Review Article | Published:

Mechanisms of weight regain after weight loss — the role of adipose tissue

Nature Reviews Endocrinology (2019) | Download Citation


One of the biggest challenges in the management of obesity is the prevention of weight regain after successful weight loss. Weight regain after weight loss has large interindividual variation. Although many factors probably contribute to this variation, we hypothesize that variability in biological responses associated with weight loss-induced shrinking of subcutaneous adipocytes has an important role. In this Review, we show that weight loss-induced variations in cellular stress, extracellular matrix remodelling, inflammatory responses, adipokine secretion and lipolysis seem to be associated with the amount of weight that is regained after successful weight loss. Weight regain could therefore, at least in part, depend on a combination of these factors. Further research on the causality of these associations could aid the development of effective strategies to prevent weight regain after successful weight loss.

Key points

  • Weight regain after successful weight loss is a major problem for many individuals, and many factors are probably involved in driving weight regain.

  • Loss of fat mass induces shrinkage of adipocytes, which is accompanied by cell stress, inflammation, altered adipokine secretion and reduced lipolysis.

  • In the absence of extracellular matrix remodelling during adipocyte shrinkage, mechanical stress builds up between the cell and the oversized extracellular matrix, which inhibits lipolysis and the release of fatty acids from adipocytes.

  • Weight loss induces an inflammatory response in adipose tissue.

  • Evidence for the involvement of epigenetic modifications, in particular microRNAs, in weight regain is sparse.

  • Knowledge of weight regain after weight loss is mainly based on associations, and so research into the causality of such associated factors is needed.


The worldwide prevalence of obesity is rising1. According to one estimate, every year, 42% of the world population tries to lose weight2. Even if the initial weight loss is successful, most individuals will regain weight over time, and only a small proportion of dieters will be able to maintain the reduced weight over the next years. For example, a meta-analysis of weight loss studies in the USA, which included 29 studies that applied hypocaloric diets with or without exercise with long-term (≥2 years) follow-up, showed that on average more than half of the weight lost is regained after 2 years and more than three-quarters is regained after 5 years3. In the Look Ahead study, which included patients with type 2 diabetes mellitus (T2DM) who are overweight, an intensive lifestyle intervention, with regular group sessions, structured meal plans and free provision of meal replacements during the first year, resulted in improved long-term weight loss outcomes compared with diabetes support and education alone4. However, the average weight regain in the intensive lifestyle intervention group was still 50% of initial weight loss after 4 years, which was maintained until 8 years. Even after bariatric surgery, weight regain can occur5,6,7. Identifying the factors that make long-term weight loss maintenance so difficult would be an important step forward for researchers and health-care professionals in the obesity field. This advance would enable investigators to explain the interindividual differences in weight regain and aid the development of better strategies for long-term weight loss outcomes.

Obesity is the consequence of small, cumulative imbalances between energy intake and energy expenditure that result in a positive energy balance and weight gain8. Many biological factors, including genetics, environmental and developmental factors, epigenetics and gastrointestinal factors but also socio-economic, behavioural and lifestyle factors and their interactions with one another have been implicated in weight gain8,9,10. In most cases, no single factor is solely responsible for the development of obesity. Rather, obesity results from a combination of factors, and these combinations can vary among individuals and over time8,9,10.

Both in the normal weight state and following the development of obesity, fat depots are homeostatically protected, be it at different levels8. Presently, it is unclear (in most cases of obesity) what factors are responsible for the defence of the elevated fat stores and whether they differ from the factors that influence energy homeostasis in the normal weight range. Some factors might overlap, but additional factors could specifically be involved in protecting the increased white adipose tissue (WAT) mass in the obese state10,11,12,13,14,15.

Factors unrelated to adipose tissue, such as eating behaviour or genetics, might have a role in weight regain after weight loss, and these factors are summarized in Table 1. Although changes in signals from the gastrointestinal system after diet-induced or surgery-induced weight loss are thought to be important for weight regain16,17,18, their role is not included in Table 1, as studies on direct associations with weight regain are lacking. As this Review focuses on adipose tissue-related factors, we do not elaborate further on non-WAT-related factors.

Table 1 Predictors of weight regain after weight loss in adults with obesity

One of the main events that occurs during a weight loss intervention is the shrinking of WAT mass19. This loss of tissue mass results predominantly from a reduction in size of adipocytes within WAT. In this Review, we explain how the shrinking of adipocytes influences their function. In addition, we discuss whether this reduction in the size of adipocytes might have a role in triggering the regain of WAT mass in people who try to maintain the weight loss they have achieved. In Fig. 1, we present an overview of the potential adipose tissue-related mechanisms that we discuss.

Fig. 1: Potential adipose tissue-related mechanisms involved in weight regain after weight loss.
Fig. 1

Weight loss induces shrinkage of adipocytes. This shrinkage might affect several potential mechanisms, which are described in detail in this Review. Interindividual variations in response to these mechanisms might determine weight maintenance or weight regain. ECM, extracellular matrix; WAT, white adipose tissue.

Excess white adipose tissue

WAT expands in size when an individual’s overall energy balance is positive (that is, when the intake of energy is greater than the amount of energy expended). The expansion of WAT can occur via two mechanisms, hyperplasia and hypertrophy20.

Preadipocytes, which are present in the stromal vascular fraction of WAT, can develop into mature adipocytes21. The maturation of preadipocytes is accompanied by uptake and storage of lipid, which leads to an increase in cell size22. The increase in the number of adipocytes by preadipocyte recruitment is referred to as hyperplasia. In humans, hyperplasia can occur before adulthood, but above 20 years of age, the number of adipocytes is stable, and only ~8% are renewed each year23.

Therefore, a positive energy balance in adults will typically lead to the expansion of WAT via hypertrophy of mature adipocytes. Interestingly, a study reported that following overfeeding, different fat depots might behave differently with regards to their expansion24: abdominal subcutaneous adipose tissue responded by hypertrophy, but femoral subcutaneous adipose tissue responded by hyperplasia. These differences in response are the result of a complex regulatory mechanism that includes signals from hormones and growth factors20,25,26.

Inefficient subcutaneous fat cell lipolysis (that is, increased in vitro basal lipolysis and reduced stimulated lipolysis) has been identified as a process that promotes weight gain and increases fat cell volume27. With regards to hypertrophy, there seems to be a limit to the extent to which adipocytes can enlarge, which is controlled by the extracellular matrix (ECM). In mice, knocking out collagen VI — the main component of the adipocyte ECM — abolishes this limitation and allows further expansion of the adipocytes, which is accompanied by improvements to the status of metabolic parameters, such as insulin sensitivity28.

A study that used transcriptomics to analyse WAT from individuals with obesity showed an increased amount of interstitial fibrosis, which was associated with infiltration of inflammatory cells29. In several human studies, a negative correlation was found between fibrosis and hypertrophy, supporting the concept that the ECM is indeed restricting hypertrophy30,31. Furthermore, on the basis of animal studies, WAT expansion has been proposed to lead to a state of hypoxia owing to either the blood flow to the WAT not keeping up with the tissue expansion or the adipocytes expanding beyond the diffusion range of the available oxygen32. In both scenarios of adipose tissue expansion, hypoxia will change the metabolic properties of WAT and promote fibrosis and inflammation33,34. This response to hypoxia might be mediated by the upregulation of HIF1α and/or HIF2α, which are oxygen-sensitive transcription factors32,35. In humans, however, measurements of oxygen pressure in subcutaneous WAT have resulted in contradictory findings, showing either a decreased or an increased oxygen pressure when comparing WAT from lean individuals with WAT from obese individuals36,37 and a reduction in oxygen tension with weight loss38. These discrepant findings might be partly due to the technological complexity of the measurement.

Immune cells are commonly found within WAT. For example, the stromal vascular fraction of adipose tissue contains immunomodulatory cells such as macrophages, of which the number is directly related to the level of adiposity and adipocyte size39. In healthy WAT, adipocytes and immune cells communicate and maintain a balanced situation in which anti-inflammatory cytokines counteract pro-inflammatory cytokines40,41. In obesity, however, hypertrophy of adipocytes leads to increased cellular stress by fibrosis and hypoxia. This increase in cellular stress results in a higher secretion of chemoattractants, which augments the invasion of immune cells, the transition of macrophages to a more aggressive immunological phenotype and the increased production of pro-inflammatory cytokines from WAT42,43,44. On the other hand, in lean men, overfeeding for 56 days resulted in increased deposition of connective tissue in the adipose tissue without a change in the number of macrophages or inflammatory cells45.

A study in mice suggested that local insulin resistance in WAT leads to the secretion of MCP1 (also known as CCL2) and the activation of pro-inflammatory macrophages46. Increased cell death might also sustain a chronic condition of low-grade inflammation. The important questions that remain unanswered, however, are whether the changes observed in adipose tissue following the development of obesity are reversed by weight loss and whether the changes have a role in weight regain after weight loss.

The role of adipocytes in weight regain

Changes in adipocyte size

When people gain weight through a positive energy balance, the surplus of energy can be stored as triglycerides in adipose tissue. In adulthood, increased fat storage predominantly occurs in existing adipocytes and leads to the formation of large hypertrophic adipocytes23,25. A study in rats47 showed that, under calorie restriction, adipocytes lose fat, and the average volume of the adipocytes declines. The authors of this study found, however, that upon refeeding the average adipocyte volume returned to the original value. They also noticed an increase in the number of very small cells (<20 µm) during calorie restriction, which were thought to be newly formed adipocytes. Following refeeding, new adipocytes joined the distribution of mature adipocytes, leading to a slight increase in the total number of cells. Therefore, in rats, weight regain seems to predominantly result from the refilling of mature adipocytes with triglycerides and is only partly due to the differentiation of new adipocytes.

In participants of the YoYo study48 (a clinical trial comparing the effects of two rates of weight loss — a 12-week low-calorie diet (LCD) or a 5-week very-low-calorie diet (VLCD) — on weight regain), adipocytes in WAT also reduced in size during weight loss. Furthermore, and in line with the animal study, investigators reported an increase in the percentage of small adipocytes and a reduction in the percentage of large adipocytes, as well as a gradual return of adipocytes to their original size distribution after cessation of the weight loss strategy49 (Fig. 2). The technique used in this study, however, did not allow for an estimation of changes in total cell number.

Fig. 2: Adipocyte diameter distribution in different phases of weight loss and weight regain.
Fig. 2

The distribution of adipocyte diameter (d, in µm) in abdominal subcutaneous adipose tissue before weight loss, at the end of weight loss, after 4 weeks of weight maintenance and after 9 months of follow-up in the YoYo study48 is shown. At the end of weight loss, the percentage of small adipocytes (with a diameter <50 µm) is increased and that of large adipocytes (70–90 µm) is reduced compared with baseline. During the subsequent 4-week strict weight maintenance period, small adipocytes refill with lipid, resulting in a decreased percentage of small adipocytes and an increase in the percentage of large adipocytes. There are no further changes in the distribution of adipocyte diameters during the follow-up period in which fat mass increases.

Data suggest that the adipogenic capacity of the stromal vascular fraction of WAT is impaired in individuals who are obese50. In a study of patients with obesity undergoing weight loss51, the authors found that the adipogenic capacity of stromal cells in biopsy samples of subcutaneous WAT improved following weight loss. In the same study, expression of the RUNX2 gene, which encodes a transcription factor for osteogenesis, was markedly downregulated after weight loss, which suggests inhibition of alternative lineage programmes and preferential adipogenic differentiation. The newly formed adipocytes also displayed an altered secretion profile with increased levels of leptin. However, in the YoYo intervention study, no significant change in expression of RUNX2 was observed in WAT biopsy samples directly after weight loss or at 4 weeks after returning to energy balance (M.A.B. and E.C.M.M., unpublished observations), whereas plasma leptin levels were dramatically decreased after weight loss52,53. These data suggest that in humans, as in rats, weight regain after weight loss results predominantly from the refilling of mature adipocytes (Fig. 3).

Fig. 3: Model of resolution of weight loss-induced adipocyte stress.
Fig. 3

Mature adipocytes are surrounded by an extracellular matrix (ECM) (part a). In the absence of ECM remodelling during adipocyte shrinkage, mechanical stress builds up between the cell and the oversized ECM (part b). In principle, this stress can be resolved in three ways: by refilling of the adipocytes, which will be accompanied by weight regain (parts ba), by ECM adjustment with maintenance of weight loss (part c) or by preadipocyte differentiation (part d). The latter does not seem to be a major contributor to weight regain.

Adipocyte stress model

Adipose tissue is a loose connective tissue. Therefore, it is not surprising that adipocytes secrete many different proteins and deposit a dense ECM. The adipocyte ECM is formed when preadipocytes differentiate into mature adipocytes. Adipocytes require a strong ECM to prevent mechanical disruption that could occur as the cell increases in size during lipid storage. In vitro experiments in 3T3-L1 adipocytes have shown that insulin can stimulate ECM formation, a process that probably occurs via the transcriptional upregulation of genes encoding enzymes needed to biochemically modify ECM proteins54. Another study in cultured mature 3T3-L1 adipocytes reported that mature adipocytes show considerable production of collagens to maintain the ECM55. Production and post-translational modification of collagens, and of ECM proteins in general, require a considerable amount of energy56.

When in negative energy balance, adipocytes activate lipolysis and release fatty acids as fuel for other tissues57,58. Consequently, during negative energy balance, adipocytes shrink, and in parallel, the ECM is remodelled to fit the smaller cell surface. Following weight loss after bariatric surgery, subcutaneous adipose tissue displays major collagen remodelling, with increased collagen degradation and crosslinking, features that are in agreement with ECM adaptation during the loss of fat mass59. Under conditions of calorie restriction, however, the energy available to adipocytes might not always be sufficient for the cell to undergo ECM remodelling56. In the absence of ECM remodelling during adipocyte shrinkage, mechanical stress builds up between the cell and the oversized ECM. The force of the mechanical stress is highest at the focal adhesions, which are the membrane-embedded protein complexes that link the intracellular actin filaments to the ECM60 (Fig. 4). Our hypothesis is that this mechanical stress inhibits lipolysis and the release of fatty acids from adipocytes. Moreover, we believe that adipocytes that are under stress are preconditioned for renewed fat accumulation and trigger the host to increase energy intake. Therefore, in this scenario, adipocytes would start to accumulate fat again, and the increasing cell volume would reduce the mechanical stress (Fig. 3) — which means weight regain for the host15.

Fig. 4: Structure of the focal adhesion as the centre for mechanical stress.
Fig. 4

The structure of the integrin-containing focal adhesions in the adipocyte membrane connecting the cytoskeleton (intracellular actin filaments) to the extracellular matrix (ECM) enabling the passage of mechanical signals from the ECM to the cell is shown. Increased levels of β-actin and heat shock protein 27 (HSP27) after weight loss are compatible with increased mechanical stress and are associated with higher weight regain63. Variations in COL23A1, encoding collagen XXIII; FBLN5, encoding fibulin 5; LAMB1, encoding laminin subunit-β1 (which is a component of laminin 111); and FN1, encoding fibronectin were found to be associated with weight regain65. Epidermal growth factor receptor (EGFR)-mediated phosphorylation of α-actinin, a β-actin crosslinking protein and target of focal adhesion kinase, might be involved in the risk of weight regain69.

A study of female participants who underwent an 8-week LCD in the European DiOGenes intervention61 compared the adipose tissue transcriptome between individuals who maintained weight loss (0–10% weight regain) and individuals who regained weight (10–20% weight regain) during the 6-month follow-up. The investigators found that the focal adhesion pathway was upregulated during the LCD in individuals who regained weight versus individuals who maintained weight loss. Upregulation of the focal adhesion pathway suggests that mechanical stress after the diet was higher in adipocytes of weight regainers than in those of weight maintainers.

Another group62 studied changes in stress-related proteins in biopsy samples of WAT over an 8-week period of calorie restriction. From a cohort of 200 participants, the investigators selected 18 men and women who lost at least 10% of their body weight over the 8-week calorie restriction period. Median weight regain was 6 kg over the 12-month ad libitum follow-up period. The authors reported that, at the end of the weight loss period, the levels of heat shock protein 60 (HSP60) and HSP70 were increased significantly in people who regained >6 kg compared with people who regained <6 kg over the 12-month follow-up period. The investigators also found a trend for increased levels of β-actin and HSP27 in people who regained >6 kg. β-Actin and HSP27 are components of actin filaments, and HSP27 phosphorylation is a regulator of actin stress fibres and focal adhesion formation63. Furthermore, increased levels of HSP60 and β-actin were also observed in cultured adipocytes after glucose restriction62. The weight loss intervention did not have any effect on SOD1 and SOD2, which are markers of oxidative stress, suggesting that oxidative stress was not influencing weight regain. The findings of this study suggest that weight regain following weight loss induced by an LCD intervention results from increased mechanical stress in the adipose tissue and/or adipocytes.

Another study that was focused on genes that encode stress-related proteins analysed microarray data from the YoYo intervention study. The investigators included 107 genes and found expression changes in 17 of these genes during the different phases of the weight loss intervention (weight loss phase by means of an LCD or a VLCD (12 or 5 weeks, respectively), the subsequent 4-week weight stabilization phase and the combined weight loss and weight stabilization phase) that correlated with weight regain during follow-up64. In the VLCD group, eight of the nine genes whose expression changes during the weight stabilization phase were correlated with weight regain during the weight stabilization phase were linked to nutrient sensing, glucose handling and actin filament dynamics of the adipose tissue64.

Further to the aforementioned studies, genetic and genomic evidence supports the adipocyte stress model for weight regain and a role for the ECM. On the basis of genome-wide association study data from the European DiOGenes intervention study, researchers found that variation in genes encoding proteins of the ECM is associated with the risk of weight regain in a sex-specific manner65. In males, weight regain-associated variations were reported in the genes that encode periostin (POSTN), laminin subunit-β1 (LAMB1), collagen type XXIII α1 (COL23A1) and fibulin 5 (FBLN5), whereas in females, a weight regain-associated variation was reported for the gene that encodes fibronectin 1 (FN1). In addition, these data suggest a link with focal adhesions (Fig. 4), as collagen XXIII and the LAMB1-containing laminin 111 interact with α2β1 integrin66 and, in mice, FBLN5 competes with FN1 for binding to β1 integrin67.

In the YoYo intervention study, the expression change of 16 out of 277 genes encoding ECM proteins over different phases of the VLCD intervention correlated with weight regain68. For 11 of these genes, the correlated change in expression occurred over the 4-week weight stabilization phase. Using these genes to build a correlation co-expression network, the genes that could be clustered in the network seemed to suggest changes in adipose tissue inflammation, insulin sensitivity and ECM remodelling during the weight-stable phase. Moreover, stress-related genes and genes that encode ECM proteins seemed to interact in relation to the risk of weight regain69. Gene–gene interaction analysis suggested that epidermal growth factor receptor-mediated phosphorylation of α-actinin, a β-actin crosslinking protein and target of focal adhesion kinase, is involved in the risk of weight regain69 (Fig. 4).

Fatty acid metabolism and weight regain

The pull model8, which is proposed as a mechanism to explain why some individuals on an isocaloric diet might develop more fat mass than others, predicts that adipocyte-autonomous processes can pull substrates preferentially into adipocytes. One group has suggested that weight loss can trigger an increased pull of substrates into the adipocytes, which in turn leads to weight regain70. In support of this model, data show that ≥1 year after bariatric surgery, high maximal capacity for long-chain fatty acid uptake persists in isolated adipocytes despite a reduction in adipocyte size71. In vivo research suggests that an increase in the storage capacity of fatty acids per volume of adipose tissue after bariatric surgery exactly compensates for the decrease in fat mass during weight loss72.

Proteomics experiments on human adipose tissue biopsy samples taken before and 3 weeks after a 5-week VLCD intervention revealed that shortly after weight loss glucose transporter type 4 (GLUT4) is translocated to the cell membrane, and perilipin 1 isoform B, which is proposed to be involved in fatty acid uptake and positively correlated with LPL, is strongly upregulated73. These data suggest that adipocytes, apart from developing stress during the weight loss phase, prepare for renewed energy uptake immediately after host energy balance is restored (such as upon termination of an LCD)74. Adipocyte stress and an increased capacity for glucose and fatty acid uptake might be linked; however, cause and effect in this respect have not yet been studied.

In the YoYo study, we measured in vivo meal-derived fatty acid uptake by subcutaneous adipose tissue before and at the end of the intervention (which consisted of a diet-induced weight loss phase followed by a 4-week weight stabilization phase). We applied the arteriovenous balance technique across abdominal subcutaneous adipose tissue, where blood is sampled from catheters inserted in a superficial epigastric vein draining subcutaneous WAT and in an artery or heated hand vein, thus retrieving arterialized venous blood. This method enables measurement of plasma concentration differences across subcutaneous WAT and, in combination with the measurement of WAT blood flow by the xenon wash-out technique, can be used to calculate net fluxes of metabolites. We combined this methodology with a 13C-palmitate tracer added to a high-fat mixed meal, which allowed us to also measure meal-derived fatty acid uptake in subcutaneous WAT75. We found that meal-derived fatty acid uptake per volume of WAT was unchanged after weight loss, which is consistent with data from elsewhere76 that were obtained using a different stable isotope tracer technique. Moreover, in our study, weight loss-induced changes in meal-derived fatty uptake were not associated with weight regain75. These data do not support the pull model, at least not for meal-derived fatty acids in the postprandial state. Nevertheless, the average diameter of the adipocytes increased during the 4-week energy balance period compared with immediately at the end of the energy-restricted diet, suggesting a change in adipose tissue fatty acid metabolism after weight loss. One possibility is that de novo lipogenesis increases in the adipocytes following weight loss, which is supported by studies that have reported an upregulation of fatty acid synthase73,77.

The effect of weight loss on adipose tissue lipolysis has been the topic of a number of studies. A review of the literature57 concluded that a negative energy balance stimulates fasting adipose tissue lipolysis and its responsiveness to neural and hormonal stimulation. The degree of stimulation of fasting lipolysis seems to depend on the size and duration of the negative energy balance57. With prolonged weight loss maintenance, that is, without a concomitant negative energy balance, fasting lipolysis and its responsiveness to stimulation are reduced compared with baseline57.

In the YoYo study, we found a positive association between the decrease in fasting plasma concentrations of free fatty acid (FFA) from baseline to the end of the intervention phase and subsequent weight regain52. Low FFA concentrations might signal a shortage of energy stores, which has been suggested to promote energy intake70. Therefore, one of the factors that could explain variations in weight regain after weight loss between individuals is the extent to which fasting lipolysis is suppressed after weight reduction. Whether the observed differences in the suppression of fasting lipolysis are due to variations in the extent to which adipocytes decrease in size is unclear. Whereas the average reduction in adipocyte size was not reported to be associated with weight regain52, a role for differences in the change in size distribution cannot be excluded70.

A low fasting fatty acid rate of appearance phenotype has been described in obesity, which was beneficial for insulin sensitivity78. The authors compared adipose tissue factors (plasma metabolites, mRNA and protein expression in subcutaneous WAT) in adults with obesity who had a low or high fasting fatty acid rate of appearance. Individuals with a low fasting fatty acid rate of appearance were characterized by reduced HSL gene expression, increased expression of genes involved in fatty acid re-esterification, reduced ECM deposition and reduced fibrosis, suggesting an increased capacity for adipocyte expansion. Other mechanisms that might be involved are interindividual differences in the response of adipose tissue blood flow and genetics.

Following weight loss, there is an associated reduction in resting energy expenditure, which provides another line of evidence supporting a role for reduced fasting lipolysis in weight regain. Weight loss lowers resting energy expenditure ~15% more than would be expected on the basis of the change in body composition (this process is referred to as adaptive thermogenesis)79,80. This extra reduction results in a risk of weight regain if it is not compensated for by increased activity-induced energy expenditure or a lower energy intake81. To date, the molecular basis for this phenomenon has not been defined.

After an 8-week VLCD, we observed that the decrease in resting energy expenditure correlated with the change in abundance of hydroxyacyl-CoA-dehydrogenase (HADH) in subcutaneous adipose tissue82. HADH is a crucial enzyme for mitochondrial β-oxidation, suggesting that reduced activation of the β-oxidation underlies the extra weight loss-associated decrease in resting energy expenditure. However, a reduction in the activation of lipolysis could be the reason for the observed extra decrease in resting energy expenditure, as it was shown that changes in the activity of HADH and other enzymes of β-oxidation in adipose tissue correlate positively with changes in the plasma levels of FFA83. A study investigated mitochondrial capacity in WAT by measuring the expression of genes involved in nuclear-encoded mitochondrial pathways, mitochondrial DNA gene expression, subunits of the oxidative phosphorylation machinery and the number of mitochondria in a weight loss intervention. The investigators found that baseline mitochondrial capacity was negatively associated with weight regain84. In another study, transcriptomics analysis showed that, in individuals with continued weight loss, mitochondrial gene expression was upregulated, whereas it was downregulated in individuals with weight regain on an ad libitum diet after a period of energy restriction85.

These observations suggest that, when a deficit in energy from the diet during calorie restriction cannot be compensated for by lipolysis and/or β-oxidation, energy expenditure will adjust to a reduced level, which persists after weight loss and increases the risk of weight regain. Thus, changes in the concentration of plasma FFA and HADH in adipose tissue during weight loss are possible markers for the risk of weight regain.

One further factor that might have a role in an individual’s susceptibility to weight regain after weight loss is the degree to which brown adipose tissue (BAT) activity changes or browning of WAT occurs. Studies have shown that BAT volume increases after bariatric surgery in mice86 as well as in humans87,88. By contrast, no effect of caloric restriction on browning of subcutaneous abdominal WAT was found in individuals with obesity89, although this effect of weight loss has been demonstrated in mice90,91. At this moment, no studies have investigated the potential association between the weight loss-induced increase in BAT activity or browning of WAT and weight regain in animals or humans and, therefore, the importance of these factors is currently unclear.

Adipose tissue as an endocrine organ

Adipose tissue has been considered an important endocrine organ since the discovery of leptin almost 25 years ago92, although the role of adipose tissue as a source of hormones was first described in 1987 in the context of adipose tissue-derived oestrogens and breast cancer in women with obesity93. Currently, we know that the adipose tissue secretome is very diverse and that it has an important role not only in the crosstalk between adipose tissue and other organs but also in a paracrine or autocrine manner94. Apart from the classical adipokines such as leptin and adiponectin, various other factors are secreted by adipose tissue, such as enzymes, lipokines and extracellular vesicles that transport bioactive molecules, such as lipids, proteins and microRNAs95. Furthermore, adipocytes secrete metabolites such as FFA and glycerol (the potential role of FFA in weight regain is discussed earlier in the manuscript). Factors secreted by cells other than adipocytes (such as macrophages) in adipose tissue also contribute to the secretome of adipose tissue96. Several studies52,97,98,99 have investigated the relationship between weight loss-induced changes to the secretome of adipose tissue and weight regain, which we describe in detail in the following section.

Changes to the adipose tissue secretome

Leptin is an obvious candidate in the regulation of the replenishment of fat depots after weight loss. Several studies have investigated the association between plasma leptin levels and weight regain97. Although some studies suggest that either a large decrease in leptin during weight loss or the baseline leptin level predicts weight regain, most studies did not detect such associations97. Studies that investigated the association between the weight loss-induced change in leptin concentration and weight regain are complicated by the fact that energy restriction has a very prominent effect on lowering plasma leptin concentrations on top of the effect of fat mass loss100. Measuring changes in plasma leptin concentrations at the end of a weight loss intervention, when participants are still in a negative energy balance, might therefore give different results than when leptin is measured after weight loss in the absence of a negative energy balance.

With the possible measurement differences in mind, in the YoYo study, we measured leptin after a 4-week period in which participants were weight stable after a weight loss intervention. Neither the change in leptin from baseline to the 4-week weight-stable measurement time point nor the leptin level at this measurement time point was associated with weight regain52. These data add to the evidence that there is no association between weight loss-induced changes in leptin plasma concentration and weight regain. Concomitant changes in leptin sensitivity, however, cannot be excluded. A 2017 study101 reported that leptin receptor expression in human adipose tissue was increased after bariatric surgery, although this might not reflect receptor expression in the hypothalamus.

Leptin links adipose tissue dynamics directly to brain responses. Adiponectin, along with several other cytokines from adipose tissue, is also known or suggested to cross the blood–brain barrier and act centrally102. Intracerebroventricular administration of adiponectin has been shown to result in the loss of fat mass in mice and rats by increasing energy expenditure independent of leptin103,104. In the YoYo study52, we looked at a number of adipokines with known or suspected functions in the regulation of energy metabolism and body weight, such as IL-6, retinol-binding protein 4 (RBP4), apelin, adiponectin, vaspin, nesfatin 1 and angiotensin-converting enzyme (ACE) activity. We found that weight loss-induced changes in RBP4 and ACE activity were associated with weight regain, but no association was found for the other factors.

RBP4, a pro-inflammatory adipokine that is associated with insulin resistance105,106,107, is predominantly secreted by the liver but is also secreted by adipose tissue106,107. Secretion of RBP4 is elevated in obese WAT owing to the presence of hypertrophic adipocytes107. Presently, whether weight regain and plasma concentration of RBP4 are causally related remains unclear, but the pro-inflammatory action of RBP4 could be involved.

ACE is a carboxypeptidase that is expressed in many tissues including adipose tissue but predominantly in the vascular endothelium and the lung. The enzyme is anchored to cell membranes and is shed into the plasma by enzymatic cleavage108. The weight loss-induced change in plasma ACE concentration was reported to predict continued weight loss on an ad libitum diet after a period of energy restriction in women in the European multicentre DiOGenes study98. Furthermore, a positive association between the weight loss-induced change in serum ACE concentration and weight regain during the ad libitum phase was found in men in the same study99. By contrast, in the YoYo study, the change in plasma ACE activity over the weight loss intervention (consisting of a weight loss phase followed by a 4-week weight stabilization phase) was negatively associated with weight regain52. The discrepancy between the two studies could be related to the different time points at which ACE was measured: at the end of the weight loss intervention in the DiOGenes study99 or after a subsequent weight stabilization phase in the YoYo study52. Whether a causal relation exists between ACE and weight regain remains to be determined, and the underlying mechanism is also unclear. We know that ACE might have a role in many processes other than the regulation of blood pressure, such as inflammation, fibrosis and the regulation of food intake by the hypothalamus108, but whether plasma concentrations of ACE or ACE activity reflect inflammation, fibrosis and regulation of food intake is unclear.

MicroRNAs in weight regain

Research has shown that microRNAs influence adipose tissue development and function, and as such, they might also be involved in regulating weight regain after weight loss109,110. Specific microRNAs can either promote or inhibit adipogenesis, for instance, via the regulation of adipose stem cell development and by lineage determination111. RNAs of the miR-30 family are able to target RUNX2 and therefore influence the decision of stem cells to become committed to the adipogenic or osteogenic lineage112. Another microRNA that can influence adipocyte morphology and lineage commitment is miR-181d. This microRNA determines the enzymatic activity of ADAMTS1, which is involved in ECM remodelling and in determining adipose tissue mass, insulin sensitivity and lipid homeostasis113. Lipid handling by adipocytes, including lipolysis, is influenced by microRNA targeting of peroxisome proliferator-activated receptor-γ (PPARγ)114.

In humans, miR-26 is upregulated during early adipogenesis and seems to have a role in the browning of adipocytes, regulating lipid accumulation, mitochondrial morphology and uncoupled respiration115. The shift between white and beige adipogenesis is mediated by the miR-26 target ADAM17.

The microRNA profile of adipose tissue is changed in individuals who are obese116. A role for microRNAs, in particular, miR-365 and miR-574, is proposed in adipose tissue hypertrophy via the regulation of transcription factor EBF1 (ref.117). As such, microRNAs are thought to contribute to the development of obesity-related comorbidities such as insulin resistance and cardiovascular disorders. Indeed, research has identified several microRNAs that can influence insulin sensitivity. Interestingly, as is the case with adipokines, microRNAs contained in exosomes can be excreted from the adipose tissue118. In this way, they can travel to other tissues and organs to modify whole-body metabolism.

Obesity-induced changes in the concentration of microRNA within adipocytes might promote the chemotaxis of macrophages and other immune cells towards the adipocyte119,120. As such, microRNAs could contribute to a state of chronic low-grade inflammation within adipose tissues, as observed in obesity. In addition, adipocytes alter their microRNA profile when exposed to TNF or other cytokines121. Thus, under the influence of inflammation, the secretion of microRNAs from adipose tissues changes, and this might contribute to the development of insulin resistance. Of note, one study reported that 2 years after laparoscopic Roux-en-Y gastric bypass surgery in 16 women with morbid obesity, the obesity-related and inflammation-related impairment of microRNA target gene expression had improved122.

Altogether, the effect that microRNAs seem to have on adipogenesis, adipose tissue functioning and whole-body metabolism makes it highly probable that they are involved in the risk of weight regain after weight loss. Notably, the consequences of weight cycling can be modified by microRNAs. In mice, high levels of hypothalamic miR-219 after weight cycling are associated with increased risk of metabolic syndrome123. More knowledge of possible similar influences of adipose tissue-derived microRNAs on weight regain would be welcome, as microRNAs could be a promising therapeutic target for preventing weight regain after weight loss and the consequences of weight cycling118.

Regulation of gene expression by microRNAs is usually categorized as epigenetic modification. In this respect it is noteworthy that changes in DNA methylation, which is another type of epigenetic modification, were observed in human subcutaneous adipose tissue during weight regain or continued weight loss in the course of a 1-year weight loss intervention124. Similarly, chromatin modifications might influence weight loss and regain125. However, these studies are only descriptive and require functional follow-up studies.

Adipose tissue inflammation

As mentioned before, obesity is often accompanied by chronic low-grade inflammation of the adipose tissue with an increase in pro-inflammatory markers in the circulation126. Although one might expect that the inflammatory status would be lowered upon weight loss, gene expression measurements during weight intervention studies show that the adipose tissue inflammatory pathways are in fact upregulated during calorie restriction and are gradually downregulated during the 3–9-month follow-up periods after the weight loss intervention127,128,129. Following weight loss after bariatric surgery, adipose tissue retains its level of inflammation as if it memorizes the obese state despite continued weight reduction130. Furthermore, during the active weight loss phase following bariatric surgery, plasma markers of low-grade inflammation are unchanged or increased and decline only during the weight maintenance phase52,131,132. In the YoYo study, co-expression analysis of ECM genes in the 4-week weight-stable period following a VLCD intervention showed that genes encoding various leukocyte-specific integrins correlated with the risk of weight regain68. These gene expression data suggest that individuals in whom adipose tissue retains a high level of leukocyte activity in the first weeks after weight loss — which might be a marker for ongoing tissue stress — are at an increased risk of weight regain. On the other hand, individuals who have a resistance to adipose tissue inflammation would be expected to be at a reduced risk of weight regain after weight loss. These findings seem to comply with the existence of an obesogenic memory130, which in mice might be mediated by CD4+ T cells133.

As described earlier, a weight loss-induced increase in levels of plasma RBP4 was associated with weight regain in the YoYo study52. Data from the DiOGenes study also suggested that inflammation is involved in weight regain134. A continued inflammatory state would be expected to increase adipose tissue lipolysis and energy expenditure135,136, which could be considered beneficial for the prevention of fat mass regain. However, how these two opposing mechanisms might interact in the process of weight regain following weight loss remains to be seen.

Models to predict weight regain

Although many factors, such as weight loss-induced effects on adipose tissue but also behavioural and environmental factors, have been shown to correlate with weight regain after diet-induced weight loss, the correlation of each factor alone with weight regain is usually low, and the independent contribution of each factor is unknown. Few studies have tested whether a combination of these factors can be used to predict successful weight loss maintenance.

One study137 used data from the DiOGenes trial, which studied weight regain after weight loss by caloric restriction in 191 individuals without T2DM who were overweight or obese. In this study, the authors analysed the adipose tissue transcriptome at baseline and investigated how it changed in response to a LCD. They combined this analysis with clinical variables, including BMI, Matsuda index (a measure of insulin sensitivity), age and sex, at baseline and how BMI and Matsuda index changed during the LCD with the aim of developing a model to identify individuals who are susceptible to weight regain after weight loss.

The authors found that 18 of the 1,173 differentially expressed genes during the LCD were associated with both BMI changes and Matsuda index changes during the calorie restriction phase. At the end of the 6-month follow-up, the model that was the most accurate at classifying individuals who had BMI changes above or below the median from baseline included weight loss-induced changes in clinical variables and baseline adipose tissue gene expression of the 18 identified genes137. Compared with a model that included only the clinical variables, this model was significantly more accurate at predicting whether an individual would regain weight or maintain the reduced weight following the LCD intervention (P = 0.012; area under the receiver operating characteristic curve 0.73 and 0.87, respectively).

In a study of 100 individuals138, the investigators combined various psychosocial variables with simple clinical baseline variables with the aim of predicting successful long-term (60 months) weight reduction (≥10% from baseline) after an 8-week VLCD intervention. Including the psychosocial variables did not improve the prediction model compared with a model based on baseline clinical variables alone (sex, BMI, waist-to-hip ratio and insulin sensitivity) (88% versus 86% of participants classified correctly).

A further study134 analysed predictors of body weight change using a Bayesian analysis in a study that included a 6-week energy restriction phase followed by a 6-week weight maintenance diet prescription phase. A large number of potential predictors were studied, including metabolic and inflammatory variables and gut microbiota composition. A combination of baseline fasting plasma insulin and IL-6 concentrations, plasma leukocyte number and the number of HAM56+ cells (macrophages) in subcutaneous WAT was able to predict resistance to weight loss and proneness to weight regain with a prediction accuracy of 75%134.

These models are far from being able to accurately predict individual weight loss success, and research in this area needs to be further expanded before this goal will become a reality. Large data sets that include a variety of potential predictors of different types (biological, behavioural and environmental) will be necessary for this type of analysis.

Strategies to prevent weight regain

Several strategies unrelated to adipose tissue to limit weight regain after successful weight loss have been investigated in adults with obesity, some of which have been effective (Box 1). In this section, we discuss strategies based on adipose tissue-related factors that show promise in predicting weight regain (Fig. 5).

Fig. 5: Working model for the role of weight loss-induced changes in adipose tissue in weight regain.
Fig. 5

Loss of fat mass induces shrinkage of adipocytes, which is accompanied by cell stress, inflammation, altered adipokine secretion (for example, retinol-binding protein 4 (RBP4)) and reduced lipolysis. These changes may affect energy balance by reducing energy expenditure (EE) and increasing energy intake, thus promoting regain of fat mass and body weight and relieving stress. In addition, the cells express an increased capacity for glucose uptake (glucose transporter type 4 (GLUT4)) and fatty acid uptake (perilipin 1 isoform B (PLINB)), which facilitates renewed fat storage73. ECM, extracellular matrix.

One strategy could be to try to reduce the cellular stress response that is associated with weight loss-induced reduction in adipocyte size by increasing the fluidity of the adipocyte membrane. Omega-3 polyunsaturated fatty acids have been shown to increase membrane fluidity139, but whether using them as a treatment will reduce the weight loss-induced stress response is unknown.

Another strategy could be to stimulate lipolysis, which would reduce the nutrient shortage signalling to the brain70 and increase energy expenditure in individuals who have low plasma FFA levels after weight loss. Strategies that increase browning of WAT, such as cold exposure140 or nutraceuticals (including conjugated linoleic acid)141, are options that could be tested in individuals with low plasma FFA levels after weight loss. Regular exercise can also stimulate lipolysis and has been shown to be beneficial for the prevention of weight regain after weight loss (Box 1). Whether regular exercise induces browning of WAT in humans is still a matter of debate, but most human studies have reported no increases in browning following exercise training142,143,144,145. Animal studies suggest that regular exercise can diminish the drive for weight regain by shifting energy metabolism to decrease de novo lipogenesis and increase plasma FFA concentration146,147. The effect of pharmaceutical agents that increase lipolysis also deserves study in the context of weight regain, although lipolysis-stimulating agents such as β-adrenergic agonists will be of limited practical use because they are associated with adverse effects, including muscle tremor, elevated blood pressure, tachycardia and insomnia148.

Although the mechanisms underlying weight regain are far from clear, a potential strategy to limit weight regain is to try to reduce inflammation in individuals who maintain high levels of inflammation after weight loss. Several approaches to achieve a reduction in inflammation can be considered: anti-inflammatory drug treatment and exercise and dietary strategies (including alterations to diet composition or dietary supplements). Several anti-inflammatory drugs have been tested in individuals with metabolic disturbances associated with obesity, but their effects on body weight have not been reported107,149,150. Furthermore, none of these drugs have been tested in the context of weight maintenance after weight loss.

Some investigators have suggested TNF inhibition in combination with recombinant IL-33 administration as a potential strategy to maintain healthy adipose tissue immune function107. Regular exercise is known to reduce inflammation in humans, and several mechanisms for this effect have been proposed, such as a reduction in adipose tissue mass, increased production of anti-inflammatory cytokines by active skeletal muscle and a reduction in the release of circulating pro-inflammatory cytokines by macrophages and monocytes151. Reduced macrophage infiltration in adipose tissue and beneficial phenotypic switching of adipose tissue macrophages by exercise have been reported in rats151. However, the effect of exercise on adipose tissue inflammation in humans remains unclear152. One study153 reported an increase in CD163+ cells after exercise training in the adipose tissue of individuals who were overweight or had obesity. CD163+ cells are supposed to be anti-inflammatory. Several studies have shown that individuals who are active after a weight loss intervention regain less weight than individuals who are inactive48,154,155,156,157 (Box 1), but whether a reduction in inflammation is involved remains unclear.

Inflammation can also be influenced by diet. One of the potentially anti-inflammatory nutraceuticals is resveratrol. The anti-inflammatory properties of resveratrol have been clearly demonstrated in animal studies150. The suggested mechanisms are primarily decreased activation of NF-κB and reduced adipose tissue macrophage infiltration150. Nevertheless, the studies on the anti-inflammatory action of resveratrol in humans have been equivocal150. A 2015 clinical trial in individuals with obesity and metabolic syndrome found no effect of low-dose or high-dose resveratrol treatment on adipose tissue inflammation, circulating inflammatory markers or body composition158. Another strategy to reduce inflammation might be supplementation with omega-3 polyunsaturated fatty acids, which reduce inflammation via a variety of mechanisms159. To date, however, neither resveratrol nor omega-3 polyunsaturated fatty acids have been tested in the context of weight loss maintenance.

Diet composition has been shown to influence the level of inflammatory markers in the circulation, and an index (called the Dietary Inflammatory Index) has been developed to reflect the inflammatory properties of the diet160,161. In a non-overweight adult cohort of Spanish individuals, the Dietary Inflammatory Index score was not associated with BMI at baseline, but a high score was associated with increased body weight and an increased risk of developing overweight or obesity after 8 years of follow-up162. In the YoYo study, we found that a high Dietary Inflammatory Index score was associated with increased weight regain over the follow-up period163. The circulating concentrations of individual inflammatory markers, such as IL-6 and RBP4, showed mixed results, with weight loss-induced changes in RBP4 being found to be associated with weight regain, whereas changes in IL-6 were not52. Clinical trials that compare diets with a low or high Dietary Inflammatory Index score after a weight loss intervention are needed to answer the question of whether the inflammatory properties of diet have a role in weight regain and whether the underlying mechanism is related to adipose tissue function.

Mouse studies have suggested that local insulin resistance in adipose tissue leads to secretion of MCP1 and activation of pro-inflammatory macrophages46. Therefore, another strategy to prevent weight regain after weight loss could be targeting adipose tissue insulin resistance. In addition, as we described earlier, targeting adipose tissue microRNAs for the prevention of weight regain might be a promising strategy, but this approach awaits further exploration. The same can be said for epigenetic marks like DNA methylation and chromatin modification.

Finally, altering adipokine production and/or secretion or mimicking their effect by drugs is another strategy that might prove successful at preventing weight regain, although the role of adipokines in weight regain is still far from clear. A study164 investigated the effect of metreleptin (a synthetic leptin analogue for the treatment of lipodystrophy) in combination with pramlintide treatment on body weight after an initial weight loss period by dietary energy restriction in individuals with obesity. The addition of metreleptin to the treatment regimen was associated with a further decrease in body weight compared with pramlintide alone, but the duration of the study was too short to provide information on weight regain164.


In this Review, we have summarized the evidence that shows that the interindividual differences in weight loss-induced changes in subcutaneous adipose tissue have a role in the interindividual differences in weight regain. As most of the evidence comes from associations, further research on the causality of these associations might eventually lead to the development of targeted strategies to effectively prevent weight regain after successful weight loss. Promising research areas are weight loss-induced variations in cellular stress, ECM remodelling, inflammatory responses, adipokine secretion, lipolysis and epigenetic modifications, including regulation by microRNAs.

Additional information

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.


  1. 1.

    NCD Risk Factor Collaboration (NCD-RisC). Trends in adult body-mass index in 200 countries from 1975 to 2014: a pooled analysis of 1698 population-based measurement studies with 19.2 million participants. Lancet 387, 1377–1396 (2016).

  2. 2.

    Santos, I., Sniehotta, F. F., Marques, M. M., Carraca, E. V. & Teixeira, P. J. Prevalence of personal weight control attempts in adults: a systematic review and meta-analysis. Obes. Rev. 18, 32–50 (2017).

  3. 3.

    Anderson, J. W., Konz, E. C., Frederich, R. C. & Wood, C. L. Long-term weight-loss maintenance: a meta-analysis of US studies. Am. J. Clin. Nutr. 74, 579–584 (2001).

  4. 4.

    Look AHEAD Research Group. Eight-year weight losses with an intensive lifestyle intervention: the look AHEAD study. Obesity (Silver Spring) 22, 5–13 (2014).

  5. 5.

    Christou, N. V., Look, D. & Maclean, L. D. Weight gain after short- and long-limb gastric bypass in patients followed for longer than 10 years. Ann. Surg. 244, 734–740 (2006).

  6. 6.

    Magro, D. O. et al. Long-term weight regain after gastric bypass: a 5-year prospective study. Obes. Surg. 18, 648–651 (2008).

  7. 7.

    Odom, J. et al. Behavioral predictors of weight regain after bariatric surgery. Obes. Surg. 20, 349–356 (2010).

  8. 8.

    Schwartz, M. W. et al. Obesity pathogenesis: an Endocrine Society scientific statement. Endocr. Rev. 38, 267–296 (2017).

  9. 9.

    Hopkins, M. & Blundell, J. E. Energy balance, body composition, sedentariness and appetite regulation: pathways to obesity. Clin. Sci. 130, 1615–1628 (2016).

  10. 10.

    MacLean, P. S., Blundell, J. E., Mennella, J. A. & Batterham, R. L. Biological control of appetite: a daunting complexity. Obesity (Silver Spring) 25 (Suppl. 1), S8–S16 (2017).

  11. 11.

    Dulloo, A. G., Jacquet, J., Miles-Chan, J. L. & Schutz, Y. Passive and active roles of fat-free mass in the control of energy intake and body composition regulation. Eur. J. Clin. Nutr. 71, 353–357 (2017).

  12. 12.

    Melby, C. L., Paris, H. L., Foright, R. M. & Peth, J. Attenuating the biologic drive for weight regain following weight loss: must what goes down always go back up? Nutrients 9, 468 (2017).

  13. 13.

    Ochner, C. N., Barrios, D. M., Lee, C. D. & Pi-Sunyer, F. X. Biological mechanisms that promote weight regain following weight loss in obese humans. Physiol. Behav. 120, 106–113 (2013).

  14. 14.

    Sumithran, P. & Proietto, J. The defence of body weight: a physiological basis for weight regain after weight loss. Clin. Sci. 124, 231–241 (2013).

  15. 15.

    Mariman, E. C. Human biology of weight maintenance after weight loss. J. Nutrigenet. Nutrigenomics 5, 13–25 (2012).

  16. 16.

    Sumithran, P. et al. Long-term persistence of hormonal adaptations to weight loss. N. Engl. J. Med. 365, 1597–1604 (2011).

  17. 17.

    Munzberg, H., Laque, A., Yu, S., Rezai-Zadeh, K. & Berthoud, H. R. Appetite and body weight regulation after bariatric surgery. Obes. Rev. 16 (Suppl. 1), 77–90 (2015).

  18. 18.

    Lean, M. E. & Malkova, D. Altered gut and adipose tissue hormones in overweight and obese individuals: cause or consequence? Int. J. Obes. (Lond.) 40, 622–632 (2016).

  19. 19.

    Verhoef, S. P., Camps, S. G., Bouwman, F. G., Mariman, E. C. & Westerterp, K. R. Physiological response of adipocytes to weight loss and maintenance. PLOS ONE 8, e58011 (2013).

  20. 20.

    Haczeyni, F., Bell-Anderson, K. S. & Farrell, G. C. Causes and mechanisms of adipocyte enlargement and adipose expansion. Obes. Rev. 19, 406–420 (2018).

  21. 21.

    Park, K. W., Halperin, D. S. & Tontonoz, P. Before they were fat: adipocyte progenitors. Cell Metab. 8, 454–457 (2008).

  22. 22.

    Engin, A. Fat cell and fatty acid turnover in obesity. Adv. Exp. Med. Biol. 960, 135–160 (2017).

  23. 23.

    Spalding, K. L. et al. Dynamics of fat cell turnover in humans. Nature 453, 783–787 (2008).

  24. 24.

    Tchoukalova, Y. D. et al. Regional differences in cellular mechanisms of adipose tissue gain with overfeeding. Proc. Natl Acad. Sci. USA 107, 18226–18231 (2010).

  25. 25.

    Arner, E. et al. Adipocyte turnover: relevance to human adipose tissue morphology. Diabetes 59, 105–109 (2010).

  26. 26.

    Jonker, J. W. et al. A PPARgamma-FGF1 axis is required for adaptive adipose remodelling and metabolic homeostasis. Nature 485, 391–394 (2012).

  27. 27.

    Arner, P., Andersson, D. P., Backdahl, J., Dahlman, I. & Ryden, M. Weight gain and impaired glucose metabolism in women are predicted by inefficient subcutaneous fat cell lipolysis. Cell Metab. 28, 45–54 (2018).

  28. 28.

    Khan, T. et al. Metabolic dysregulation and adipose tissue fibrosis: role of collagen VI. Mol. Cell. Biol. 29, 1575–1591 (2009).

  29. 29.

    Henegar, C. et al. Adipose tissue transcriptomic signature highlights the pathological relevance of extracellular matrix in human obesity. Genome Biol. 9, R14 (2008).

  30. 30.

    Muir, L. A. et al. Adipose tissue fibrosis, hypertrophy, and hyperplasia: correlations with diabetes in human obesity. Obesity (Silver Spring) 24, 597–605 (2016).

  31. 31.

    Divoux, A. et al. Fibrosis in human adipose tissue: composition, distribution, and link with lipid metabolism and fat mass loss. Diabetes 59, 2817–2825 (2010).

  32. 32.

    Trayhurn, P. Hypoxia and adipocyte physiology: implications for adipose tissue dysfunction in obesity. Annu. Rev. Nutr. 34, 207–236 (2014).

  33. 33.

    Sun, K., Tordjman, J., Clement, K. & Scherer, P. E. Fibrosis and adipose tissue dysfunction. Cell Metab. 18, 470–477 (2013).

  34. 34.

    Lin, N. & Simon, M. C. Hypoxia-inducible factors: key regulators of myeloid cells during inflammation. J. Clin. Invest. 126, 3661–3671 (2016).

  35. 35.

    Lin, Q. & Yun, Z. The hypoxia-inducible factor pathway in adipocytes: the role of HIF-2 in adipose inflammation and hypertrophic cardiomyopathy. Front. Endocrinol. (Lausanne) 6, 39 (2015).

  36. 36.

    Pasarica, M. et al. Reduced adipose tissue oxygenation in human obesity: evidence for rarefaction, macrophage chemotaxis, and inflammation without an angiogenic response. Diabetes 58, 718–725 (2009).

  37. 37.

    Goossens, G. H. et al. Increased adipose tissue oxygen tension in obese compared with lean men is accompanied by insulin resistance, impaired adipose tissue capillarization, and inflammation. Circulation 124, 67–76 (2011).

  38. 38.

    Vink, R. G. et al. Diet-induced weight loss decreases adipose tissue oxygen tension with parallel changes in adipose tissue phenotype and insulin sensitivity in overweight humans. Int. J. Obes. (Lond.) 41, 722–728 (2017).

  39. 39.

    Ellulu, M. S., Patimah, I., Khaza’ai, H., Rahmat, A. & Abed, Y. Obesity and inflammation: the linking mechanism and the complications. Arch. Med. Sci. 13, 851–863 (2017).

  40. 40.

    Domenis, R. et al. Pro inflammatory stimuli enhance the immunosuppressive functions of adipose mesenchymal stem cells-derived exosomes. Sci. Rep. 8, 13325 (2018).

  41. 41.

    Engin, A. B. Adipocyte-macrophage cross-talk in obesity. Adv. Exp. Med. Biol. 960, 327–343 (2017).

  42. 42.

    Fantuzzi, G. Adipose tissue, adipokines, and inflammation. J. Allergy Clin. Immunol. 115, 911–919 (2005).

  43. 43.

    Rosen, E. D. & Spiegelman, B. M. What we talk about when we talk about fat. Cell 156, 20–44 (2014).

  44. 44.

    Kratz, M. et al. Metabolic dysfunction drives a mechanistically distinct proinflammatory phenotype in adipose tissue macrophages. Cell Metab. 20, 614–625 (2014).

  45. 45.

    Alligier, M. et al. Subcutaneous adipose tissue remodeling during the initial phase of weight gain induced by overfeeding in humans. J. Clin. Endocrinol. Metab. 97, E183–E192 (2012).

  46. 46.

    Shimobayashi, M. et al. Insulin resistance causes inflammation in adipose tissue. J. Clin. Invest. 128, 1538–1550 (2018).

  47. 47.

    Maclean, P. S., Bergouignan, A., Cornier, M. A. & Jackman, M. R. Biology’s response to dieting: the impetus for weight regain. Am. J. Physiol. Regul. Integr. Comp. Physiol. 301, R581–R600 (2011).

  48. 48.

    Vink, R. G., Roumans, N. J., Arkenbosch, L. A., Mariman, E. C. & van Baak, M. A. The effect of rate of weight loss on long-term weight regain in adults with overweight and obesity. Obesity (Silver Spring) 24, 321–327 (2016).

  49. 49.

    Lenz, M. et al. Estimating real cell size distribution from cross-section microscopy imaging. Bioinformatics 32, i396–i404 (2016).

  50. 50.

    Isakson, P., Hammarstedt, A., Gustafson, B. & Smith, U. Impaired preadipocyte differentiation in human abdominal obesity: role of Wnt, tumor necrosis factor-alpha, and inflammation. Diabetes 58, 1550–1557 (2009).

  51. 51.

    Rossmeislova, L. et al. Weight loss improves the adipogenic capacity of human preadipocytes and modulates their secretory profile. Diabetes 62, 1990–1995 (2013).

  52. 52.

    Vink, R. G., Roumans, N. J., Mariman, E. C. & van Baak, M. A. Dietary weight loss-induced changes in RBP4, FFA, and ACE predict weight regain in people with overweight and obesity. Physiol. Rep. 5, e13450 (2017).

  53. 53.

    Rosenbaum, M., Kissileff, H. R., Mayer, L. E., Hirsch, J. & Leibel, R. L. Energy intake in weight-reduced humans. Brain Res. 1350, 95–102 (2010).

  54. 54.

    Bouwman, F., Renes, J. & Mariman, E. A combination of protein profiling and isotopomer analysis using matrix-assisted laser desorption/ionization-time of flight mass spectrometry reveals an active metabolism of the extracellular matrix of 3T3-L1 adipocytes. Proteomics 4, 3855–3863 (2004).

  55. 55.

    Wang, P. et al. Insulin modulates the secretion of proteins from mature 3T3-L1 adipocytes: a role for transcriptional regulation of processing. Diabetologia 49, 2453–2462 (2006).

  56. 56.

    Mariman, E. C. & Wang, P. Adipocyte extracellular matrix composition, dynamics and role in obesity. Cell. Mol. Life Sci. 67, 1277–1292 (2010).

  57. 57.

    Rossmeislova, L., Malisova, L., Kracmerova, J. & Stich, V. Adaptation of human adipose tissue to hypocaloric diet. Int. J. Obes. (Lond.) 37, 640–650 (2013).

  58. 58.

    Duncan, R. E., Ahmadian, M., Jaworski, K., Sarkadi-Nagy, E. & Sul, H. S. Regulation of lipolysis in adipocytes. Annu. Rev. Nutr. 27, 79–101 (2007).

  59. 59.

    Liu, Y. et al. Accumulation and changes in composition of collagens in subcutaneous adipose tissue after bariatric surgery. J. Clin. Endocrinol. Metab. 101, 293–304 (2016).

  60. 60.

    Schwarz, U. S. et al. Calculation of forces at focal adhesions from elastic substrate data: the effect of localized force and the need for regularization. Biophys. J. 83, 1380–1394 (2002).

  61. 61.

    Mutch, D. M. et al. A distinct adipose tissue gene expression response to caloric restriction predicts 6-mo weight maintenance in obese subjects. Am. J. Clin. Nutr. 94, 1399–1409 (2011).

  62. 62.

    Roumans, N. J. et al. Weight loss-induced stress in subcutaneous adipose tissue is related to weight regain. Br. J. Nutr. 115, 913–920 (2016).

  63. 63.

    Schneider, G. B., Hamano, H. & Cooper, L. F. In vivo evaluation of hsp27 as an inhibitor of actin polymerization: hsp27 limits actin stress fiber and focal adhesion formation after heat shock. J. Cell. Physiol. 177, 575–584 (1998).

  64. 64.

    Roumans, N. J. T. et al. Weight loss-induced cellular stress in subcutaneous adipose tissue and the risk for weight regain in overweight and obese adults. Int. J. Obes. (Lond.) 41, 894–901 (2017).

  65. 65.

    Roumans, N. J. et al. Variation in extracellular matrix genes is associated with weight regain after weight loss in a sex-specific manner. Genes Nutr. 10, 56 (2015).

  66. 66.

    Veit, G. et al. Collagen XXIII, novel ligand for integrin α2β1 in the epidermis. J. Biol. Chem. 286, 27804–27813 (2011).

  67. 67.

    Schluterman, M. K. et al. Loss of fibulin-5 binding to β1 integrins inhibits tumor growth by increasing the level of ROS. Dis. Model. Mech. 3, 333–342 (2010).

  68. 68.

    Roumans, N. J., Vink, R. G., Fazelzadeh, P., van Baak, M. A. & Mariman, E. C. A role for leukocyte integrins and extracellular matrix remodeling of adipose tissue in the risk of weight regain after weight loss. Am. J. Clin. Nutr. 105, 1054–1062 (2017).

  69. 69.

    Roumans, N. J. T., Wang, P., Vink, R. G., van Baak, M. A. & Mariman, E. C. M. Combined analysis of stress- and ECM-related genes in their effect on weight regain. Obesity (Silver Spring) 26, 492–498 (2018).

  70. 70.

    MacLean, P. S., Higgins, J. A., Giles, E. D., Sherk, V. D. & Jackman, M. R. The role for adipose tissue in weight regain after weight loss. Obes. Rev. 16 (Suppl. 1), 45–54 (2015).

  71. 71.

    Ge, F. et al. Facilitated long chain fatty acid uptake by adipocytes remains upregulated relative to BMI for more than a year after major bariatric surgical weight loss. Obesity (Silver Spring) 24, 113–122 (2016).

  72. 72.

    Grenier-Larouche, T. et al. Fatty acid metabolic remodeling during type 2 diabetes remission after bariatric surgery. Diabetes 66, 2743–2755 (2017).

  73. 73.

    Bouwman, F. G., Wang, P., van Baak, M., Saris, W. H. & Mariman, E. C. Increased β-oxidation with improved glucose uptake capacity in adipose tissue from obese after weight loss and maintenance. Obesity (Silver Spring) 22, 819–827 (2014).

  74. 74.

    Eastman, Q. Very low calorie diet makes adipocytes “scream”. J. Proteome Res. 8, 5408 (2009).

  75. 75.

    Vink, R. G. et al. Adipose tissue meal-derived fatty acid uptake before and after diet-induced weight loss in adults with overweight and obesity. Obesity (Silver Spring) 25, 1391–1399 (2017).

  76. 76.

    Santosa, S., Hensrud, D. D., Votruba, S. B. & Jensen, M. D. The influence of sex and obesity phenotype on meal fatty acid metabolism before and after weight loss. Am. J. Clin. Nutr. 88, 1134–1141 (2008).

  77. 77.

    Johansson, L. E. et al. Differential gene expression in adipose tissue from obese human subjects during weight loss and weight maintenance. Am. J. Clin. Nutr. 96, 196–207 (2012).

  78. 78.

    Van Pelt, D. W., Guth, L. M., Wang, A. Y. & Horowitz, J. F. Factors regulating subcutaneous adipose tissue storage, fibrosis, and inflammation may underlie low fatty acid mobilization in insulin-sensitive obese adults. Am. J. Physiol. Endocrinol. Metab. 313, E429–E439 (2017).

  79. 79.

    Schwartz, A. & Doucet, E. Relative changes in resting energy expenditure during weight loss: a systematic review. Obes. Rev. 11, 531–547 (2010).

  80. 80.

    Camps, S. G., Verhoef, S. P. & Westerterp, K. R. Weight loss, weight maintenance, and adaptive thermogenesis. Am. J. Clin. Nutr. 97, 990–994 (2013).

  81. 81.

    Hall, K. D. & Kahan, S. Maintenance of lost weight and long-term management of obesity. Med. Clin. North Am. 102, 183–197 (2018).

  82. 82.

    Camps, S. G. et al. Weight loss-induced changes in adipose tissue proteins associated with fatty acid and glucose metabolism correlate with adaptations in energy expenditure. Nutr. Metab. (Lond.) 12, 37 (2015).

  83. 83.

    Bouwman, F. G. et al. The physiologic effects of caloric restriction are reflected in the in vivo adipocyte-enriched proteome of overweight/obese subjects. J. Proteome Res. 8, 5532–5540 (2009).

  84. 84.

    Jokinen, R. et al. Adipose tissue mitochondrial capacity associates with long-term weight loss success. Int. J. Obes. (Lond.) 42, 817–825 (2017).

  85. 85.

    Marquez-Quinones, A. et al. Adipose tissue transcriptome reflects variations between subjects with continued weight loss and subjects regaining weight 6 mo after caloric restriction independent of energy intake. Am. J. Clin. Nutr. 92, 975–984 (2010).

  86. 86.

    Chen, Y., Yang, J., Nie, X., Song, Z. & Gu, Y. Effects of bariatric surgery on change of brown adipocyte tissue and energy metabolism in obese mice. Obes. Surg. 28, 820–830 (2018).

  87. 87.

    Vijgen, G. H. et al. Increase in brown adipose tissue activity after weight loss in morbidly obese subjects. J. Clin. Endocrinol. Metab. 97, E1229–E1233 (2012).

  88. 88.

    Dadson, P. et al. Brown adipose tissue lipid metabolism in morbid obesity: effect of bariatric surgery-induced weight loss. Diabetes Obes. Metab. 20, 1280–1288 (2018).

  89. 89.

    Barquissau, V. et al. Caloric restriction and diet-induced weight loss do not induce browning of human subcutaneous white adipose tissue in women and men with obesity. Cell Rep. 22, 1079–1089 (2018).

  90. 90.

    Neinast, M. D. et al. Activation of natriuretic peptides and the sympathetic nervous system following Roux-en-Y gastric bypass is associated with gonadal adipose tissues browning. Mol. Metab. 4, 427–436 (2015).

  91. 91.

    Fabbiano, S. et al. Caloric restriction leads to browning of white adipose tissue through type 2 immune signaling. Cell Metab. 24, 434–446 (2016).

  92. 92.

    Zhang, Y. et al. Positional cloning of the mouse obese gene and its human homologue. Nature 372, 425–432 (1994).

  93. 93.

    Siiteri, P. K. Adipose tissue as a source of hormones. Am. J. Clin. Nutr. 45, 277–282 (1987).

  94. 94.

    Wang, P., Mariman, E., Renes, J. & Keijer, J. The secretory function of adipocytes in the physiology of white adipose tissue. J. Cell. Physiol. 216, 3–13 (2008).

  95. 95.

    Choi, C. H. J. & Cohen, P. Adipose crosstalk with other cell types in health and disease. Exp. Cell Res. 360, 6–11 (2017).

  96. 96.

    Hocking, S. L., Wu, L. E., Guilhaus, M., Chisholm, D. J. & James, D. E. Intrinsic depot-specific differences in the secretome of adipose tissue, preadipocytes, and adipose tissue-derived microvascular endothelial cells. Diabetes 59, 3008–3016 (2010).

  97. 97.

    Strohacker, K., McCaffery, J. M., MacLean, P. S. & Wing, R. R. Adaptations of leptin, ghrelin or insulin during weight loss as predictors of weight regain: a review of current literature. Int. J. Obes. (Lond.) 38, 388–396 (2014).

  98. 98.

    Wang, P. et al. Blood profile of proteins and steroid hormones predicts weight change after weight loss with interactions of dietary protein level and glycemic index. PLOS ONE 6, e16773 (2011).

  99. 99.

    Wang, P. et al. Circulating ACE is a predictor of weight loss maintenance not only in overweight and obese women, but also in men. Int. J. Obes. (Lond.) 36, 1545–1551 (2012).

  100. 100.

    Rosenbaum, M. et al. Effects of weight change on plasma leptin concentrations and energy expenditure. J. Clin. Endocrinol. Metab. 82, 3647–3654 (1997).

  101. 101.

    Tamez, M. et al. Adipocyte size and leptin receptor expression in human subcutaneous adipose tissue after Roux-en-Y gastric bypass. Obes. Surg. 27, 3330–3332 (2017).

  102. 102.

    Bluher, M. & Mantzoros, C. S. From leptin to other adipokines in health and disease: facts and expectations at the beginning of the 21st century. Metabolism 64, 131–145 (2015).

  103. 103.

    Qi, Y. et al. Adiponectin acts in the brain to decrease body weight. Nat. Med. 10, 524–529 (2004).

  104. 104.

    Park, S., Kim, D. S., Kwon, D. Y. & Yang, H. J. Long-term central infusion of adiponectin improves energy and glucose homeostasis by decreasing fat storage and suppressing hepatic gluconeogenesis without changing food intake. J. Neuroendocrinol. 23, 687–698 (2011).

  105. 105.

    Smith, U. & Kahn, B. B. Adipose tissue regulates insulin sensitivity: role of adipogenesis, de novo lipogenesis and novel lipids. J. Intern. Med. 280, 465–475 (2016).

  106. 106.

    Noy, N. Vitamin A in regulation of insulin responsiveness: mini review. Proc. Nutr. Soc. 75, 212–215 (2016).

  107. 107.

    Brestoff, J. R. & Artis, D. Immune regulation of metabolic homeostasis in health and disease. Cell 161, 146–160 (2015).

  108. 108.

    Bernstein, K. E. et al. A modern understanding of the traditional and nontraditional biological functions of angiotensin-converting enzyme. Pharmacol. Rev. 65, 1–46 (2013).

  109. 109.

    McGregor, R. A. & Choi, M. S. microRNAs in the regulation of adipogenesis and obesity. Curr. Mol. Med. 11, 304–316 (2011).

  110. 110.

    Hilton, C., Neville, M. J. & Karpe, F. MicroRNAs in adipose tissue: their role in adipogenesis and obesity. Int. J. Obes. (Lond.) 37, 325–332 (2013).

  111. 111.

    Valenti, M. T., Dalle Carbonare, L. & Mottes, M. Role of microRNAs in progenitor cell commitment and osteogenic differentiation in health and disease (review). Int. J. Mol. Med. 41, 2441–2449 (2018).

  112. 112.

    Zaragosi, L. E. et al. Small RNA sequencing reveals miR-642a-3p as a novel adipocyte-specific microRNA and miR-30 as a key regulator of human adipogenesis. Genome Biol. 12, R64 (2011).

  113. 113.

    Chen, S. Z. et al. The miR-181d-regulated metalloproteinase Adamts1 enzymatically impairs adipogenesis via ECM remodeling. Cell Death Differ. 23, 1778–1791 (2016).

  114. 114.

    Liu, W. et al. LncRNA Gm15290 sponges miR-27b to promote PPARgamma-induced fat deposition and contribute to body weight gain in mice. Biochem. Biophys. Res. Commun. 493, 1168–1175 (2017).

  115. 115.

    Karbiener, M. et al. MicroRNA-26 family is required for human adipogenesis and drives characteristics of brown adipocytes. Stem Cells 32, 1578–1590 (2014).

  116. 116.

    Engin, A. B. MicroRNA and adipogenesis. Adv. Exp. Med. Biol. 960, 489–509 (2017).

  117. 117.

    Belarbi, Y. et al. MicroRNAs-361-5p and miR-574-5p associate with human adipose morphology and regulate EBF1 expression in white adipose tissue. Mol. Cell. Endocrinol. 472, 50–56 (2017).

  118. 118.

    Fatima, F. & Nawaz, M. Long distance metabolic regulation through adipose-derived circulating exosomal miRNAs: a trail for RNA-based therapies? Front. Physiol. 8, 545 (2017).

  119. 119.

    Arner, P. & Kulyte, A. MicroRNA regulatory networks in human adipose tissue and obesity. Nat. Rev. Endocrinol. 11, 276–288 (2015).

  120. 120.

    Arner, E. et al. Adipose tissue microRNAs as regulators of CCL2 production in human obesity. Diabetes 61, 1986–1993 (2012).

  121. 121.

    Zhu, L. et al. MiR-335, an adipogenesis-related microRNA, is involved in adipose tissue inflammation. Cell Biochem. Biophys. 68, 283–290 (2014).

  122. 122.

    Ortega, F. J. et al. Surgery-induced weight loss is associated with the downregulation of genes targeted by MicroRNAs in adipose tissue. J. Clin. Endocrinol. Metab. 100, E1467–E1476 (2015).

  123. 123.

    Schroeder, M., Drori, Y., Ben-Efraim, Y. J. & Chen, A. Hypothalamic miR-219 regulates individual metabolic differences in response to diet-induced weight cycling. Mol. Metab. 9, 176–186 (2018).

  124. 124.

    Bollepalli, S. et al. Subcutaneous adipose tissue gene expression and DNA methylation respond to both short- and long-term weight loss. Int. J. Obes. (Lond.) 42, 412–423 (2018).

  125. 125.

    Martinez, J. A., Milagro, F. I., Claycombe, K. J. & Schalinske, K. L. Epigenetics in adipose tissue, obesity, weight loss, and diabetes. Adv. Nutr. 5, 71–81 (2014).

  126. 126.

    Maurizi, G., Della Guardia, L., Maurizi, A. & Poloni, A. Adipocytes properties and crosstalk with immune system in obesity-related inflammation. J. Cell. Physiol. 233, 88–97 (2018).

  127. 127.

    Capel, F. et al. Contribution of energy restriction and macronutrient composition to changes in adipose tissue gene expression during dietary weight-loss programs in obese women. J. Clin. Endocrinol. Metab. 93, 4315–4322 (2008).

  128. 128.

    Vink, R. G. et al. Adipose tissue gene expression is differentially regulated with different rates of weight loss in overweight and obese humans. Int. J. Obes. (Lond.) 41, 309–316 (2017).

  129. 129.

    Capel, F. et al. Macrophages and adipocytes in human obesity: adipose tissue gene expression and insulin sensitivity during calorie restriction and weight stabilization. Diabetes 58, 1558–1567 (2009).

  130. 130.

    Schmitz, J. et al. Obesogenic memory can confer long-term increases in adipose tissue but not liver inflammation and insulin resistance after weight loss. Mol. Metab. 5, 328–339 (2016).

  131. 131.

    Snel, M. et al. Immediate and long-term effects of addition of exercise to a 16-week very low calorie diet on low-grade inflammation in obese, insulin-dependent type 2 diabetic patients. Food Chem. Toxicol. 49, 3104–3111 (2011).

  132. 132.

    Malisova, L. et al. Expression of inflammation-related genes in gluteal and abdominal subcutaneous adipose tissue during weight-reducing dietary intervention in obese women. Physiol. Res. 63, 73–82 (2014).

  133. 133.

    Zou, J. et al. CD4+T cells memorize obesity and promote weight regain. Cell. Mol. Immunol. 15, 630–639 (2017).

  134. 134.

    Kong, L. C. et al. Insulin resistance and inflammation predict kinetic body weight changes in response to dietary weight loss and maintenance in overweight and obese subjects by using a Bayesian network approach. Am. J. Clin. Nutr. 98, 1385–1394 (2013).

  135. 135.

    Wang, H. & Ye, J. Regulation of energy balance by inflammation: common theme in physiology and pathology. Rev. Endocr. Metab. Disord. 16, 47–54 (2015).

  136. 136.

    van den Berg, S. M., van Dam, A. D., Rensen, P. C., de Winther, M. P. & Lutgens, E. Immune modulation of brown(ing) adipose tissue in obesity. Endocr. Rev. 38, 46–68 (2017).

  137. 137.

    Armenise, C. et al. Transcriptome profiling from adipose tissue during a low-calorie diet reveals predictors of weight and glycemic outcomes in obese, nondiabetic subjects. Am. J. Clin. Nutr. 106, 736–746 (2017).

  138. 138.

    Sumithran, P., Purcell, K., Kuyruk, S., Proietto, J. & Prendergast, L. A. Combining biological and psychosocial baseline variables did not improve prediction of outcome of a very-low-energy diet in a clinic referral population. Clin. Obes. 8, 30–38 (2018).

  139. 139.

    Caires, R. et al. Omega-3 fatty acids modulate TRPV4 function through plasma membrane remodeling. Cell Rep. 21, 246–258 (2017).

  140. 140.

    Sidossis, L. & Kajimura, S. Brown and beige fat in humans: thermogenic adipocytes that control energy and glucose homeostasis. J. Clin. Invest. 125, 478–486 (2015).

  141. 141.

    Shen, W. & McIntosh, M. K. Nutrient regulation: conjugated linoleic acid’s inflammatory and browning properties in adipose tissue. Annu. Rev. Nutr. 36, 183–210 (2016).

  142. 142.

    Tsiloulis, T. et al. No evidence of white adipocyte browning after endurance exercise training in obese men. Int. J. Obes. (Lond.) 42, 721–727 (2017).

  143. 143.

    Norheim, F. et al. The effects of acute and chronic exercise on PGC-1alpha, irisin and browning of subcutaneous adipose tissue in humans. FEBS J. 281, 739–749 (2014).

  144. 144.

    Nakhuda, A. et al. Biomarkers of browning of white adipose tissue and their regulation during exercise- and diet-induced weight loss. Am. J. Clin. Nutr. 104, 557–565 (2016).

  145. 145.

    Pino, M. F., Parsons, S. A., Smith, S. R. & Sparks, L. M. Active individuals have high mitochondrial content and oxidative markers in their abdominal subcutaneous adipose tissue. Obesity (Silver Spring) 24, 2467–2470 (2016).

  146. 146.

    Steig, A. J. et al. Exercise reduces appetite and traffics excess nutrients away from energetically efficient pathways of lipid deposition during the early stages of weight regain. Am. J. Physiol. Regul. Integr. Comp. Physiol. 301, R656–R667 (2011).

  147. 147.

    Giles, E. D. et al. Exercise decreases lipogenic gene expression in adipose tissue and alters adipocyte cellularity during weight regain after weight loss. Frontiers Physiol. 7, 32 (2016).

  148. 148.

    Bartus, R. T. et al. β2-adrenoceptor agonists as novel, safe and potentially effective therapies for amyotrophic lateral sclerosis (ALS). Neurobiol. Dis. 85, 11–24 (2016).

  149. 149.

    Esser, N., Paquot, N. & Scheen, A. J. Anti-inflammatory agents to treat or prevent type 2 diabetes, metabolic syndrome and cardiovascular disease. Expert Opin. Investig. Drugs 24, 283–307 (2015).

  150. 150.

    Poulsen, M. M. et al. Resveratrol and inflammation: challenges in translating pre-clinical findings to improved patient outcomes. Biochim. Biophys. Acta 1852, 1124–1136 (2015).

  151. 151.

    Gleeson, M. et al. The anti-inflammatory effects of exercise: mechanisms and implications for the prevention and treatment of disease. Nat. Rev. Immunol. 11, 607–615 (2011).

  152. 152.

    Lancaster, G. I. & Febbraio, M. A. The immunomodulating role of exercise in metabolic disease. Trends Immunol. 35, 262–269 (2014).

  153. 153.

    Auerbach, P. et al. Differential effects of endurance training and weight loss on plasma adiponectin multimers and adipose tissue macrophages in younger, moderately overweight men. Am. J. Physiol. Regul. Integr. Comp. Physiol. 305, R490–R498 (2013).

  154. 154.

    van Baak, M. A. et al. Leisure-time activity is an important determinant of long-term weight maintenance after weight loss in the Sibutramine Trial on Obesity Reduction and Maintenance (STORM trial). Am. J. Clin. Nutr. 78, 209–214 (2003).

  155. 155.

    Kerns, J. C. et al. Increased physical activity associated with less weight regain six years after “the biggest loser” competition. Obesity (Silver Spring) 25, 1838–1843 (2017).

  156. 156.

    Ostendorf, D. M. et al. Objectively measured physical activity and sedentary behavior in successful weight loss maintainers. Obesity (Silver Spring) 26, 53–60 (2018).

  157. 157.

    de Luis, D. A. et al. Biochemical, anthropometric and lifestyle factors related with weight maintenance after weight loss secondary to a hypocaloric mediterranean diet. Ann. Nutr. Metab. 71, 217–223 (2017).

  158. 158.

    Kjaer, T. N. et al. Resveratrol reduces the levels of circulating androgen precursors but has no effect on, testosterone, dihydrotestosterone, PSA levels or prostate volume. A 4-month randomised trial in middle-aged men. Prostate 75, 1255–1263 (2015).

  159. 159.

    Calder, P. C. Long-chain fatty acids and inflammation. Proc. Nutr. Soc. 71, 284–289 (2012).

  160. 160.

    Shivappa, N. et al. Associations between dietary inflammatory index and inflammatory markers in the Asklepios Study. Br. J. Nutr. 113, 665–671 (2015).

  161. 161.

    Shivappa, N., Steck, S. E., Hurley, T. G., Hussey, J. R. & Hebert, J. R. Designing and developing a literature-derived, population-based dietary inflammatory index. Publ. Health Nutr. 17, 1689–1696 (2014).

  162. 162.

    Ramallal, R. et al. Inflammatory potential of diet, weight gain, and incidence of overweight/obesity: the SUN Cohort. Obesity 25, 997–1005 (2017).

  163. 163.

    Muhammad, H. F. L. et al. Dietary intake after weight loss and the risk of weight regain: macronutrient composition and inflammatory properties of the diet. Nutrients 9, 1205 (2017).

  164. 164.

    Ravussin, E. et al. Enhanced weight loss with pramlintide/metreleptin: an integrated neurohormonal approach to obesity pharmacotherapy. Obesity (Silver Spring) 17, 1736–1743 (2009).

  165. 165.

    Aller, E. E. et al. Weight loss maintenance in overweight subjects on ad libitum diets with high or low protein content and glycemic index: the DIOGENES trial 12-month results. Int. J. Obes. (Lond.) 38, 1511–1517 (2014).

  166. 166.

    Johansson, K., Neovius, M. & Hemmingsson, E. Effects of anti-obesity drugs, diet, and exercise on weight-loss maintenance after a very-low-calorie diet or low-calorie diet: a systematic review and meta-analysis of randomized controlled trials. Am. J. Clin. Nutr. 99, 14–23 (2014).

  167. 167.

    The STORM Study Group. Effect of sibutramine on weight maintenance after weight loss: a randomised trial. Lancet 356, 2119–2125 (2000).

  168. 168.

    Richelsen, B. et al. Effect of orlistat on weight regain and cardiovascular risk factors following a very-low-energy diet in abdominally obese patients: a 3-year randomized, placebo-controlled study. Diabetes Care 30, 27–32 (2007).

  169. 169.

    Wadden, T. A. et al. Weight maintenance and additional weight loss with liraglutide after low-calorie-diet-induced weight loss: the SCALE Maintenance randomized study. Int. J. Obes. (Lond.) 37, 1443–1451 (2013).

  170. 170.

    Vazquez, C. et al. Meal replacement with a low-calorie diet formula in weight loss maintenance after weight loss induction with diet alone. Eur. J. Clin. Nutr. 63, 1226–1232 (2009).

  171. 171.

    Westerterp-Plantenga, M. S., Lejeune, M. P. & Kovacs, E. M. Body weight loss and weight maintenance in relation to habitual caffeine intake and green tea supplementation. Obes. Res. 13, 1195–1204 (2005).

  172. 172.

    Dutton, G. R. et al. Comparison of an alternative schedule of extended care contacts to a self-directed control: a randomized trial of weight loss maintenance. Int. J. Behav. Nutr. Phys. Act. 14, 107 (2017).

  173. 173.

    Voils, C. I. et al. Maintenance of weight loss after initiation of nutrition training: a randomized trial. Ann. Intern. Med. 166, 463–471 (2017).

  174. 174.

    Crain, A. L., Sherwood, N. E., Martinson, B. C. & Jeffery, R. W. Mediators of weight loss maintenance in the Keep It Off trial. Ann. Behav. Med. 52, 9–18 (2017).

  175. 175.

    Ryan, A. S., Serra, M. C. & Goldberg, A. P. Metabolic benefits of prior weight loss with and without exercise on subsequent 6-month weight regain. Obesity (Silver Spring) 26, 37–44 (2018).

  176. 176.

    Wong, M. H. et al. Caloric restriction induces changes in insulin and body weight measurements that are inversely associated with subsequent weight regain. PLOS ONE 7, e42858 (2012).

  177. 177.

    Hanvold, S. E. et al. Plasma amino acids, adiposity, and weight change after gastric bypass surgery: are amino acids associated with weight regain? Eur. J. Nutr. 57, 2629–2637 (2017).

  178. 178.

    Sawamoto, R. et al. Predictors of successful long-term weight loss maintenance: a two-year follow-up. Biopsychosoc. Med. 11, 14 (2017).

  179. 179.

    Calugi, S., Marchesini, G., El Ghoch, M., Gavasso, I. & Dalle Grave, R. The influence of weight-loss expectations on weight loss and of weight-loss satisfaction on weight maintenance in severe obesity. J. Acad. Nutr. Diet 117, 32–38 (2017).

  180. 180.

    Greenberg, I., Stampfer, M. J., Schwarzfuchs, D., Shai, I. & Group, D. Adherence and success in long-term weight loss diets: the dietary intervention randomized controlled trial (DIRECT). J. Am. Coll. Nutr. 28, 159–168 (2009).

  181. 181.

    Vogels, N. & Westerterp-Plantenga, M. S. Categorical strategies based on subject characteristics of dietary restraint and physical activity, for weight maintenance. Int. J. Obes. (Lond.) 29, 849–857 (2005).

  182. 182.

    Fabricatore, A. N. et al. Predictors of attrition and weight loss success: results from a randomized controlled trial. Behav. Res. Ther. 47, 685–691 (2009).

  183. 183.

    Lillis, J. et al. Weight loss intervention for individuals with high internal disinhibition: design of the Acceptance Based Behavioral Intervention (ABBI) randomized controlled trial. BMC Psychol. 3, 17 (2015).

  184. 184.

    Butryn, M. L., Thomas, J. G. & Lowe, M. R. Reductions in internal disinhibition during weight loss predict better weight loss maintenance. Obesity (Silver Spring) 17, 1101–1103 (2009).

  185. 185.

    Abu Dayyeh, B. K., Jirapinyo, P. & Thompson, C. C. Plasma ghrelin levels and weight regain after Roux-en-Y gastric bypass surgery. Obes. Surg. 27, 1031–1036 (2017).

  186. 186.

    Brock, D. W. et al. Perception of exercise difficulty predicts weight regain in formerly overweight women. Obesity (Silver Spring) 18, 982–986 (2010).

  187. 187.

    Price, D. W. et al. Depression as a predictor of weight regain among successful weight losers in the diabetes prevention program. Diabetes Care 36, 216–221 (2013).

  188. 188.

    Larsen, L. H. et al. Analyses of single nucleotide polymorphisms in selected nutrient-sensitive genes in weight-regain prevention: the DIOGENES study. Am. J. Clin. Nutr. 95, 1254–1260 (2012).

  189. 189.

    Nicklas, B. J. et al. Genetic variation in the peroxisome proliferator-activated receptor-gamma2 gene (Pro12Ala) affects metabolic responses to weight loss and subsequent weight regain. Diabetes 50, 2172–2176 (2001).

  190. 190.

    McCaffery, J. M. et al. FTO predicts weight regain in the Look AHEAD clinical trial. Int. J. Obes. (Lond.) 37, 1545–1552 (2013).

  191. 191.

    Delahanty, L. M. et al. Genetic predictors of weight loss and weight regain after intensive lifestyle modification, metformin treatment, or standard care in the Diabetes Prevention Program. Diabetes Care 35, 363–366 (2012).

  192. 192.

    Masuo, K. et al. Rebound weight gain as associated with high plasma norepinephrine levels that are mediated through polymorphisms in the β2-adrenoceptor. Am. J. Hypertens. 18, 1508–1516 (2005).

  193. 193.

    Crujeiras, A. B. et al. Association of weight regain with specific methylation levels in the NPY and POMC promoters in leukocytes of obese men: a translational study. Regul. Pept. 186, 1–6 (2013).

Download references

Reviewer information

Nature Reviews Endocrinology thanks José Fernandez-Real and other anonymous reviewers for their contribution to the peer review of this work.

Author information


  1. NUTRIM School for Nutrition and Translational Research in Metabolism, Department of Human Biology, Maastricht University, Maastricht, Netherlands

    • Marleen A. van Baak
    •  & Edwin C. M. Mariman


  1. Search for Marleen A. van Baak in:

  2. Search for Edwin C. M. Mariman in:


The authors contributed equally to all aspects of the article.

Competing interests

The authors declare no competing interests.

Corresponding author

Correspondence to Marleen A. van Baak.


Energy balance

The balance between energy intake and energy expenditure.

Fasting fatty acid rate of appearance

The rate of release of fattyacids into the plasma in thefasting state, which can bedetermined by measuring thedilution of an infused fatty acidtracer in plasma.

Weight cycling

The repeated loss and regain of body weight.

Bayesian analysis

A statistical paradigm that answers research questions about unknown parameters using probability statements.

About this article

Publication history