which TG are hydrolysed into glycerol and non-esterified fatty acids (NEFA, the released form of TG). This process is mainly regulated by AR in the SNS45, 46. The SNS releases NE and epinephrine through nerve endings and adrenal glands, which can activate second messenger pathways (involving G proteins and cAMP) through AR, ultimately leading to PKA-mediated phosphorylation of HSL and activation of lipolysis47. Catecholamine release is correlated with exercise intensity48. Aerobic exercise mediates moderate secretion, which enhances lipolysis through β-AR of adipocytes to meet fat consumption during exercise, while high-intensity exercise induces excessive secretion of catecholamines and inhibits lipolysis through a negative feedback mechanism involving α-AR49, 50.
Because minimal fat is burned during exercise, it is generally believed that HIIT can reduce fat based on post-exercise TG consumption23, 24, 51. A commonly mentioned view is that HIIT could increase excess post-exercise oxygen consumption (EPOC), but there are still controversies among existing results52. Another reasonable hypothesis is that due to the intense load of HIIT, during the recovery period, adipose tissue would release more NEFA for tissue healing or reconfiguration in muscle, lung and other tissues24, which suggests HIIT promotes adaptive catabolism of adipose tissue. Studies have verified the increase in β-AR and lipolysis in adipocytes after long-term HIIT26, 28, supporting the hypothesis of adaptative changes. As mentioned earlier, when facing catabolic stress, OP individuals are more likely to show adaptation (fat is lost easily), while OR individuals are more likely to maintain homeostasis35, 36. Therefore, a reasonable explanation why the VAT mass of OP rats decreases more than that of OR rats after HIIT is the visceral adipocytes of the former group generate a stronger adaptation to HIIT, and the lipolytic pathways are easier to start-up to ensure an energy supply when facing fasting, cold, exercise or other stress.
To test this hypothesis, we evaluated the expression and phosphorylation of HSL (the rate-limiting enzyme of lipolysis) as well as glycerol release (lipolysis marker) of adipocyte in vitro. We found that after 12 h of fasting (inducing catecholamine release), there was no difference in HSL expression between the groups, but the level of phosphorylated HSLser660 in the exercise groups was higher than in the control groups, indicating an activating effect of HIIT on VAT lipolysis. Our in vitro experiment confirmed greater lipolysis in the H-OP group after catecholamine stimulation, suggesting that there is stronger catabolism in the OP rats. In summary, these results suggest greater adaptation of lipolysis pathways in OP compared with OR rats.
Increased β3-AR Expression May Play a Key Role in Mediating Adaptive Change to HIIT
Although VAT lipolysis and NTS in the H-OP group showed stronger adaptations to HIIT, the upstream mechanisms that mediate these change remained unknown. Adipose tissue is regulated extensively by the neuroendocrine network. Although the SNS plays a primary role (Bartness et al. 2014),53 parasympathetic nerves, natriuretic peptides, glucocorticoids and parathyroid hormone are also involved in regulating lipolysis54. Even downstream of the SNS, catecholamines and AR are not the only ‘transmitter–receptor’ pathway. Recent studies have reported other transmitters, receptors or intermediary cells such as neuropeptide Y (NPY), G-protein coupled receptor 3 (GPR3) and adipose mesenchymal cells (MSCs) involved in the SNS-mediated regulation of adipose catabolism55–57. Due to the existence of many pathways, the role of the SNS and β3-AR needed to be further confirmed in this study.
Because the hypothalamus is the major regulator of energy homeostasis and directly controls the SNS and food intake53, 58, we speculated that the hypothalamus of OP rats could adapt to HIIT, thereby changing the feeding behaviour and SNS-regulated adipose metabolism. Unexpectedly, similarly to the food intake, 12 weeks of training did not influence the TH expression of VAT, indicating that HIIT does not promote NE release of sympathetic nerves. However, β3-AR expression of the exercise groups was increased significantly, and the increase was greater in the H-OP compared with the H-OR group, indicating that HIIT could increase more strongly the receptor number and the exercise adaptation of OP rats. To determine whether higher β3-AR expression produced stronger sensitivity to catecholamines, isolated adipocytes were stimulated by an ISO gradient (0.1–10 µM, to imitate levels of catabolism pressure) in vitro. The glycerol release of the H-OP group was significantly higher than the H-OR group, suggesting that adipocytes of OP rats are more sensitive to catecholamines and could produce more NEFA to export or local oxidation under the same SNS signal.
So far, three beta isoforms of AR have been reported (β1, β2 and β3). Although β3-AR predominates in white adipose tissue59, studies have confirmed that gene polymorphisms of all three isoforms could affect the fat catabolism induced by aerobic exercise60–62, although the relationship between HIIT and different isoforms had been unclear. To determine whether the VAT adaptation to HIIT is only related to β3-AR or also to other isoforms, we blocked this receptor with SR59230a, a selective antagonist, concomitantly with 10 µM ISO stimulation. Without the participation of β3-AR, the increased glycerol release in the H-OP group disappeared. These data confirm that increased expression of β3-AR, rather than β1-AR or β2-AR, plays a key role in increasing the sensitivity of adipocytes to catecholamine. In summary, after 12 weeks of HIIT, greater β3-AR expression in OP rats led to an elevated sensitivity to catecholamines compared with the OR rats. This change underlies the greater VAT lipolysis capabilities of OP rats.