EVOLUTIONARY IMPORTANCE OF INFLAMMATION AND INSULIN RESISTANCE

The adipose tissue is a necessary survival characteristic of species without constant access to food. Humans developed the ability to store fat for periods of famine. We now have virtually unlimited access to food and accumulate excess energy for no useful purpose. A similar reasoning is applicable for energy output. We regulate energy output voluntarily by physical exercise and inactivity is recognized as an important cause of obesity. Besides famine, infection was the other major threat for survival of prehistoric man. A wide array of genes in the human genome are committed to two integrated problems: a) the inflammatory response to survive trauma, infection, and other threats to organ integrity; and b) the metabolic adaptation to scarce energy input during periods of food deprivation. These evolutionary pressures selected individuals able to mount efficacious inflammatory responses and an "energy-sparing" metabolic attitude (i.e., insulin resistance). Primitive man had a very active lifestyle, and atherosclerosis and insulin resistance did not represent health problems for our progenitors. In our era, sedentary habits, food abundance, and the reduced infection death toll have paved the way to the epidemic of insulin resistance syndrome, diabetes, and atherosclerotic complications.

INFLAMMATION AND THE CYTOKINE SYSTEM OF THE ADIPOSE TISSUE

The immunologic system burns up about 15% of resting metabolic rate and therefore it represents a major energy-consumer among human organ systems. Such a high consumption clearly requires adequate sources of energy, and the adipose tissue appears well equipped to provide such a high-energy input. The metabolic potential of adipose tissue and the energy needs of the immune system have probably led to evolutionary development of feedback relationships between the two systems. Thus, when the immune system is challenged, an important flow of energy is required. As a result, cytokines like TNF- are recruited to induce lipolysis¹ and to mobilize energy. Conversely, when adipose tissue stores are depleted, energy sparing becomes a priority, even at the cost of imposing a decrease in the level of immunologic surveillance. Therefore, in this situation, switching off the immune response would appear to be an appropriate counter-regulatory response. Cachexia is indeed associated with a significant impairment of cell-mediated immunity, phagocyte function, complement factors, and cytokine production. TNF- is a very fundamental "internal regulator" of adipose tissue metabolism, and IL-6 and IL-1 are relevant control factors, as well. Indeed, both TNF- and IL-6² inhibit lipoprotein lipase, while TNF- down-regulates insulin-dependent glucose uptake via glucose transporter 4 and insulin receptor autophosphorylation, thus modulating lipid accumulation.

Established inflammation-triggers like lipopolysaccharides are able to stimulate the synthesis and the release of cytokines from adipocytes, but factors responsible for the regulation of these substances in the adipose tissue are still not completely understood. TNF- has well-defined effects on adipocyte trophism in that it controls cell size and may induce apoptosis. Furthermore, this cytokine has a biphasic action on leptin secretion (initial stimulation followed by suppression)³, and inhibits the expression of ₃ receptors. The effect of TNF- on ₃ receptors is of relevance because these receptors mediate sympathetically-mediated lipolysis in white adipose cells and thermogenesis in brown adipose tissue⁴.

Notwithstanding, the dominant regulatory role of TNF- within the adipose tissue, leptin, and IL-6, but not TNF-, appears to be the major signals linking the adipose tissue to systemic immunologic response. To date there is no evidence that TNF- is released systemically by the adipose tissue, while it is well established that leptin⁵ and IL-6⁶ have important effects on the hypothalamus, where they suppress appetite. IL-6 promotes the synthesis of C-reactive protein and other acute-phase proteins in the liver, while leptin modulates insulin secretion in the pancreas⁷. Furthermore, leptin is strictly related to body fat and sex, and has a variety of actions, which span from interference with sympathetic activity to hematopoiesis and reproductive function⁸.

In 1996, the discovery of adiponectin, a collagen-like protein of the collectin family that is synthesized exclusively in the adipose tissue and which is the most abundant adipose tissue-derived plasma protein, expanded knowledge of adipocyte proteins. The relationship of this protein with fat and body mass is opposite to that of leptin; adiponectin is significantly reduced in obese subjects in comparison to lean, healthy control subjects⁹. Like plasma leptin, adiponectin plasma concentration also appears to be sex-dependent¹⁰. Adiponectin is emerging as a pleiotropic cytokine linked not only to fat mass and energy balance but also to other fundamental body functions like hematopoiesis and immunity¹¹.

ADIPOKINES IN END-STAGE RENAL DISEASE (ESRD): THE ACUTE-PHASE RESPONSE, LEPTIN, AND ADIPONECTIN

During evolution, organisms developed the acute-phase response so as to localize and limit the injury and to clear necrotic products, and to confine the effects of the original insult. An important part of this defensive response involves C-reactive protein and complement. This response is orchestrated by cytokines like TNF-, IL-1, IL-6, and other pro-inflammatory cytokines, which are produced both by immunocompetent cells and by other cells (e.g., cardiomyocytes in the heart or smooth muscle cells, or endothelial cells in the vascular system). The predominant effects of these proteins are local, at the site of injury, but these cytokines also have systemic effects. IL-6 is the cytokine with the longest serum half-life and with the most well-defined endocrine effects.

Although the cause(s) of cachexia in ESRD are incompletely understood, evidence is emerging that this complication is multifactorial, and that inflammation is a prominent correlate of it¹². The acute-phase response has nutritional implications because it is energy-intensive and requires large quantities of essential amino acids. When this process goes unabated, the inflammatory response is no longer protective and inflammation becomes a strong risk factor for mortality and cardiovascular complications. The relationship between acute-phase reactants/cytokines and all-cause and cardiovascular mortality has been intensively investigated in patients with ESRD. It has been coherently observed that acute-phase reactants like CRP are independently associated to atherosclerosis¹³, death, and cardiovascular complications^14,15. IL-6 correlates with the severity of atherosclerosis¹⁶, and high TNF-, IL-1, and IL-6 are associated with shorter survival¹⁷ in dialysis patients. It is presently unknown whether cytokines generated in the adipose tissue in uremic patients contribute to systemic complications.

Probably due to its anorexigen effect, leptin is the most studied adipokine in ESRD. In ESRD, abolished renal clearance is the primary factor responsible for hyperleptinemia, and in uremic patients, the physiologic relationship between leptin and the BMI is shifted to the right in comparison to that in healthy subjects¹⁸. On the other hand, the ob gene is overexpressed in uremic patients with high CRP, suggesting that inflammation may contribute to hyperleptinemia¹⁹. Yet careful observations by Don et al²⁰ demonstrated that leptin behaves as an inverse acute-phase reactant (i.e., plasma leptin decreases) during spontaneous episodes of the acute-phase response, a phenomenon presumably due to a simultaneous, inflammation-induced, decrease in IGF-1. Another important reason for the interest in leptin in ESRD depends on its direct link with plasma insulin and on its inverse association with erythropoietin dose. The relationship between leptin and insulin is also confirmed in the Cardiovascular Risk Extended Evaluation in Dialysis (CREED) study cohort. Indeed, in this study leptin was directly related to plasma insulin, as well as to the HOMA index (a measure of insulin sensitivity) Figure 1. Of note, the link between leptin and this index was largely independent of BMI and other risk factors. Leptin is an independent predictor of adverse cardiovascular events in the general population. Because leptin has no direct atherogenic properties, this data may simply indicate that the plasma concentration of this protein may be a risk marker. To date, no study in uremic patients has explored whether leptin predicts survival and cardiovascular events in ESRD.

Figure 1.

Relationship between adiponectin and leptin. Data are median values and interquartile ranges.

Full figure and legend (21K)

Adiponectin has been investigated little in patients with chronic renal diseases. We have shown that, independent of treatment modality and diabetes, the plasma concentration of this adipokine is markedly raised in dialysis patients, and that it is related to several fundamental metabolic fuels and hormones like glucose, insulin, and triglycerides²¹. Though adiponectin and leptin are both markedly increased in dialysis patients, the two hormones are inversely, rather than directly, related Figure 2. This phenomenon suggests that factors other than abolished renal function play a role in the regulation of the plasma concentration of these substances in uremic patients. In our study, adiponectin was unrelated to CRP, but serial measurements during intercurrent episodes of inflammation will be required to prove that that this cytokine is truly "inflammation-insensitive."

Figure 2.

Direct relationship between leptin and the HOMA index in dialysis patients (hemodialysis N = 228, CAPD = 51) in the CREED study (Cardiovascular Risk Extended Evaluation in Dialysis patients). The relationship between leptin and the HOMA index remained significant also after adjustment for the BMI (partial r = 0.24, P < 0.01).

Full figure and legend (22K)

Interest on adiponectin derives from its potential protective role for the cardiovascular system. Adiponectin inhibits TNF-–induced monocyte adhesion and adhesion molecule expression in vitro, and modulates the endothelial response to inflammatory stimuli²². In line with this hypothesis, it has been recently reported that adiponectin is reduced in patients with coronary heart disease²² and in type 2 diabetics⁹, in which the plasma concentration of this protein tends to be inversely related to the severity of cardiovascular damage. Atherosclerosis is an inflammatory disease and inflammation (as quantified by CRP) is an independent marker of the severity of atherosclerosis in uremic patients¹³. If the inflammatory response detected in atherosclerotic lesions is counteracted in vivo by adiponectin, this protein may have a role for the prevention and/or the retardation of atherogenesis in various diseases, including chronic renal failure. The multiple links of adiponectin to several metabolic risk factors like glucose, triglycerides, insulin, and HDL cholesterol in uremic patients²¹ are all coherently in line with the hypothesis that adiponectin is a protective factor. In this regard, it is of great interest that plasma adiponectin is an inverse predictor of incident cardiovascular events in dialysis patients, and that this link is independent of traditional and nontraditional risk factors (36). However, adiponectin was much increased (20.8 6.8 g/mL), not only in patients with a low incidence of cardiovascular events Figure 3, but also in those with a high incidence (9.3 2.7 g/mL vs. 5.9 2.6 g/mL in healthy subjects), though to a much lesser extent. This observation clearly suggests that the biological phenomena underlying the cardiovascular protective role of adiponectin must be down-regulated in end-stage renal disease, thus resetting the relationship between this protein and cardiovascular damage and clinical complications in these patients at a higher level. This hypothesis has implications well beyond renal failure because it suggests that supranormal circulating levels may exert a protective effect. In vitro and in vivo studies, as well as observations in human subjects are clearly required to better understand the role of this most interesting cytokine in human diseases.

Figure 3.

COX events free survival curves of patients divided on the basis of the median value (cut-off: 13.2 g/mL) of plasma adiponectin. Data were adjusted for all significant predictors of fatal and nonfatal cardiovascular outcomes (age, smoking, and previous cardiovascular events). The number of cardiovascular events was 32 (28%) in the first group and 47 (41%) in the second group.

Full figure and legend (8K)

References

References