| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Departments of Medicine and Obstetrics and Gynecology, Columbia University College of Physicians and Surgeons, New York, New York 10032
Address all correspondence and requests for reprints to: Dr. Sharon L. Wardlaw, Department of Medicine, Columbia University College of Physicians and Surgeons, 630 West 168th Street, New York, New York 10032. E-mail: sw22{at}columbia.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, IL-1ß, and IL-6 (5). Leptin-deficient mice are also more sensitive to the toxic effects of TNF (6). The increased sensitivity to endotoxin and TNF in ob/ob mice can be reversed by leptin administration. In normal mice, fasting is associated with a fall in leptin levels and an increased susceptibility to endotoxic shock, which can be reversed by leptin treatment (7). In the rodent, endotoxin has been reported to stimulate leptin gene expression in adipose tissue and leptin release into peripheral blood, responses thought to be mediated by IL-1ß (8, 9, 10, 11). In contrast to the rodent, a stimulatory effect of endotoxin on leptin release into peripheral blood has not been found in human subjects. Leptin levels, however, are known to be increased in critically ill, septic patients, and high levels are positively correlated with survival (12, 13, 14). It is unclear why the leptin responses to endotoxin differ in the rodent and human studies. In only one of the three previously reported human studies was leptin measured for more than 12 h after endotoxin injection. It is thus not known whether endotoxin would stimulate leptin in the human with a longer period of observation or with chronic administration. The effects of endotoxin on leptin in the nonhuman primate are also unknown. In this study we therefore examined the effects of acute and chronic endotoxin administration on plasma leptin levels in the monkey. We also measured leptin levels in blood samples from four postmenopausal women who received a low dose of endotoxin.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Nine adult female rhesus monkeys (Macaca mulatta), weighing 5.1–12 kg, were used in these experiments. Monkeys were housed in individual cages in a temperature-controlled room (19–22 C) with a 12-h light, 12-h dark photocycle and were fed 20 Purina monkey chow biscuits (
6.8 g each; Ralston Purina Co., St. Louis, MO) twice daily at 1000 and 1500 h. This was supplemented with fresh fruit or vegetables daily. All animals were ovariectomized at least 2 months before the studies to eliminate fluctuations in estradiol (E2) levels, because E2 has been shown to affect cytokine and neuroendocrine responses to endotoxin. All protocols were approved by the Columbia University institutional animal care and use committee and were conducted in accordance with the NIH Guide for the Care and Use of Laboratory Animals.
Monkey experimental protocols
Study 1. Plasma leptin concentrations were measured at baseline in the morning at 0900 h and then 2, 5, 24, and 48 h after iv injection of 100–200 µg endotoxin (lipopolysaccharide W Escherichia coli 055:B5, Difco Laboratories, Detroit, MI) in five monkeys. Blood samples were obtained by venipuncture without sedation, a procedure to which the animals had previously been habituated.
Study 2. Blood samples were obtained during a control period at 0900, 1300, and 1700 h in five monkeys. Endotoxin (100 µg) was then injected at 1700 h, and blood sampling was continued at 16 h (0900 h), 20 h (1300 h), 24 h (1700 h), and 48 h (0900 h) after endotoxin injection. To determine whether estrogen affects the leptin response to endotoxin in the monkey, we repeated the same experiment in five monkeys after E2 replacement for 3 wk (via sc implanted SILASTIC brand capsules, Dow Corning, Inc. Corp., Midland, MI).
Study 3. The effect of repeated administration of endotoxin on hormone and cytokine responses was studied in nine monkeys. Blood samples were obtained during a control period at 0900, 1300, and 1700 h on d 1. Endotoxin (100 µg) was then injected iv at 1700 h on d 1, at 0900 h on d 2, at 1700 h on d 2, at 0900 h on d 3, at 1700 h on d 3, and at 0900 h on d 4 for a total of six injections over the 4-d period. Blood samples were collected at 0900, 1300, and 1700 h throughout the 4-d period for leptin, ACTH, cortisol, and IL-6 measurements. Blood samples were centrifuged, and plasma was separated and stored at -20 C for cytokine and hormonal assays. Food intake was measured 2 d before the study and during the 4-d experimental protocol. Food intake was measured by counting the number of uneaten biscuits (±0.5 biscuits) twice daily at 0900 and 1700 h.
Human experimental protocol
In a previous experiment blood samples were obtained from six healthy postmenopausal female subjects, 42–68 yr old, to analyze cytokine, ACTH, and cortisol responses over a 7-h period after a low dose of endotoxin (2–3 ng/kg) (15). Four of the six subjects returned the following day to have a blood sample drawn 24 h after endotoxin injection. Plasma leptin levels were measured in the current study in blood samples from these four subjects. Informed consent was obtained from all subjects, and the study was approved by the Columbia-Presbyterian Medical Center institutional review board.
Subjects were taking no medications other than vitamin or mineral supplements. The mean (±SEM) concentration of E2 in peripheral blood was 7.3 ±.0.8 pg/ml. An iv catheter was inserted between 0800 and 0900 h after an overnight fast, and normal saline containing 5% glucose was infused at 60 ml/h. At 1000 h, purified endotoxin (2–3 ng/kg) was administered iv. Plasma leptin concentrations were measured hourly for 7 h and then at 24 h after endotoxin injection. Subjects ate a light lunch at 1700 h and then had dinner at their usual time. They then fasted overnight before obtaining the 24-h blood sample the next morning. Blood samples were centrifuged within 1 h, and plasma was separated and stored at -20 C for hormonal assays. Two endotoxin preparations were used. The first two subjects received endotoxin (purified lipopolysaccharide prepared from Escherichia coli, U.S. Pharmacopoeia Endotoxin Reference Standard, EC-5) obtained from U.S. Pharmacopoeia (Bethesda, MD). Because the EC-5 preparation became unavailable, the next two subjects received endotoxin (U.S. Standard Reference Endotoxin, PDS no. 67801) obtained from the Pharmaceutical Development Section, NIH (Bethesda, MD).
Hormone and cytokine assays
Leptin was assayed in monkey and human samples with a double antibody primate RIA kit that uses an antiserum to human leptin that cross-reacts fully with monkey leptin (Linco Research, Inc., St. Charles, MO). ACTH was measured in unextracted plasma by immunoradiometric assay (Nichols Institute Diagnostics, San Juan Capistrano, CA). Serum cortisol was assayed in unextracted plasma by solid phase RIA (Diagnostic Products, Los Angeles, CA). IL-6 was assayed by specific monoclonal sandwich immunoassay with a human ELISA kit (R Systems, Inc., Minneapolis, MN) that we have validated for use in the rhesus monkey (16). E2 was measured with a solid phase, chemiluminescent immunoassay (Immulite, Diagnostic Products, Los Angeles, CA).
Data analysis
Hormone levels before and after endotoxin administration were compared by Wilcoxon signed-rank test in the monkey studies and by paired t test in the human study. The area under the hormone response curve (AUC) was calculated by trapezoid analysis, and the responses before and after endotoxin injection were compared by paired Wilcoxon signed-rank test. The effects of endotoxin injection on hormone and cytokine responses in monkeys were also analyzed by ANOVA with repeated measures and using Bonferroni-Dunn post hoc analysis. Analyses were performed with statistical software (StatView, Abacus Concepts, Inc., Berkeley, CA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Study 1.
In an initial study in five monkeys, leptin levels did not change significantly during the first 5 h after endotoxin administration, but did increase significantly above baseline at 24 h (Fig. 1
). The mean (±SEM) plasma leptin concentration measured at baseline was 6.4 ± 2.1 vs. 12.3 ± 4.4 ng/ml at 24 h (P = 0.043).
|
). There was a significant increase in leptin over time after endotoxin treatment (P < 0.001). The mean plasma leptin concentration during the 8-h control period was 6.83 ± 3.0 ng/ml and increased to a mean of 10.1 ± 3.6 ng/ml during the same 8-h period the following day, 16–24 h after endotoxin injection (P = 0.043). The AUC calculated for leptin during the 16- to 24-h period after endotoxin treatment was 48% above the AUC during the 8-h control period (P = 0.043). This experiment was then repeated after the monkeys received E2 replacement for 3 wk. The mean plasma E2 level increased from less than 20 to 66 ± 6.4 pg/ml. The effect of endotoxin on plasma leptin in the E2-treated animals was similar to that seen in the untreated animals. Plasma leptin increased from a mean of 5.8 ± 2.5 ng/ml during the 8-h control period to a mean of 8.4 ± 3.6 ng/ml 16–24 h after endotoxin treatment (P = 0.043). The AUC calculated for leptin during the 16- to 24-h period after endotoxin injection was 47% above the AUC during the 8-h control period (P = 0.043).
|
Study 3.
The effects of repeated injections of endotoxin over a 3-d period were studied in nine monkeys. Daily food intake decreased in all monkeys after endotoxin administration (P < 0.0001). Mean daily intake decreased from 38.6 ± 0.6 biscuits at baseline to 20.5 ± 3.4, 10.6 ± 1.5, and 19.7 ± 3.3 at 24, 48, and 72 h, respectively, after endotoxin injection. During the control period animals ate 40% of their food between 0900–1700 h and 60% between 1700–0900 h. The same pattern persisted after endotoxin. As shown in Fig. 3, there was a significant stimulation of IL-6, ACTH, cortisol, and leptin release over time (P < 0.001). Responses to endotoxin decreased with repeated injections, and only the IL-6 and cortisol responses remained significant on the second and third days of treatment (Fig. 3). Leptin increased from a mean of 9.1 ± 2.4 ng/ml on the control day to a mean of 13.5 ± 3.0 ng/ml (P = 0.008) during the first day of treatment. Mean leptin levels in the entire group were no longer significantly elevated during the second and third days of treatment. However, if the monkeys were divided into two groups based on mean baseline leptin levels, two distinct patterns of leptin response to endotoxin administration emerged (Fig. 4). Monkeys with baseline plasma leptin levels greater than 10 ng/ml exhibited a sustained increase in leptin throughout the 3-d treatment period (P < 0.020). Monkeys with baseline leptin levels less than 10 ng/ml had a transient increase in leptin for 24 h, with a subsequent decrease in leptin levels. As shown in Fig. 4, there was a highly significant correlation (r = 0.81; P = 0.008) between the mean baseline leptin level and the percent increase in leptin over baseline on the last day of treatment.
|
|
This dose of endotoxin produced a mean peak temperature of 38.3 ± 0.1 C. As reported previously, mean peak ACTH and cortisol levels increased to 411 ± 144 pg/ml and 30.8 ± 3.2 µg/dl, respectively (15). Plasma leptin concentrations did not change significantly during the 7-h period after endotoxin injection (Fig. 5). However, leptin levels increased in all four women from a mean baseline of 8.34 ± 3.1 to 13.1 ± 4.3 ng/ml 24 h after endotoxin administration (P = 0.038; Fig. 5).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and of the hypothalamic-pituitary-adrenal axis was attenuated by estrogen replacement (15). Speculation about the role of estrogen in modulating the leptin response to endotoxin led us to study female ovariectomized monkeys with and without estrogen replacement. We found little difference in the leptin response to endotoxin in the estrogen-deficient and the estrogen-replaced monkeys. The time of endotoxin injection and the pattern of food intake in the experimental subjects also need to be considered when comparing our human study to previous studies. One human study was performed at night without a change in the pattern of food intake (19), whereas the other two were performed in the morning, similar to our study (17, 18). The pattern of food intake described in the study by Granowitz et al. (18) is similar to that in our study, but the duration of that study was only 12 h. It is unlikely that the increase in plasma leptin in our subjects is related to a shift in the diurnal rhythm of leptin secondary to a shift in meal pattern. Such a change has been reported when the timing of all three meals was shifted by 7 h (20). In our study, although lunch was delayed, dinner was eaten at the usual time on both days. Food intake was, however, not quantitated in the human subjects, and a control study with a saline infusion and the same pattern of food intake was not performed. Although an effect of food intake cannot be completely ruled out in the human subjects, a similar pattern of leptin increase was noted in the monkeys despite a marked reduction in food intake after endotoxin injection.
In contrast to the rodent, no short-term stimulatory effect of endotoxin on leptin release into peripheral blood has been detected in any of the human or monkey studies. In previous experiments in rats, mice, and hamsters, endotoxin and the inflammatory cytokines IL-1ß and TNF have been shown to stimulate leptin gene expression in adipose tissue and peptide release into peripheral blood (8, 9, 10, 11, 21, 22). Endotoxin was reported to increase leptin release into peripheral blood as early as 2 h after injection, and the increase persisted for 6 h. The stimulation of leptin gene expression in adipose tissue was reported to be maximal 6 h after endotoxin injection (9). Several studies demonstrate that the stimulation of leptin by endotoxin is mediated by IL-1ß. Endotoxin failed to stimulate leptin in IL-1ß-deficient mice or in normal mice treated with the soluble IL-1 receptor (9, 11). TNF may also play a role in regulating the increase in leptin induced by endotoxin, although there is some controversy about the effects of TNF on leptin in cultured adipocytes (21). There is some evidence that suggests that the early 2 h stimulatory effect of endotoxin on leptin release may be neurally mediated, as this effect can be blocked by anesthesia (22). The failure to detect acute changes in leptin release in response to endotoxin in our studies suggests that the neural mechanisms that may be important in the rodent do not mediate the leptin response to endotoxin in the primate.
Endotoxin also stimulates the HPA axis, which may exert secondary effects on leptin release. Glucocorticoids directly stimulate leptin synthesis in adipocytes in vitro and in human adipose tissue in vivo (23, 24). Treatment with dexamethasone has been reported to increase plasma leptin levels in human subjects at 24–48 h (24). It is thus likely that endotoxin affects leptin expression and release by more than one mechanism and that there are species differences with respect to the mechanisms involved in the short-term leptin response to endotoxin. It could be argued that the failure to detect acute changes in plasma leptin levels in the human studies results from the relatively low doses of endotoxin that were used compared with the rodent studies. However, no acute effect of endotoxin on leptin release was seen in the monkey despite the fact that a much larger dose of endotoxin (100–200 µg) was used. It is thus likely that there are some differences in the mechanisms that regulate endotoxin-induced leptin release in the rodent compared with the primate, although interspecies differences in endotoxin sensitivity must also be considered. The role of IL-1ß in mediating the leptin response to endotoxin has not been studied in the primate. However, in a study of human cancer patients, the administration of IL-1 was reported to increase plasma leptin levels 24 h after injection (25). This is consistent with a role for IL-1 in mediating the effects of endotoxin on leptin in the human and with the time course of the leptin response to endotoxin in the current study. TNF has also been reported to increase serum leptin levels 12 h after injection in human cancer patients (26).
The stimulation of leptin release by endotoxin in the current study is consistent with previous studies documenting an increase in the concentration of leptin in the blood of patients hospitalized with sepsis (12, 13, 14). Leptin levels were significantly higher, and IL-6 levels were significantly lower in patients who survived the septic episode compared with nonsurvivors (14). A positive correlation between circulating levels of leptin and the antiinflammatory cytokines, IL-1 receptor antagonist and IL-10, was also noted in septic patients (13). In animal studies, alterations in immune and inflammatory responses have been reported in rodents with genetic leptin deficiency and during starvation, when leptin levels are suppressed. Leptin deficiency is associated with decreased responsiveness to T cell-activating stimuli, resulting in impaired T cell immunity (27), and increased sensitivity to monocyte/macrophage-activating stimuli, resulting in increased sensitivity to endotoxin (5). Leptin-deficient mice (ob/ob) are more sensitive to the lethal effects of bacterial endotoxin and TNF (5, 6). This increased sensitivity to endotoxin and TNF in ob/ob mice can be reversed by leptin administration. The normal induction of the antiinflammatory cytokines, IL-1ra and IL-10, by endotoxin is attenuated in ob/ob mice (5). The increased susceptibility to endotoxic shock in normal fasted mice can also be reversed by leptin treatment (7). Thus, leptin appears to exert important antiinflammatory effects under physiological conditions in the human and the rodent.
Leptin and its receptor are structurally and functionally related to the IL-6 cytokine family (28). Leptin activates signal transduction, like other members of this family, by stimulating the Janus kinase-signal transducer and activator of transcription pathway. Leptin also induces the expression of SOCS-3, a member of the suppressor of cytokine signaling (SOCS) proteins, which inhibits cytokine signal transduction (29) (30). The long form of the leptin receptor is expressed by tissues and cells of the immune system, including bone marrow, spleen, monocytes/macrophages, CD4+ and CD8+ T lymphocytes, and CD34+ cells (4, 31). Accumulating evidence suggests that leptin can modulate the response to endotoxin by multiple mechanisms. Leptin may modulate the production of proinflammatory and antiinflammatory cytokines and may also affect cytokine signaling by a variety of mechanisms, including activation of SOCS-3. In a recent study in the monkey we have shown that infusion of leptin for 16 h at a dose that increases plasma levels within the physiological range attenuated the IL-6 and HPA responses to endotoxin (32). Thus, endotoxin can stimulate leptin release, which can, in turn, modulate the inflammatory response to endotoxin. In the present study there was a significant correlation between the mean baseline leptin level and the percent increase in leptin over baseline on the last day of chronic endotoxin treatment. It is not known whether increased basal leptin levels are protective during a subsequent episode of sepsis, but these data suggest that suppressed leptin levels during starvation or wasting illness could increase the vulnerability to sepsis. During an episode of sepsis, however, it has been shown that plasma leptin levels are 3-fold higher in patients who survived the septic episode than in nonsurvivors (12).
In our study in the monkey, leptin levels were increased 16–24 after endotoxin despite a 50% decrease in food intake, which under other circumstances would be expected to result in a decrease in blood leptin levels. As leptin levels are elevated in experimental models of inflammation, and leptin decreases food intake, it has been postulated that leptin may be responsible for the anorexia that occurs during chronic inflammation. Endotoxin, however, been shown to induce equivalent or even worse anorexia in leptin-deficient or leptin receptor-deficient rodents (33, 34). Thus, leptin is not required for endotoxin-induced anorexia, but it remains to be determined whether the stimulation of leptin during inflammation can contribute to the anorexia induced by endotoxin. Alternatively, if leptin attenuates endotoxin-induced proinflammatory cytokine release, many physiological effects, including anorexia, may be reduced.
In summary, our data show that endotoxin stimulates the release of leptin into peripheral blood in the human and the nonhuman primate, but the time course is different from what has been previously reported in the rodent. These results are consistent with previous reports of increased blood leptin levels in patients with sepsis and further support a potential role for leptin in modulating the response to inflammation in the human.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: AUC, Area under the hormone response curve; E2, estradiol; HPA, hypothalamic-pituitary-adrenal; SOCS, suppressor of cytokine signaling.
Received August 30, 2002.
Accepted December 13, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
in the induction of leptin by lipopolysaccharide. Endocrinology 139:2278–2283 increases serum leptin concentrations in humans. J Clin Endocrinol Metab 82:3084–3086This article has been cited by other articles:
 |
|
 |
|
  S. S. Martin, A. Qasim, and M. P. Reilly Leptin Resistance: A Possible Interface of Inflammation and Metabolism in Obesity-Related Cardiovascular Disease J. Am. Coll. Cardiol., October 7, 2008; 52(15): 1201 - 1210. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Matarese, C. Procaccini, and V. De Rosa The intricate interface between immune and metabolic regulation: a role for leptin in the pathogenesis of multiple sclerosis? J. Leukoc. Biol., October 1, 2008; 84(4): 893 - 899. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Ballou, G. D. Cruz, W. Pittroff, D. H. Keisler, and E. J. DePeters Modifying the Acute Phase Response of Jersey Calves by Supplementing Milk Replacer with Omega-3 Fatty Acids from Fish Oil J Dairy Sci, September 1, 2008; 91(9): 3478 - 3487. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Vila, C. Maier, M. Riedl, P. Nowotny, B. Ludvik, A. Luger, and M. Clodi Bacterial Endotoxin Induces Biphasic Changes in Plasma Ghrelin in Healthy Humans J. Clin. Endocrinol. Metab., October 1, 2007; 92(10): 3930 - 3934. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. D. Anderson, N. N. Mehta, M. L. Wolfe, C. C. Hinkle, L. Pruscino, L. L. Comiskey, J. Tabita-Martinez, K. F. Sellers, M. R. Rickels, R. S. Ahima, et al. Innate Immunity Modulates Adipokines in Humans J. Clin. Endocrinol. Metab., June 1, 2007; 92(6): 2272 - 2279. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Schaffler, U. Muller-Ladner, J. Scholmerich, and C. Buchler Role of Adipose Tissue as an Inflammatory Organ in Human Diseases Endocr. Rev., August 1, 2006; 27(5): 449 - 467. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. Velez, E. Costamagna, E. T. Kimura, L. Fozzatti, C. G. Pellizas, M. M. Montesinos, A. M. Lucero, A. H. Coleoni, P. Santisteban, and A. M. Masini-Repiso Bacterial Lipopolysaccharide Stimulates the Thyrotropin-Dependent Thyroglobulin Gene Expression at the Transcriptional Level by Involving the Transcription Factors Thyroid Transcription Factor-1 and Paired Box Domain Transcription Factor 8 Endocrinology, July 1, 2006; 147(7): 3260 - 3275. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. R. Vulliemoz, E. Xiao, L. Xia-Zhang, M. Ferin, and S. L. Wardlaw Melanocortin Modulation of Inflammatory Cytokine and Neuroendocrine Responses to Endotoxin in the Monkey Endocrinology, April 1, 2006; 147(4): 1878 - 1883. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. E. Weber and M. E. Spurlock Leptin alters antibody isotype in the pig in vivo, but does not regulate cytokine expression or stimulate STAT3 signaling in peripheral blood monocytes in vitro J Anim Sci, June 1, 2004; 82(6): 1630 - 1640. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Xiao, L. Xia-Zhang, N. R. Vulliemoz, M. Ferin, and S. L. Wardlaw Leptin Modulates Inflammatory Cytokine and Neuroendocrine Responses to Endotoxin in the Primate Endocrinology, October 1, 2003; 144(10): 4350 - 4353. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
