|
|
||||||||
Original Research |
Antibody Normalizes Serum Leptin in IL-2 Deficient Mice
Department of Nutrition and Food Science, University of Kentucky, and the Lexington Veterans Administration Medical Center, Lexington (L. M. G.)
Department of Internal Medicine University of Louisville, Louisville (H. S. O., C. J. M.), Kentucky
Department of Internal Medicine, Bristol-Myers Squibb, Lawrenceville Campus, Princeton, New Jersey (R. C. F.)
Address reprint requests to: Lisa Gaetke, Ph.D., R.D., Department of Nutrition and Food Science, University of Kentucky, 218 Funkhouser Building, Lexington, KY 40506-0054. E-mail: lgaetke{at}uky.edu
| ABSTRACT |
|---|
|
|
|---|
(TNF-
), a pro-inflammatory cytokine, contributes to the abnormally elevated leptin in IL-2-/- mice.
Methods: Eight week old, IL-2-/- and wild-type control (IL-2+/+), male mice were fed regular laboratory mouse food for two weeks. At the end of the study, blood was collected in the fed state, IL-2-/- and IL-2+/+ mice were injected with either anti-TNF-
monoclonal antibody or normal saline, and blood was collected in the starved state.
Results: The IL-2-/- mice consumed less food and lost weight. Administration of anti-TNF-
antibody markedly reduced serum leptin concentrations in IL-2-/- and control mice after food deprivation. Serum leptin in the IL-2-/- mice not receiving anti-TNF-
antibody increased significantly in the starved state. Serum concentrations of TNF-
were higher in IL-2-/- mice compared to controls in both the fed and starved state.
Conclusions: These results suggest that elevated TNF-
may be one mechanism for the sustained elevated leptin observed in IL-2-/- mice during food deprivation.
Key words: leptin, IL-2 deficient mice, tumor necrosis factor-
(TNF-
), inflammatory bowel disease (IBD)
Abbreviations: IBD, inflammatory bowel disease IFN-
, interferon-
IL, interleukin IL-2-/-, interleukin-2 deficient LIF, leukemia inhibitory factor LPS, lipopolysaccharide TNF-
, tumor necrosis factor-
| INTRODUCTION |
|---|
|
|
|---|
(TNF-
), in a setting of inflammation and immune dysfunction. Leptin secreted by the adipocyte plays a central role in food intake and energy balance and has materialized as a potential mediator of inappropriate satiety in inflammatory states, such as inflammatory bowel disease (IBD) [1,2]. Normally, serum leptin concentration rapidly declines with food restriction and appears to communicate an energy deficit and a wide range of adaptive responses including increased food seeking behavior to minimize loss of body weight [35]. Recently, we reported abnormally elevated serum leptin concentrations in interleukin-2 deficient (IL-2-/-) mice, a model of chronic inflammatory bowel disease [6]. IL-2-/- mice had inappropriately elevated leptin responses during food restriction compared to starved controls. Serum leptin concentrations in the food restricted IL-2-/- mice were not different from fed control mice, indicating that the IL-2-/- mice were receiving an inappropriate signal of satiety at the same time the mice were food restricted.
IL-2-/- mice have colitis with an associated wasting syndrome with elevated production of multiple pro-inflammatory cytokines, including interferon-
(IFN-
), interleukin-1 (IL-1), interleukin-6 (IL-6) and TNF-
by components of the gut and its immunoregulatory cells [710]. We, and others, have shown that certain pro-inflammatory cytokines, including IL-1, TNF-
, leukemia inhibitory factor (LIF), and probably IL-6, and lipopolysaccharide (LPS) elevate serum leptin [1,2]. An increasing number of studies have linked TNF-
with elevated leptin in inflammation and infection [1116]. However, other research examining diseases with evidence of inflammation and/or the presence of pro-inflammatory cytokines report low leptin concentrations [1720]. In addition, studies have demonstrated a regulatory role for leptin in immune function [2125]. We have found that our protocol for comparing leptin concentrations in the fed and fasted state and preventing diurinal variations may facilitate study of the pathophysiological dysregulation of the leptin system.
The purpose of this study was to investigate whether increased pro-inflammatory cytokines, particularly TNF-
, caused by chronic intestinal inflammation in IL-2-/- mice may induce elevated serum leptin and contribute to anorexia and reduced food intake in this animal model.
| MATERIALS AND METHODS |
|---|
|
|
|---|
General Procedures
Male mice (eight weeks old) were divided into four individually housed groups: homozygous IL-2tm1Hor mutants (IL-2-/-) with free access to food were divided into 1) IL-2-/- mice that received anti-TNF-
antibody in the fed state (n = 6) and 2) IL-2-/- that did not receive anti-TNF-
antibody in the fed state (n = 6). Matched wild-type (IL-2+/+) litter mates when possible (or with age-matched IL-2+/+ from other litters) as control mice with free access to food were divided into 3) IL-2+/+ that received anti-TNF-
antibody in the fed state (n = 6) and 4) IL-2+/+ that did not receive anti-TNF-
antibody in the fed state (n = 6).
After two weeks, the mice were anesthetized with methoxyfurane and blood collected by orbital puncture in the fed state at 0800. Animals were injected intraperitoneally (i.p.) with 160 µg of mouse anti-TNF-
monoclonal antibody (isotype:rat igG1) purchased from Endogen (Woburn, MA) or 160 µg of normal saline solution immediately after the fed state blood collection. Mice were then deprived of food for 24 hours, anesthetized, and exsanguinated by cardiac puncture (starved state).
In a second experiment, male (eight weeks old) IL-2-/- mice (n = 6) and wild-type litter mates (IL-2+/+) or age-matched IL-2+/+ from other litters as control mice (n = 6) were compared for serum TNF-
concentrations in the fed and starved state.
Immunoassays
Serum concentrations of leptin and all TNF-
studies, were performed using mouse ELISA kits purchased from R&D Systems (Minneapolis, MN). Serum concentrations of serum amyloid A (SAA), an acute phase protein, were detected using a mouse ELISA kit purchased from Biosource (Camarillo, CA).
Statistics
Means for total food intake for both groups were compared using independent sample t tests. Mean responses for body weight were compared by using a repeated measures analysis of variance (ANOVA) having between animal factor groups and within animal factor age (eight vs. ten weeks). Mean responses for serum leptin and serum TNF-
were compared by using a repeated measures ANOVA having the between animal factor groups (IL-2-/- vs. IL-2+/+) and within animal factor state (fed vs. starved). In all ANOVA, post hoc comparison of means was based on Fishers protected least significant difference procedure. Statistical significance was determined at the 0.05 level. All results are expressed as the mean ± SEM.
| RESULTS |
|---|
|
|
|---|
|
Concentrations
were higher in IL-2-/- mice (0.046 ± 0.003 µg/mL) and did not decline with food restriction (0.029 ± 0.004 µg/mL) in contrast to controls (0.015 ± 0.003 µg/mL in fed state vs. 0.005 ± 0.003 µg/mL in starved state, p < 0.03) (Fig. 1). These data suggested that elevated serum TNF-
concentrations which do not decline with food deprivation may be one mechanism for the elevated leptin.
|
|
antibody in the fed state had significantly elevated serum leptin concentrations in the starved state compared to all other groups in the starved state (p < 0.001) (Fig. 2). Serum leptin concentrations in the IL-2-/- mice administered anti-TNF-
antibody in the fed state (1.32 ± 0.23 µg/L) dropped dramatically in the starved state (0.099 ± 0.09 µg/L, p < 0.001). Serum leptin concentrations in IL-2+/+ control mice that did not receive anti-TNF-
antibody in the fed state (2.32 ± 0.21 µg/L) decreased as expected in the starved state (1.34 ± 0.08 µg/L), and IL-2+/+ control mice administered anti-TNF-
antibody in the fed state (2.24 ± 0.21 µg/L) dropped more quickly to a nadir (0.41 ± 0.08 µg/L) in the starved state. These data support our hypothesis that the elevated TNF-
concentration and TNF-
production in this model may play an important role in elevated serum leptin during food restriction.
Acute Phase Response Measurements
Previously, we have reported the high correlation between SAA and the severity of colitis [9,10], such that SAA was used again as a marker of the presence of an acute phase response and the degree of inflammation active in IL-2-/- mice in this study. At the end of two weeks, serum concentrations of SAA were elevated in IL-2-/- mice compared to IL-2+/+ mice (p < 0.02) (Table 1).
| DISCUSSION |
|---|
, as a result of the chronic intestinal inflammation characteristic of this mouse model, elevates leptin contributing at least in part to the anorexia observed in IL-2-/- mice. Our results support the hypothesis.
This mouse model (IL-2-/-) lacks the gene that encodes for the protein, IL-2, and fails to develop appropriate immune tolerance, which then leads to an autoimmune-mediated inflammatory process that affects several organs [7,8]. IL-2-/- mice develop a wasting syndrome associated with colitis including symptoms of decreased food intake, wasting, weight loss and increased acute phase reactants. In this model, pro-inflammatory cytokines known to stimulate leptin production are elevated [810]. Increased leptin, although somewhat more transient, has been reported in animal models of acute intestinal inflammation [27]. Here, we report for what we believe is the first time, increased serum TNF-
in an animal model of chronic intestinal inflammation in the fed and starved state compared to controls.
Once a neutralizing monoclonal antibody to TNF-
was administered to IL-2-/- mice in the fed state, serum leptin concentrations dropped in the starved state similar to, or greater than, the decreased leptin response in starved control mice. These data support our hypothesis that the elevated TNF-
concentrations in part elevate serum leptin during fasting. The drop in serum leptin in IL-2-/- mice, similar to that observed in control mice, would signal the hypothalamic centers to increase food intake, just as in controls. When anti-TNF-
antibody was administered to control mice, serum TNF-
was undetectable (data not shown). We have also studied the effect of IgG as an isotype control and have found no change in serum TNF-
concentrations (data not shown). A future study will investigate whether IL-2-/- mice injected continuously with anti-TNF-
antibody increase food intake and gain weight.
In a similar study [16], mice with bacterial peritonitis had increased plasma leptin and increased adipose tissue ob mRNA expression compared to pair-fed and control sham-treated mice. Mice with bacterial peritonitis pretreated with TNF-binding protein did not have elevated plasma leptin. Food intake was reduced in all groups despite the low plasma leptin concentrations with TNF-binding protein, and these authors concluded that leptin did not appear to regulate the anorexia associated with this mouse model. The model cited was a more acute model of infection compared to our chronic model of intestinal inflammation.
Results have not been consistent among studies investigating leptin concentrations in relation to anorexia and weight loss in cytokine-induced chronic inflammatory diseases. Patients demonstrating a systemic inflammatory response/multiple organ dysfunction had elevated serum leptin and TNF-
concentrations compared to controls [12], as did subjects administered TNF-
intravenously (i.v.) daily over five days [13]. Patients with chronic hepatitis C [14] and cirrhosis [15] had elevated leptin concentrations when compared to controls. Plasma leptin, as well as plasma cortisol, were higher in patients with acute sepsis [28].
In contrast, patients with cardiac cachexia characterized by wasting and elevated TNF-
had exceptionally low leptin concentrations in comparison to non-cachectic-heart patients, ischemic heart patients, and healthy controls [17]. Decreased leptin concentrations were reported in patients with tuberculosis and wasting [18]. In studies of IBD patients, there was no increase in serum leptin in pediatric and young adult patients with IBD compared to controls, and changes in body weight predicted leptin concentrations rather than disease type or disease activity [19,20]. The differences in conclusions of these various investigators may be due to differences in experimental design.
Our protocol examining samples taken with reference to the fed or starved/fasted state and minimizing diurnal variation may elucidate the pathophysiological changes in leptin [35,2931]. Had we only taken samples from mice with an eight to twelve hour food restriction, the leptin dysregulation would have been obscured. The "normal" concentrations observed in some IBD human studies [19,20] are inappropriately high and are similar to "normal" leptin concentrations that we observed in our food restricted IL-2-/- mice. Serum leptin concentrations in the starving IL-2-/- mice were not different from fed control mice. Thus, comparison of leptin in the fed and fasted state may more clearly reveal differences in leptin in future studies in patients with IBD.
It is clear that elevated leptin is not the only explanation for reduced feeding. Grunfeld and coworkers have demonstrated that LPS and its associated pro-inflammatory cytokines produced anorexia in the ob/ob and db/db mice that lack leptin or its receptor [32,33]. In studies where elevated leptin has not been reported, TNF-
and other proinflammatory cytokines may produce changes in energy homeostasis that mimic leptin [34]. It has also been suggested that serum leptin may vary based on severity or chronicity of the inflammatory response [35]. The significance of the observed elevated TNF-
and elevated leptin in relation to food restriction is also not clear.
Leptin is considered to be a cytokine [36,37], and may have immunoregulatory functions that extend beyond appetite regulation [2125]. Leptin has been shown to activate macrophages and induce expression of proinflammatory cytokines [21]. Leptin-deficient (ob/ob) mice given dextran sulfate sodium (DSS) to induce colitis had reduced intestinal inflammation compared to DSS-treated, wild-type, control mice, suggesting that leptin may mediate the inflammatory response [23]. Here, we did not study the effect of elevated leptin in the starved state on the inflammatory immune response in IL-2-/- mice.
Further studies are needed 1) that pay careful attention to feeding state and diurnal variation to uncover a significant role for the leptin system in humans, 2) to identify mechanisms other than leptin that may be responsible for a reduction in food intake and 3) to determine if therapeutic interventions to reduce serum leptin or its pro-inflammatory stimulants may be most effective during periods of reduced food intake.
In summary, this appears to be the first study to show increased serum TNF-
concentrations compared to controls during food restriction. When IL-2-/- mice with elevated TNF-
concentrations were administered anti-TNF-
antibody in the fed state, serum leptin concentrations dropped after 24 hours of food restriction, suggesting that elevated TNF-
may be one mechanism of sustained elevated leptin during fasting.
| ACKNOWLEDGMENTS |
|---|
|
|
|---|
Received December 19, 2002. Accepted May 13, 2003.
| REFERENCES |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |