White adipose tissue (WAT), long regarded as a “Cinderella organ,” has truly emerged into the limelight, and much of the stimulus for this relates to the current concern with obesity. This disease now affects over one in five adults in the United Kingdom, for example, with even more in the United States, and is associated with a reduced life expectancy and an increased incidence of several major diseases, particularly type II diabetes, coronary heart disease, and cancer. An important recent development is the emergence of the concept that obesity, like diabetes, is characterized by chronic low-grade inflammation (1, 2). WAT is itself recognized as an important site of the production of inflammation-related proteins, the production of which is (generally) increased in the obese (3, 4). Considerable interest was aroused just over a year ago by two reports that demonstrated that, in obesity, adipose tissue is infiltrated by macrophages (5, 6). One important factor produced by adipocytes underlying this infiltration is monocyte chemoattractant protein-1 (MCP-1) (7, 8); another may be macrophage migration inhibitory factor (MIF), and a key paper by Skurk et al. (9), in this issue of Endocrinology, demonstrates that MIF is secreted from human adipocytes and that the rate of secretion (in culture) is positively correlated with the body mass index of the subjects.
MIF, which was originally identified in activated T lymphocytes as a cytokine that inhibited the migration of macrophages from capillaries, is part of the accelerating list of protein factors and signals secreted from white adipocytes—the adipokines (3, 4, 10). The recognition that protein signals are secreted from adipocytes began in effect with adipsin in the late 1980s (11) and was followed by the proinflammatory cytokine TFNα a few years later (12). The pivotal event in our evolving perspective on WAT as a secretory organ was the discovery in 1994 of leptin (13); this resulted in the characterization of the tissue as a critical endocrine system. Currently, more than 50 different adipokines are recognized, and these are highly heterogeneous both in terms of protein structure and of function (3, 4, 10). The adipokines are implicated in a wide range of physiological processes, including appetite and energy balance, glucose homeostasis, lipid metabolism, blood pressure regulation, hemostasis, and angiogenesis (3, 4, 10). An increasing number of adipokines are directly linked to inflammation and the inflammatory response (Fig. 1), and these encompass classical cytokines (e.g. TNFα, IL-1β, IL-6, IL-10) and acute-phase proteins (e.g. haptoglobin, plasminogen activator inhibitor-1, serum amyloid A), as well as other inflammation-related signals such as MCP-1, nerve growth factor, and adiponectin (3, 4). Adiponectin, a major adipocyte-derived hormone with a role in insulin sensitivity, has a distinct antiinflammatory action (14).
Inflammation-related adipokines and signals for macrophage infiltration into WAT. Only factors for which secretion from adipocytes has been clearly demonstrated are shown. NGF, Nerve growth factor; PAI-1, plasminogen activator inhibitor-1; VEGF, vascular endothelial growth factor.
The expression and secretion of MIF by adipose tissue was first described in mice and was found to be strongly up-regulated in 3T3-L1 adipocytes by TNFα (15). Whether TNFα also stimulates the synthesis and release of MIF in human adipocytes is not yet known, but, interestingly, Skurk et al. (9) indicate that neither lipopolysaccharide, IFN-γ, nor IL-4 have any effect on MIF secretion in human fat cells differentiated in culture. Stimulation of MIF in human adipocytes by TNFα would be expected not only because of the data from the murine cell line (15), but also because this cytokine has pleiotropic effects on the production of inflammation-related adipokines in humans, including the stimulation of IL-6, MCP-1, and nerve growth factor expression (16). The capacity of human adipocytes to secrete MIF appears to be substantial, with similar levels of secretion from cells from the sc and omental depots (within the same individual). Some depot differences are evident, however; much lower rates of release are apparent with mammary fat cells (9).
Although, with human adipocytes, high concentrations of insulin do not affect the expression and release of MIF (9), in 3T3-L1 cells, a role for insulin in regulating the production of the factor is suggested (17). Recent studies on human subjects have indicated that MIF mRNA levels are increased in circulating mononuclear cells in obesity and the plasma levels of the protein are also increased (18, 19). Indeed, this up-regulation of MIF is related to body mass index, with the plasma levels being suppressed by treatment with metformin (19). Thus, MIF is one of the growing number of inflammation-related proteins whose circulating levels are increased in obesity. These proteins include IL-6, C-reactive protein, TNFα and its soluble receptors, IL-18, plasminogen activator inhibitor-1, and haptoglobin, and are the basis for the view that the obese are characterized by chronic low-grade inflammation (20–25). Importantly, the circulating level of adiponectin with its antiinflammatory effect falls in obesity (26). With the exception of IL-18, there is strong evidence that the expanded adipose tissue mass of the obese contributes either directly or indirectly to these increased circulating levels.
There is growing evidence of a causal link between what happens in adipose tissue in obesity and the development of type 2 diabetes and the metabolic syndrome (27, 28). Indeed, because expansion of the size and number of adipocytes is the key characteristic of obesity, it is unsurprising that there is a link between such events and the pathologies associated with the disorder. Although adipose tissue is clearly a source of at least some of the inflammation-related proteins whose circulating levels rise in obesity, the quantitative importance of the contribution from fat has not been established. A key issue, which has received little attention, is precisely why there should be a major increase in the production and release of inflammation-related adipokines in the obese. The parsimonious explanation, as advanced recently, is that it relates to specific events within WAT itself, raised plasma levels reflecting spillover from an “inflamed” tissue (4). The trigger may be hypoxia, through the recruitment of the transcription factor hypoxia-inducible factor-1, in clusters of adipocytes distant from the vasculature in the expanding adipose tissue mass in advance of angiogenesis (4).
The new work on MIF demonstrates that at least two major signals involved in the infiltration of WAT by macrophages are released in substantial quantities by human adipocytes, and MIF, like MCP-1, may be a key “obesity-dependent mediator of macrophage infiltration of adipose tissue” (9). The arrival of macrophages en masse is likely to result in a major amplification of the inflammatory state within adipose tissue, involving extensive cross-talk with mature adipocytes and also preadipocytes. Inflammation and its consequences, and the role of macrophages in particular, are destined to be hot topics in adipose tissue biology and obesity research over the next few years. One intriguing question is whether inflammation and macrophage infiltration are peculiar to obesity, or whether they are common to all situations in which there is a substantial expansion of adipose mass—such as in the pronounced prehibernatory fattening of ground squirrel species during which body weight can double over a matter of weeks.
Abbreviations
- MCP-1,
Monocyte chemoattractant protein-1;
- MIF,
migration inhibitory factor;
- WAT,
white adipose tissue.
