Body fat distribution is an important metabolic and cardiovascular risk factor, because the proportion of abdominal to gluteofemoral body fat correlates with obesity-associated diseases and mortality. Here, we review the evidence and possible mechanisms that support a specific protective role of gluteofemoral body fat. Population studies show that an increased gluteofemoral fat mass is independently associated with a protective lipid and glucose profile, as well as a decrease in cardiovascular and metabolic risk. Studies of adipose tissue physiology in vitro and in vivo confirm distinct properties of the gluteofemoral fat depot with regards to lipolysis and fatty acid uptake: in day-to-day metabolism it appears to be more passive than the abdominal depot and it exerts its protective properties by long-term fatty acid storage. Further, a beneficial adipokine profile is associated with gluteofemoral fat. Leptin and adiponectin levels are positively associated with gluteofemoral fat while the level of inflammatory cytokines is negatively associated. Finally, loss of gluteofemoral fat, as observed in Cushing's syndrome and lipodystrophy is associated with an increased metabolic and cardiovascular risk. This underlines gluteofemoral fat's role as a determinant of health by the long-term entrapment of excess fatty acids, thus protecting from the adverse effects associated with ectopic fat deposition.
Obesity is defined as a body mass index (BMI) >30 kg m–2 and an increased BMI has been associated with a high risk of cardiovascular disease and diabetes.1 It has long been known that body fat distribution is also an important risk factor: the proportion of abdominal to gluteofemoral body fat, as measured, for example, by the waist-to-hip ratio, correlates with obesity-associated diseases and mortality and is a stronger cardiovascular risk marker than BMI.2, 3, 4 Here, we review the evidence for gluteofemoral fat being protective and the possible mechanisms that may be involved.
Body fat distribution and risk
Several population studies support the fact that obesity-associated health risks depend on the accumulation of abdominal fat. Abdominal obesity is associated with increased blood pressure and plasma triglyceride (TG) levels.2, 3 It is also an independent predictor for the development of type 2 diabetes.5
The adverse effect of abdominal obesity arises from the combination of subcutaneous adipose tissue dysfunction and the accumulation of visceral fat that becomes apparent when individuals with increased waist-to-hip ratio are studied with imaging techniques.6, 7, 8 The particular importance of visceral fat accumulation versus subcutaneous fat depot dysfunction has recently been challenged9, 10 as there is mounting evidence to support that visceral adipocytes are phenotypically different from subcutaneous adipocytes.11
Nevertheless, it is the protective role of lower body, that is, gluteofemoral fat that is striking. The protective properties of the lower-body fat depot have been confirmed in many studies conducted in subjects with a wide range of age, BMI and co-morbidities (Table 1). Gluteofemoral fat, as measured by thigh circumference, hip circumference or leg adipose tissue mass, is independently associated with lower total- and low-density lipoprotein-cholesterol, and total- and very-low-density lipoprotein–TG levels, and higher high-density lipoprotein-cholesterol levels.12, 13, 14, 15, 16, 17 Atherosclerotic protection is not only conveyed by an improved lipid profile, but also through direct effects on vascular health: increased gluteofemoral fat mass is associated with lower aortic calcification and arterial stiffness,18, 19, 20 as well as with a decreased progression of aortic calcification.21 Leg adipose tissue mass change after a 14-week intervention study with diet and exercise is inversely associated with diastolic blood pressure, low-density lipoprotein-cholesterol levels and the number of coronary heart disease-risk factors (that is, blood pressure, fasting plasma glucose levels and levels of individual lipid fractions).22 Further, lower-body fat is inversely associated with fasting insulin levels and insulin levels after an oral glucose load, and positively associated with insulin sensitivity.12, 17, 23, 24, 25 In healthy overweight and obese women, hip circumference and thigh adipose tissue mass are associated with a lower Hba1c and increased leptin levels.26
The protective effect of gluteofemoral fat has also been confirmed in large population studies. In the AusDiab study, a larger hip circumference was associated with a lower prevalence of undiagnosed diabetes mellitus and dyslipidaemia.27 The INTERHEART study, comprising 27 000 participants, established an independent association between larger hip circumference and lower risk for myocardial infarction.4 In the prospective European Prospective Investigation into Cancer and Nutrition–Norfolk study, larger hip circumference was associated with a lower hazard ratio for coronary heart disease.28 Further, hip circumference was positively associated with plasma levels of ascorbic acid, an anti-oxidative factor thought to contribute to endothelial protection.29
Recent physiological and molecular studies elucidating the possible mechanisms that may contribute to the gluteofemoral depot's protective role will be the focus of this review. It has been suggested that subcutaneous adipose tissue acts as a buffer for the daily influx of dietary lipids, protecting other tissues from a lipid overflow with associated lipotoxicity.30 The protective properties of gluteofemoral fat could derive from a differential local handling of fatty acid uptake and release. Indeed, femoral fat accumulation that is typical of the female fat distribution pattern is associated with an elevated adipose tissue lipoprotein lipase activity.31, 32 Given the metabolic and cardiovascular protection that is conveyed by the gynoid fat distribution pattern, this supports the view of gluteofemoral fat being a protective ‘metabolic sink’.33
Another possibility is that there is differential secretion of adipose tissue-related proteins, that is, adipokines. As it will be discussed below, gluteofemoral adipose tissue could contribute to a protective adipokine profile by secreting more ‘beneficial’ adipokines and less pro-inflammatory molecules compared with abdominal fat.
Finally, conditions that result in the loss of gluteofemoral fat, as in chronic glucocorticoid excess in Cushing's syndrome (CS) or partial lipodystrophy, lead to metabolic abnormalities that underline the protective properties of this distinctive adipose tissue depot.
Fatty acid storage and release
Adipose tissue metabolism is a complex and highly regulated process that depends on several factors.34 One of the main functions of subcutaneous adipose tissue is the short-term and long-term storage of energy as TG to re-supply the organism with energy in form of non-esterified fatty acids, released during periods of exercise, fasting or starvation. Important determinants of local adipose tissue fatty acid trafficking are the rate of lipolysis, the rate of TG storage, primarily taking place after meals, and the blood flow in the tissue beds.35 Variations in these factors may be responsible for the beneficial effects of gluteofemoral fat depots.
Insulin is a major suppressor of lipolysis in adipose tissue. Body fat distribution has an effect on the inhibition of lipolysis: isolated adipocytes from upper-body obese women respond less to insulin when compared with those of lower-body obese.36 Catecholamines increase lipolysis in isolated adipocytes. However, there are marked differences between depots with regards to this response. Abdominal adipocytes showed a four- to fivefold increase in lipolysis during noradrenaline stimulation when compared with gluteal adipocytes.37 This is mediated through a higher expression of beta-adrenoceptors in the abdominal depot.38
The uptake of meal fatty acids from TG-rich chylomicrons depends on the expression of lipoprotein lipase (LPL) by adipocytes. In isolated adipocytes and fat segments, LPL mRNA expression is higher in abdominal than in gluteal cells.39 However, the situation is probably more complex, not least because of the complex post-transcriptional regulation of LPL.40 This is observed when analyzing the sexual dimorphism of fat distribution: activity of LPL is higher in abdominal adipocytes from men and in gluteal adipocytes from women when compared with the other depot respectively.31, 39 In men, testosterone suppresses LPL activity in the thigh, thus contributing to abdominal fat accumulation.41 It is therefore likely that gender-specific fat depot formation is partly regulated by LPL.
As a result of the complexity of human adipose tissue metabolism, the study of fat depots in vivo is the only way that allows for the integration of factors not present in vitro. Upper-body fat is the major contributor of systemic non-esterified fatty acids, hence showing a higher lipolysis rate when compared with the lower-body fat depot.42, 43 The rate of action of the enzyme hormone-sensitive lipase, a key enzyme in lipolysis, is lower in the gluteal than the abdominal depot.44 Starvation for 72 h results in increased lipolysis in the abdominal depot but not in gluteofemoral fat.45 During adrenaline stimulation, abdominal lipolysis appears to be higher than femoral lipolysis, when measured as glycerol release by microdialysis.46 Palmitate release, a direct marker of lipolysis, is lower in leg fat compared with the abdominal depot during systemic beta-adrenoceptor stimulation.47
Regional differences can also be found when studying fatty acid uptake in vivo. A consistent finding, much in line with lower lipolysis rates in gluteofemoral body fat, is that abdominal fat takes up meal-derived fatty acids more avidly than lower-body fat on a day-to-day short-term basis.48, 49 This supports the hypothesis of the subcutaneous fat being a metabolic ‘sink’, with the majority of daily fatty acid buffering being handled by abdominal fat. The exact fatty acid trafficking might, however, be more complex, given the presence of putative distinct very-low-density lipoprotein- and non-esterified fatty acid-turnover pathways, for instance. These, as well as the regulation of long-term fatty acid deposition in lower-body fat depots, are little understood and require further study.
Adipose tissue blood flow (ATBF) is an important determinant of fat metabolism that can only be studied in vivo. Variations in blood flow allow for the regulation of non-esterified fatty acid release into the systemic circulation and of substrate availability for LPL after a meal. Conversely, ATBF increases in periods of starvation and postprandially.50, 51 Interestingly, the postprandial rise in ATBF is not a direct insulin effect but is mainly mediated through catecholamines and nitric oxide.52, 53
ATBF changes of the abdominal fat depot have been well studied in different settings; however, little is known regarding ATBF regulation in the lower-body fat depot. Gluteal fat was shown to have lower basal ATBF,44 and in lean women, femoral ATBF showed an attenuated increase during systemic adrenaline stimulation compared with abdominal fat.46 Postprandially, ATBF increases in both the abdominal and femoral depots in women, but only in the abdominal depot in men.49 Further studies will be needed to elucidate the exact mechanisms behind differential ATBF regulation of the two depots.
In summary, upper- and lower-body fat show distinct properties with regards to lipolysis and fatty acid uptake. When looking at the short-term handling of fatty acids, the abdominal depot actively participates in daily fatty acid metabolism and seems to act as a buffer for the daily influx of dietary fatty acids.30 In contrast, the gluteofemoral depot appears to be more passive and exerts its protective properties in long-term fatty acid storage (Figure 1). This is evident in the accumulation of gluteofemoral fat associated with the typical female fat distribution.31 Female lower-body fat is only removed during periods of excessive energy demand, for example, lactation.54 In contrast, healthy women are more protected from cardiovascular events than men, until their body fat distribution changes with menopause (interestingly, toward the android distribution pattern).55, 56 Extrapolating this, gluteofemoral fat may protect our bodies, irrespective of gender, by trapping excess fatty acids and preventing chronic exposure to elevated lipid levels. The exact regulatory mechanisms of fatty acid release and storage and their effect on short- and long-term fatty acid metabolism remain to be analyzed.
Adipokine formation in regional adipose tissue
Adipose tissue is known to secrete several hormones termed adipokines. The term ‘adipokines’ accounts for the common source of these hormones57 but does not take into consideration that they are in fact secreted by different cell types within the tissue. Thus, while adipokines such as adiponectin and leptin are produced by adipocytes, interleukins are secreted mainly by resident macrophages. Wherever they come from, adipokines are part of a signalling cascade that is only partly understood, allowing for the communication between adipose tissue and other organs, primarily brain and liver. The existence of such a communication network is not surprising given the tissue's central role in fat storage and release and the need for coordination and control of appetite, hunger and satiety, which are subject to central regulation. The question that arises is whether the protective properties of lower-body fat are associated with a beneficial profile in its adipokine metabolism.
Differences in expression of adipokine genes and in secretion in vitro are well established when comparing subcutaneous and visceral fat.11, 58 However, little is known regarding differences between upper- and lower-body fat.
One of the first adipokines discovered was leptin, a 16-kDa protein that is mainly secreted by adipocytes as a product of the LEP gene.59, 60 Leptin is thought to be pivotal in energy metabolism by regulating appetite and energy intake as a function of fat mass.61 Leptin levels are associated with the subcutaneous rather than the visceral fat depot and leptin mRNA is more abundant in subcutaneous adipocytes.6, 58, 62 Leptin secretion in vitro is higher in subcutaneous adipocytes compared with visceral adipocytes and this secretion rate correlates well with serum leptin levels.63 There are no studies that have compared regional leptin mRNA expression in subcutaneous adipose tissue; however, there are indirect suggestions that basal expression does not differ between abdominal and gluteofemoral fat.64, 65 Serum leptin increases in parallel with BMI and, interestingly, is higher in women.66, 67, 68, 69, 70 The sexual dimorphism in circulating leptin levels is apparent already in childhood71, 72 and has been attributed to differences in subcutaneous fat mass and adipocyte cell size.73, 74 But could gender differences in body fat distribution be indicative of a differential leptin production between depots? Leptin levels correlate with leg fat mass and hip circumference75, 76, 77 and, in obesity, are negatively correlated to waist-to-hip ratio.66, 78, 79 In men, interstitial leptin concentrations, as measured by microdialysis, are higher in subcutaneous femoral than abdominal fat.80 This is further supported by the finding that the leg is a net leptin producer, when measured directly.81 When comparing android to gynoid obesity, differences can be found in circadian serum leptin levels, with higher amplitude of secretion in lower-body obesity.82, 83 However, despite these associations a direct differential involvement of either depot remains to be proven. Given leptin's role as a feedback signal with regards to body fat mass and a modulator of metabolic processes,84 it remains speculative whether there is any special association with the lower-body fat depot. This is an area that needs further investigation.
A beneficial association has also been shown for another adipokine—adiponectin. Adiponectin is exclusively secreted by adipocytes and is negatively correlated to body fat mass, that is, its serum levels are decreased in obesity.85, 86, 87 Interestingly, higher adiponectin levels are related to better glycaemic control and insulin sensitivity, a more favorable lipid profile and reduced inflammation in healthy individuals88, 89 and diabetic patients.90 Thus, it is believed that adiponectin provides a link between obesity and the development of insulin resistance and cardiovascular disease.91 Accordingly, regarding the influence of gluteofemoral body fat, adiponectin levels are positively associated with leg fat mass.92 Moreover, higher gluteofemoral fat mass results in higher adiponectin levels and increased insulin sensitivity.16, 93 There is a negative association between adiponectin and waist-to-hip ratio, central fat mass accumulation and visceral fat mass.6, 88, 94, 95, 96, 97 Further, in vitro studies have shown that adiponectin protein content and mRNA expression are lower in visceral adipocytes98 and that with increasing visceral fat mass adiponectin secretion from those cells is decreased, whereas secretion rates from subcutaneous adipocytes remain unaffected.99 There is also a differential expression pattern of adiponectin receptor 1 with high levels of expression in subcutaneous fat. Although there are no data regarding differences in adiponectin receptor expression levels between the upper- and lower-body fat depot, it is of interest that subcutaneous receptor expression is reduced in obesity and can be restored by weight loss.100 In summary, a favorable adiponectin profile could be facilitated by an increased lower-body fat depot.
Alongside the ‘classic’ adipokines leptin and adiponectin, several so-called inflammatory cytokines have been identified to be secreted by adipose tissue. This is recognized as one of the links between obesity and the development of cardiovascular disease, insulin resistance and diabetes, because inflammatory processes are part of the pathophysiology of these diseases. The most widely analyzed cytokines are tumor necrosis factor-alpha (TNF-α) and interleukin 6 (IL-6).
TNF-α is secreted by many different cell types, including adipocytes, as a 17-kDa protein, which is derived by cleavage from an initially synthesized 26-kDa transmembrane monomeric protein.101, 102 Although adipocytes are capable of secreting TNF-α, it is now recognized that the main proportion of TNF-α secreted by adipose tissue originates from adipose tissue-resident macrophages and other cells of the stromavascular fraction.103 TNF-α is a strong inducer of lipolysis104 and data from rodents and humans support a role in the development of insulin resistance.105, 106, 107 The latter, however, has been challenged by others108 and although TNF-α levels are increased in obesity,109, 110 there is little direct release from abdominal subcutaneous adipose tissue in vivo.111 Interestingly, in a similar set of studies it was found that abdominal subcutaneous adipose tissue releases the soluble TNF receptor type 1 and that its levels correlate with BMI.112 Given that there is no association between body fat distribution and serum levels of TNF-α,113, 114 one could hypothesize that it is the secreted levels of soluble TNF receptor type 1, which determine the adverse properties of TNF-α in abdominal obesity.115 However, because there are no data regarding TNF-α secretion by lower-body fat depots, this remains to be further analyzed.
IL-6 is an inflammatory cytokine that, similarly to TNF-α, is secreted by a variety of cells and tissues,116 including adipose tissue. There, it is secreted both from adipocytes and stromavascular fraction cells and secretion levels correlate positively with obesity.117, 118 Systemic plasma levels of IL-6 correlate with visceral fat mass and there is a weak negative association with increasing thigh fat area.119 There are no studies comparing regional IL-6 production between upper- and lower-body fat, but direct regional in vivo measurement of IL-6 and IL-6 receptor concentrations showed abdominal tissue to release IL-6 but not its receptor.111, 112 Although insulin-resistance and cardiovascular disease are viewed as partly low-grade chronic inflammatory processes, the IL-6 role in the development of insulin-resistance remains controversial.120 There is a current lack of data on regional IL-6 release and this is an area of future research.
Currently, there are several other adipokines known and some of them show interesting correlations between systemic levels and body fat distribution parameters. For instance, plasma levels of retinol-binding protein 4, a protein released from adipocytes and associated with insulin resistance,121 correlate with trunk fat mass.122, 123 Further, serum vaspin, an insulin-sensitizing adipokine, shows an interesting sexual dimorphism with levels higher in women.124 For these, and for other emerging adipokines, little is known regarding the specific role of regional adipose tissue in their production, secretion and function in human metabolism.
Gluteofemoral fat loss and disease
So far, the beneficial properties of the gluteofemoral fat depot have been highlighted. A further perspective underlining its protective features is to elucidate what happens when substantial amounts of gluteofemoral fat mass are lost. Indeed, lower-body fat loss is associated with metabolic abnormalities.
Cortisol, the main glucocorticoid in humans, which is secreted by the adrenal cortex, exerts important catabolic functions in fuel metabolism. The effect of cortisol on body fat and metabolic health is observed in states of glucocorticoid excess such as Cushing's syndrome, in which a marked fat re-distribution is pathognomonic. Fat redistribution in CS is a complex area, as glucocorticoid excess not only reduces the gluteofemoral fat depot, but also changes adipocyte function in the abdominal depot (reviewed in125). Glucocorticoid excess in CS patients results in an increase of total fat mass, mostly confined to the abdominal fat depot.126 This abdominal fat accumulation might be due to an increase in abdominal LPL activity and a decreased lipolytic capacity of the enlarged abdominal adipocytes.127 Furthermore, corticosteroid exposure results in a decrease of abdominal adipose tissue lipolysis in vivo.128 Abdominal fat accumulation in CS has been attributed to the redistribution of peripheral fat to the visceral fat depot.129, 130 However, it is interesting to note that visceral fat accumulation in CS does not correlate with lipid and glucose profiles.131 It is well established that glucocorticoid excess, as observed in CS, is associated with an adverse glucose and lipid profile that increases metabolic and cardiovascular risk.132, 133 So far, the direct association of gluteofemoral fat loss and metabolic health in CS has not been studied. However, the relative mass loss because of fat redistribution provides a plausible mechanism for the adverse effect of glucocorticoid excess on metabolic and cardiovascular risk.
The deleterious effect of gluteofemoral fat loss becomes most evident in lipodystrophic syndromes. Lipodystrophy is characterized by partial or total absence of adipose tissue and can be either inherited or acquired.134 The loss of adipose tissue leads to ectopic fat accumulation, in line with the view of subcutaneous fat being a ‘metabolic sink’ for excess energy.33, 135 Ectopic fat accumulation occurs in the liver, pancreas and muscle. Further, depending on the type of lipodystrophy, there is fat accumulation in specific non-atrophic fat depots, for instance the neck.
Patients suffering from autosomal dominant familial partial lipodystrophy Dunnigan-type lose primarily the subcutaneous adipose tissue of their arms and legs.135 These patients show severe insulin resistance, lipid profile abnormalities, hypertension and develop diabetes.136, 137, 138 Several mutations in the LMNA gene as well as in the PPARG gene were identified as the molecular mechanism responsible for the development of familial partial lipodystrophy.134 The PPARG-mutations provide a possible link between adipose tissue dysfunction and the metabolic profile associated with this type of lipodystrophy, given that adipose tissue appears to be unresponsive to physiological regulatory mechanisms in these patients.138
In contrast, patients with acquired partial lipodystrophy (Barraquer–Simons’ syndrome), who progressively lose their adipose tissue from the face and upper body but deposit large amounts of fat in the gluteofemoral region, show mild insulin resistance and have a lower prevalence of diabetes compared with other lipodystrophic syndromes.134, 139
Body fat distribution is a major determinant of metabolic health and gluteofemoral adipose tissue exerts specific functional properties that are associated with an improved metabolic and cardiovascular risk profile. The protective properties of gluteofemoral fat have been confirmed in large population studies. Evidence arising from in vitro and in vivo studies suggests that this is due to a differential regulation of lower-body fatty acid release and uptake at the level of the adipocyte that results in the long-term entrapment of fatty acids in this depot and protection from ectopic fat accumulation. This becomes even more evident in states of gluteofemoral fat loss as in glucocorticoid excess and lipodystrophic syndromes that are associated with increased metabolic and cardiovascular risk. In addition, gluteofemoral adipose tissue could convey protection through a beneficial adipokine profile. This is suggested by the favorable association of leptin and adiponectin levels with gluteofemoral fat mass, while it remains to be proven for other emerging adipokines. The recent discovery of novel gene loci involved in the regulation of body fat distribution, beyond the male–female dichotomy, provides an appealing area for further investigation.140 Future research will also show if the protective properties of gluteofemoral body fat can be used to attenuate metabolic and cardiovascular risk, as already observed in the redistribution of body fat associated with the clinical use of thiazolidinediones.141, 142
We thank the Wellcome Trust for support of our work.