Regulation of T Cell-Mediated Hepatic Inflammation by Adiponectin and Leptin
http://www.100md.com
内分泌学杂志 2005年第5期
Divisions of Infectious Diseases (J.A.S., R.F., C.A.D., G.F.) and Endocrinology, Metabolism, and Diabetes (A.M.M., R.H.E.), University of Colorado Health Sciences Center, Denver, Colorado 80262; Rockefeller University (E.A., J.M., J.M.F.), New York, New York 10021; and Howard Hughes Medical Institute (J.M.F.), Chevy Chase, Maryland 20815
Address all correspondence and requests for reprints to: Dr. Giamila Fantuzzi, Department of Human Nutrition, University of Illinois, 1919 West Taylor Street, M/C 517, Chicago, Illinois 60612. E-mail: giamila@uic.edu.
Abstract
Concanavalin A-induced hepatotoxicity was compared in lipodystrophic aP2-nSREBP-1c transgenic mice (LD mice) lacking adipose tissue, obese leptin-deficient ob/ob mice, and lean wild-type (WT) mice. Serum leptin and adiponectin were low in LD mice, whereas ob/ob mice had undetectable leptin, but high adiponectin. Protection from hepatotoxicity was observed in ob/ob, but not in LD mice, despite low cytokine levels and reduced T cell activation and hepatic natural killer T cells in both groups. Administration of adiponectin protected LD mice from hepatotoxicity without altering cytokine levels. In contrast, administration of leptin heightened disease susceptibility by restoring cytokine production. Neutralization of TNF protected LD mice from liver damage. Increased in vivo susceptibility to the hepatotoxic effect of TNF was observed in LD mice. In vitro, adiponectin protected primary hepatocytes from TNF-induced death, whereas leptin had no protective effect. In conclusion, although leptin increases susceptibility to hepatotoxicity by regulating cytokine production and T cell activation, adiponectin protects hepatocytes from TNF-induced death.
Introduction
UNTIL RECENTLY, THE white adipose tissue (WAT) had been considered an inert tissue, mainly devoted to energy storage. However, with the discovery of several adipocyte-derived factors, collectively known as adipokines, WAT is currently regarded as an active player in the regulation of metabolism (1). The observations that obesity is associated with systemic inflammation and that malnutrition is accompanied by immunodeficiency have pointed to a possible role of adipokines in the regulation of immunity and inflammation (2). Leptin and adiponectin, two of the most studied adipokines, play direct regulatory roles in a variety of inflammatory conditions, both in vivo and in vitro (3).
Adiponectin is the most abundant circulating adipokine in both mice and humans (4). Several cell types, including macrophages, express adiponectin receptors (5). Adiponectin directly affects the inflammatory response by regulating both the production and the activity of cytokines (6, 7, 8) and can also act as an antiapoptotic agent in a variety of cell types (9, 10, 11). In fact, recent data indicate that adiponectin plays an antiinflammatory role in both acute and chronic inflammatory liver disease in vivo in mice (12, 13, 14).
The role of leptin in the regulation of immunity and inflammation has been well described. Leptin-deficient ob/ob mice are protected from inflammation/autoimmunity in several experimental models (15, 16, 17, 18). In particular, ob/ob mice exhibit markedly reduced disease severity in the experimental model of fulminant autoimmune hepatitis induced by administration of the T cell mitogen Concanavalin A (Con A), whereas reconstitution with exogenous leptin restores their response (15, 19). Cytokines, particularly TNF, play a crucial role in Con A-induced liver damage (20, 21). CD4+ T lymphocytes as well as natural killer (NK) T cells also contribute to the disease (22). In ob/ob mice, a reduced response to Con A-induced hepatitis is associated with reduced levels of selected proinflammatory cytokines and a lower percentage of intrahepatic NK T cells (15, 19).
Despite increasing evidence that adipocyte-derived factors modulate inflammatory reactions, most reports to date have focused on dissecting the role of single adipokines. In the present study we evaluated the role of WAT itself in regulating inflammation. We used lipodystrophic aP2-nSREBP-1c transgenic mice (LD mice), in which WAT is virtually absent due to the lack of differentiated white adipocytes (23). Insulin resistance, diabetes mellitus, as well as pronounced hepatomegaly with steatosis are present in LD mice, highlighting the pivotal role of WAT-derived factors in regulating metabolic homeostasis. In particular, administration of exogenous leptin reverses the glycemic alterations of LD mice, a therapeutic effect also observed in patients with generalized lipodystrophy (24, 25).
In the current study we used the model of Con A-induced hepatitis in LD mice to evaluate the role of WAT in regulating cytokine production and liver damage after a stimulus directed to T lymphocytes. A second experimental model, depletion of WAT by chronic administration of leptin, was used to confirm results obtained in LD mice. Reconstitution studies indicated the presence of a coordinate interplay between leptin and adiponectin in the regulation of T cell-mediated hepatic inflammation. Moreover, the role of adiponectin in protecting hepatocytes from TNF-induced cell death was evaluated both in vivo and in vitro.
Materials and Methods
Mice
Animal protocols were approved by the animal studies committee of University of Colorado Health Sciences Center and the Rockefeller University. LD mice, generated as previously described (23), were purchased from the Jackson Laboratory (Bar Harbor, ME). LD males (C57BL/6JxSJL background) were crossed to C57BL/6J females. LD females and littermate controls [wild type (WT)] generated from this cross were used. Female leptin-deficient obese ob/ob mice (B6.V-Lepob/J) and their lean littermates (WT) on a C57BL/6J background were obtained from The Jackson Laboratory. Mice of each strain were 8–12 wk old at the time of the experiment.
Administration of Con A
Con A (type IV-S, Sigma-Aldrich Corp., St. Louis, MO) was injected iv in the tail vein at a dose of 200 μg/mouse, as previously described (15). Mice were killed by cervical dislocation under isoflurane anesthesia at various time points after the administration of Con A for evaluation of cytokine and alanine aminotransferase (ALT) levels as well as liver histology. For neutralization of TNF activity, soluble TNFRp55 (Amgen, Thousands Oaks, CA) was administered ip 1 h before Con A injection at a dose of 200 μg/mouse.
Administration of TNF and D-galactosamine (D-Gal)
WT and LD mice received an ip injection of 100 ng murine recombinant TNF (R&D Systems, Minneapolis, MN) together with 20 mg D-Gal. Serum ALT levels were measured 6 h later.
Administration of leptin and adiponectin
LD mice were treated with recombinant murine leptin (5 μg/d; Amgen) and/or adiponectin (30 μg/d; Alexis Corp., San Diego, CA) in PBS for 6 d using sc placed Alzet miniosmotic pumps (Durect Corp., Palo Alto, CA). Body weight was measured daily.
Depletion of WAT
Depletion of WAT was performed as previously described (26). Alzet pumps (Durect Corp.) with an exchange rate of 0.5 μl/h were filled aseptically with either sterile PBS solution or leptin (Amgen) at a concentration of 5 μg/μl and implanted sc into C57BL/6J mice. Leptin was withdrawn by removing the pumps after 8 d of leptin treatment. At this point, mice were fed for 48 h a diet normocaloric to their caloric intake during leptin administration.
Flow cytometry
Hepatic mononuclear cells were isolated from the liver as previously described (19). Flow cytometry followed routine procedures using 1 x 105 cells/sample. Staining with annexin V and propidium iodide (BD Pharmingen, San Diego, CA) was used for evaluation of the rate of apoptosis. Surface marker expression was evaluated using anti-CD3, anti-CD69, and anti-NK1.1 antibodies from Caltag Laboratories (Burlingame, CA). Analysis was conducted on a FACSCalibur (BD Pharmingen) using the CellQuest analysis program (BD Pharmingen).
Leptin, adiponectin, cytokine, and ALT measurements
Leptin levels were measured using a mouse leptin ELISA (R&D Systems). The sensitivity of the assay is 0.1 ng/ml. Adiponectin levels were measured using an RIA (Linco Research, Inc., St. Charles, MO; sensitivity, 1 ng/ml). TNF, IL-10, and IL-18 levels were measured using an electrochemiluminescence method as previously described (18). The sensitivity of these assays is 10 pg/ml for TNF and IL-10 and 30 pg/ml for IL-18. IL-4, IL-6, and IFN levels were measured using an ELISA (BD Pharmingen). The sensitivity of these assays is 20 pg/ml. Inter- and intraassay variabilities for each of the assays used were less than 20%. Time points for cytokine measurement were chosen according to previously published data and were as follows: TNF, IL-4, and IL-10, 2 h after Con A; IL-6, IL-18, and IFN, 6 h after Con A. ALT levels were measured by the clinical laboratory of University Hospital (Denver, CO).
TUNEL assay
Formalin-fixed sections of livers obtained 24 h after Con A were processed using a TUNEL kit from Promega Corp. (Madison, WI) according to the manufacturer’s instructions.
Hepatocyte isolation and culture
Primary hepatocytes were isolated from WT mice as previously described (27). Cells were incubated with murine recombinant adiponectin (30 μg/ml) and/or leptin (100 ng/ml) for 1 h before the addition of murine recombinant TNF (100 ng/ml) and actinomycin D (20 nM) (28). Cell death was evaluated after an overnight incubation using a lactate dehydrogenase kit from Promega Corp.
Statistical analysis
Data are expressed as the mean ± SEM. The statistical significance of differences between treatment and control groups was determined by factorial ANOVA. Statistical analyses were performed using XLStat software (Addinsoft, Brooklyn, NY).
Results
Leptin levels in LD mice
LD mice lack differentiated WAT, the primary source of circulating leptin (23). As expected, serum leptin levels were extremely low in LD mice compared with WT mice, whereas serum leptin was below the limit of detection (0.1 ng/ml) in leptin-deficient ob/ob mice (Fig. 1)
FIG. 1. Serum leptin in WT, LD, and ob/ob mice. Serum was obtained from WT, LD, and ob/ob mice for measurement of leptin. Data are expressed as nanograms per milliliter and are the mean ± SEM (n = 5 mice/group). a, P < 0.01 vs. WT, by ANOVA.
Pattern of cytokine production in LD and ob/ob mice after administration of Con A
We previously reported that administration of Con A to ob/ob mice leads to a significantly blunted cytokine response compared with that in WT mice (15). Evaluation of circulating TNF, IL-4, IL-6, IL-18, IL-10, and IFN levels after Con A administration revealed a comparable pattern of cytokine production in LD and ob/ob mice compared with WT mice. As shown in Fig. 2, both LD and ob/ob mice had significantly reduced serum TNF, IL-4, IL-6, and IL-18 levels compared with WT mice. On the contrary, serum IL-10 levels were higher in both LD and ob/ob mice compared with WT mice, whereas serum IFN levels were not significantly different among the three groups. Serum cytokine levels were below detection limit in vehicle-injected mice and are thus not shown in the figure. Therefore, LD and ob/ob have a similar response to Con A in terms of cytokine production.
FIG. 2. Serum cytokine levels in WT, LD, and ob/ob mice injected with Con A. Mice received an iv injection of Con A. Serum was obtained at 2 h for measurements of TNF, IL-4, and IL-10 and at 6 h for measurements of IFN, IL-6, and IL-18. Data are expressed as nanograms per milliliter and are the mean ± SEM (n = 5 mice/group). a, P < 0.05; b, P < 0.01 (vs. WT, by ANOVA).
Characterization of liver T lymphocytes in LD mice
In ob/ob mice, resistance to liver damage induced by Con A is associated with reduced numbers of hepatic NK T cells and diminished T cell activation (19). Similarly, as shown in Fig. 3A, LD mice had a low percentage of hepatic NK T cells compared with WT mice. Administration of Con A significantly reduced the percentage of NK T cells in the liver of WT, but not LD, mice, in agreement with previous observations in ob/ob mice (19). Furthermore, expression of the activation marker CD69 on hepatic CD3+ cells 24 h after Con A injection was significantly reduced in LD mice compared with WT mice (Fig. 3B), again confirming results previously obtained in ob/ob mice (19).
FIG. 3. Hepatic immune cell populations in WT and LD mice. Mice received an iv injection of Con A. Hepatic mononuclear cells were isolated 24 h later. Data are expressed as a percentage and are the mean ± SEM (n = 5 mice/group). a, P < 0.01; b, P < 0.05 (vs. WT vehicle, by ANOVA). c, P < 0.01 (vs. LD vehicle, by ANOVA). d, P < 0.01 (vs. WT Con A, by ANOVA).
Con A-induced hepatitis in LD mice: comparison with ob/ob mice
As previously reported (23), basal ALT levels were significantly elevated in LD and ob/ob mice compared with WT mice (87.25 ± 11.57, 109.50 ± 22.31, and 37.00 ±3.51 U/liter in LD, ob/ob, and WT mice, respectively; P < 0.05 for either LD or ob/ob vs. WT; n = 5).
As shown in Fig. 4, administration of Con A induced marked hepatotoxicity, as measured by a 136-fold increase in serum ALT levels in WT mice and an 85-fold increase in LD mice, but only a 2.6-fold increase in ob/ob mice. These data confirmed that leptin-deficient ob/ob mice are resistant to hepatitis induced by Con A administration (15). In contrast and unexpectedly, LD mice, despite having extremely low leptin levels, were as susceptible to Con A as WT mice.
FIG. 4. Serum ALT levels in WT, LD, and ob/ob mice injected with Con A. Mice received an iv injection of Con A. Serum was obtained 6 and 24 h later for ALT measurement. Data are the mean ± SEM (n = 10 mice/group). a, P < 0.001; b, P < 0.05 (vs. respective vehicle, by ANOVA). c, P < 0.001 (vs. WT or LD Con A, by ANOVA).
Reduced adiponectin levels in LD mice
The results reported above indicate that the patterns of cytokine production and lymphocyte activation after administration of Con A were similar in LD and ob/ob mice. Furthermore, both LD and ob/ob mice had virtually no circulating leptin. Yet, ob/ob mice were resistant to Con A-induced liver damage, whereas LD mice were not. The primary difference between LD and ob/ob mice is the lack of WAT in the former compared with the extreme abundance of WAT in the latter. Therefore, we reasoned that adipocyte-derived factors other than leptin might contribute to the observed differences in susceptibility to Con A. To this aim, we measured circulating levels of the most abundant WAT-derived factor, adiponectin. Serum adiponectin levels were markedly reduced in LD mice compared with either WT or ob/ob mice (Fig. 5). In contrast, a minor increase in serum adiponectin levels was observed in ob/ob compared with WT mice, in agreement with previous results (28).
FIG. 5. Serum adiponectin levels in WT, LD, and ob/ob mice. Serum was obtained from WT, LD, and ob/ob mice for measurement of adiponectin. Data are expressed as micrograms per milliliter and are the mean ± SEM (n = 5 mice/group). a, P < 0.01 vs. WT; b, P < 0.01 vs. LD (by ANOVA).
Con A-induced hepatitis in WAT-depleted mice
To investigate whether other circumstances associated with reduced WAT also modulate sensitivity to Con A in a manner comparable to that of congenital lipodystrophy, a model of WAT depletion induced by chronic administration of leptin was used (26). In this study, leptin is infused until adipose tissue in completely depleted. At this point, leptin is withdrawn while the animals’ food intake is maintained at the same level as before leptin withdrawal. With this model, reduced adipose tissue and leptin levels can be sustained for several weeks in the absence of any other nutritional alterations. As indicated in Table 1, circulating leptin and adiponectin levels were both extremely low in WAT-depleted mice compared with WT mice. When WAT-depleted mice were injected with Con A, a pattern similar to that in LD mice was observed. In fact, despite reduced production of TNF, Con A induced a sharp increase in serum ALT levels in both control and WAT-depleted mice. Therefore, the simultaneous reduction of leptin and adiponectin levels led to a similar response in two different models, LD mice and WAT-depleted mice.
TABLE 1. Adipokine levels and effect of Con-A in WAT-depleted mice
Effects of adiponectin and leptin reconstitution in LD mice
To evaluate how adiponectin and leptin regulate the response to Con A, LD mice received a chronic infusion of adiponectin, leptin, or both for 6 d. As expected, a significant reduction in body weight was observed with administration of leptin, with or without concomitant administration of adiponectin (Fig. 6A). Due to the absence of WAT, in LD mice the effect of leptin on body weight was restricted to a decrease in liver weight. In fact, average liver weight decreased from 3.56 ± 0.20 g in vehicle-treated LD mice to 1.98 ± 0.20 g in LD mice treated with leptin and 2.02 ± 0.21 g in LD mice treated with both adiponectin and leptin. The ratio of liver weight/body weight was reduced from 14.63 ± 0.78 in vehicle-treated LD mice to 9.06 ± 0.51 and 8.73 ± 0.72 in leptin- and adiponectin- plus leptin-treated LD mice, respectively. Administration of adiponectin alone did not significantly affect body or liver weight. Either leptin or adiponectin was capable of improving basal ALT levels in LD mice; the combination of adiponectin and leptin did not further decrease serum ALT (Fig. 6B). These data are in agreement with previous reports indicating that leptin administration improves liver function in LD mice (29).
FIG. 6. Effect of adiponectin and leptin administration in LD mice. Lipodystrophic mice received a chronic infusion of vehicle, adiponectin (ADP), leptin (LEP), or their combination (ADP + LEP) for 6 d. Untreated wild-type mice were used as controls. A, Body weight was recorded daily. B, Serum was obtained on d 7 for measurement of ALT levels. Data are the mean ± SEM (n = 7 mice/group). a, P < 0.05; b, P < 0.01; c, P < 0.001 (vs. WT, LD vehicle, or LD ADP, by ANOVA). d, P < 0.01 (vs. WT, by ANOVA). e, P < 0.05 (vs. LD vehicle, by ANOVA).
Effects of adiponectin and leptin administration on Con A- induced hepatitis in LD mice
Administration of Con A induced a marked and comparable increase in serum ALT levels in WT and vehicle-treated LD mice (see Fig. 7A; compare with levels shown in Fig. 6B). In contrast, Con A-induced ALT levels were significantly lower in adiponectin-treated compared with vehicle-treated LD mice, indicating that replacement of adiponectin protected LD mice from the hepatotoxic effect of Con A. In contrast, administration of leptin, with or without simultaneous adiponectin treatment, led to an exacerbation of Con A-induced liver damage.
FIG. 7. Effects of adiponectin and leptin administration on Con A-induced liver damage and cytokine production in LD mice. Lipodystrophic mice received a chronic infusion of vehicle, adiponectin (ADP), leptin (LEP), or their combination (ADP + LEP) for 6 d. Untreated wild-type mice were used as controls. On d 7, mice received an iv injection of Con A. A, Serum ALT levels were measured 24 h after Con A administration. B, Serum TNF was measured 2 h after Con A administration. C, Serum IL-4 was measured 2 h after Con A administration. Data are the mean ± SEM (n = 7 mice/group). a, P < 0.01 vs. WT; b, P < 0.01 vs. LD vehicle; c, P < 0.05 vs. WT, LD vehicle, or LD ADP (by ANOVA).
High levels of TNF and IL-4 were observed in the serum of Con A-injected WT mice (Fig. 7, B and C). In LD mice, administration of adiponectin did not affect serum TNF or IL-4 levels compared with vehicle administration, whereas treatment with leptin (with or without adiponectin) restored cytokines to levels observed in WT mice (Fig. 7, B and C). Serum cytokine levels were below the detection limit in each mouse that did not receive Con A (data not shown). These data are in agreement with the previously observed effect of leptin in modulating Con A-induced cytokine production in ob/ob mice (15).
The effects of adiponectin and leptin administration on liver histology and hepatocyte cell death were also evaluated (Fig. 8). As demonstrated by hematoxylin and eosin staining, Con A induced cell infiltration and necrosis to a similar degree in WT and LD mice. TUNEL staining also indicated a similar degree of cell death in the two groups. Administration of adiponectin protected LD mice from immune cell infiltration and liver necrosis and completely prevented Con A-induced cell death. In contrast, high levels of cell infiltration, tissue necrosis, and TUNEL positivity were observed in LD mice receiving leptin, with or without adiponectin.
FIG. 8. Effect of adiponectin and leptin administration on Con A-induced alterations in hepatic histology. Lipodystrophic mice received a chronic infusion of vehicle, adiponectin (ADP), leptin (LEP), or their combination (ADP + LEP) for 6 d. Untreated WT mice were used as controls. On d 7, mice received an iv injection of Con A. The liver was removed 24 h later for histological evaluation (left panels) and TUNEL staining (right panels). One representative staining of five performed in each group is shown.
Collectively, these data demonstrate that administration of adiponectin to LD mice induced a pattern of response to Con A similar to that present in ob/ob mice, i.e. protection from Con A-induced liver damage associated with low cytokine levels. In contrast, similar to what was observed in leptin-reconstituted ob/ob mice (15), administration of leptin to LD mice led to restoration of cytokine production to levels not significantly different from those observed in WT mice.
TNF neutralization reduces Con A-induced hepatitis in LD mice
The data presented above indicate that in LD mice, Con A can induce hepatitis even when serum TNF levels are significantly reduced compared with those in WT mice. To evaluate whether Con A-induced liver damage in LD mice was still dependent upon the presence of TNF, a TNF receptor, sTNFRp55, was administered to neutralize TNF activity. As shown in Fig. 9, pretreatment with sTNFRp55 significantly protected both WT and LD mice from the increase in serum ALT levels induced by Con A, indicating that TNF is indeed a crucial cytokine for Con A-mediated hepatitis in LD mice.
FIG. 9. Neutralization of TNF activity protects both WT and LD mice from Con A-induced hepatotoxicity. WT and LD mice received an injection of sTNRp55 1 h before iv administration of Con A. Serum was obtained 24 h later for evaluation of ALT levels. Data are the mean ± SEM (n = 5 mice/group). a, P < 0.01 vs. respective vehicle, by ANOVA.
Increased susceptibility to the hepatotoxic effect of TNF in LD mice
To verify whether LD mice have an increased susceptibility to the direct hepatotoxic effects of TNF, the model of TNF- plus D-Gal-induced liver damage was used. This is a commonly used model to evaluate the hepatotoxicity of TNF. Briefly, the administration of D-Gal sensitizes the hepatocytes to TNF’s activity, allowing for the use of low doses of this cytokine to study hepatocyte death. Six hours after the administration of TNF- and D-Gal, serum ALT levels were 131.33 ± 39.22 and 708.15 ± 302.00 U/liter in WT vs. LD mice, respectively (n = 15 mice/group; P < 0.01), representing 3.5- and 8-fold increases over basal ALT levels, respectively.
Adiponectin protects hepatocytes from TNF-induced cell death
We next verified whether adiponectin could protect hepatocytes from TNF-induced cell death. Cell death was induced in cultures of WT primary murine hepatocytes by overnight incubation with TNF and actinomycin D, and the effects of adiponectin and leptin were evaluated. Adiponectin significantly (P < 0.05) reduced TNF-induced cell death when used either alone or in the presence of leptin. In fact, TNF-induced cell death was 70.65 ± 1.47% in the absence of either leptin or adiponectin, 55.87 ± 1.35% in the presence of adiponectin alone, and 57.02 ± 2.00 in the presence of both adiponectin and leptin. Leptin alone did not significantly alter TNF-induced hepatocyte death (69.83 ± 3.67%).
Discussion
In the present report we show evidence that two adipocyte-derived factors, adiponectin and leptin, act in a coordinate fashion to regulate T cell-mediated liver damage in the model of Con A administration. These data extend previous studies from our group demonstrating that leptin deficiency is associated with protection from T cell-mediated hepatitis (15, 19). The present data suggest that adiponectin modulates the sensitivity of hepatocytes to immune-mediated damage. Hence, two of the most abundant adipokines act as critical mediators in regulating hepatic inflammation.
Based on previous observation that leptin-deficient ob/ob mice are protected from Con A-induced hepatitis, we were initially puzzled by results indicating a different pattern of response in LD and WAT-depleted mice, in which leptin is barely detectable. However, a major difference exists between ob/ob and LD or WAT-depleted mice, namely, the presence of high levels of WAT-derived adiponectin in ob/ob mice compared with the virtual absence of this adipokine in the other two strains. We therefore investigated whether adiponectin could be responsible for the observed results.
Adiponectin plays a protective role in both acute and chronic inflammatory liver disease in mice (12, 13, 14). In agreement with these data, we observed that administration of adiponectin protected LD mice from Con A-induced hepatitis. However, protection was only observed in mice in which leptin levels were low. In fact, the concomitant administration of adiponectin together with leptin did not result in a protective effect, suggesting a predominant role for leptin.
Several cytokines contribute to hepatotoxicity after administration of Con A; TNF probably being the most important (30). Activated T lymphocytes as well as NK T cells are also critical factors (30, 31). In leptin-deficient ob/ob mice, protection from Con A-induced hepatitis is associated with reduced production of proinflammatory cytokine, increased levels of the protective cytokine IL-10, as well as diminished T cell activation and low numbers of hepatic NK T cells (15, 19). The cytokine and lymphocyte patterns of response to Con A were similar in ob/ob and LD mice, suggesting that leptin is a critical mediator of cytokine production and T cell activation in response to Con A in both mouse models. These data were confirmed by the observation that administration of leptin restored TNF and IL-4 levels in LD mice.
However, it was quite remarkable that in LD mice, a full-blown response to Con A in terms of hepatocyte cell death and cell infiltration occurred even in the presence of a significantly blunted cytokine response, diminished lymphocyte activation, and reduced liver NK T cells. These data suggest that adiponectin modulates the sensitivity of hepatocytes to immune-mediated cell death. TNF is the most important mediator in Con A-induced hepatocyte death (30). In LD mice, liver damage occurred even in the presence of low levels of TNF. However, neutralization of TNF protected LD mice from hepatic necrosis, suggesting that the presence of low levels of adiponectin is associated with enhanced sensitivity to this cytokine. This interpretation is also supported by the observation that administration of adiponectin protected LD mice from Con A-induced hepatitis without altering cytokine levels. Furthermore, adiponectin protected hepatocytes from TNF-induced cell death in vitro. At variance with the in vivo results reported above, under in vitro conditions, the protective effect of adiponectin on hepatocyte survival was observed even in the presence of leptin. This discrepancy is probably due to the lack of a role for leptin in the direct regulation of TNF-induced hepatocyte death in vitro compared with its prominent effect in the modulation of cytokine production in vivo. This conclusion is consistent with previous reports suggesting that leptin’s effects on the liver are indirect (32). Our data on the role of adiponectin in protecting hepatocytes from TNF-induced death are in agreement with previous reports indicating that adiponectin inhibits TNF-induced signal transduction in endothelial cells (8). Furthermore, adiponectin protects both endothelial and pancreatic ?-cells from apoptosis (9, 10). A recent report also indicates that administration of adiponectin reduces hepatotoxicity in obese KK-Ay mice treated with either endotoxin or TNF and D-Gal, again pointing to a hepatoprotective role for adiponectin (12). Additional antiinflammatory activities of adiponectin, such as direct effects on T cell activation, could also be involved.
In summary, both adiponectin and leptin contribute to the regulation of T cell-mediated hepatic inflammation, albeit through separate actions and mechanisms, leading to functionally opposite effects. Leptin increases susceptibility to Con A by maintaining high numbers of hepatic NK T cells as well as mediating T cell activation and cytokine production. Adiponectin, in contrast, protects hepatocytes from cytokine-mediated cell death. Conditions associated with high levels of leptin and low levels of adiponectin, such as the obesity and the metabolic syndrome, would therefore put the liver at a heightened risk of immune-mediated damage. A proposed model for the concerted actions of leptin and adiponectin in the Con A model is presented in Fig. 10.
FIG. 10. A proposed model for the concerted role of leptin and adiponectin in the regulation of Con A-induced hepatotoxicity. Con A stimulates T lymphocytes to produce a variety of cytokines and to interact with macrophages and Kupffer cells, which are then also induced to produce cytokines. By acting on leptin receptors present on T lymphocytes and macrophages, leptin increases the production of IL-18, IL-6, TNF, and IL-4, while reducing IL-10 levels. In the Con A model, leptin does not modulate the production of the T cell-derived cytokine, IFN. Leptin also contributes to regulating the hepatotoxic effects of Con A by modulating number of intrahepatic NK T cells as well as T cell activation. T cell- and macrophage-derived TNF is the most critical mediator of the induction of hepatocyte cell death after the administration of Con A. Adiponectin acts on hepatocytes to protect them from TNF-induced death.
In conclusion, WAT-derived factors exert crucial roles in regulating T cell-mediated inflammation. However, different adipokines appear to act in opposite directions, by either up- or down-regulating inflammatory reactions. Although this study has focused only on leptin and adiponectin, it is likely that additional WAT-derived factors act together with cytokines, adipokines, and other mediators in controlling inflammation. The final overall effect of WAT in the regulation of inflammatory reactions will probably depend on the metabolic and immune status of the organism, the type of stimulus involved in the induction of inflammation, and the organ involved in the inflammatory reaction.
References
Fruhbeck G, Gomez-Ambrosi J, Muruzabal FJ, Burrell MA 2001 The adipocyte: a model for integration of endocrine and metabolic signaling in energy metabolism regulation. Am J Physiol 280:E827–E847
Fantuzzi G, Faggioni R 2000 Leptin in the regulation of immunity, inflammation and hematopoiesis. J Leukocyte Biol 68:437–446
Rajala MW, Scherer PE 2003 The adipocyte: at the crossroads of energy homeostasis, inflammation, and atherosclerosis. Endocrinology 144:3765–3773
Beltowski J 2003 Adiponectin and resistin: new hormones of white adipose tissue. Med Sci Monit 9:RA55–RA61
Chinetti G, Zawadski C, Fruchart JC, Staels B 2004 Expression of adiponectin receptors in human macrophages and regulation by agonists of the nuclear receptors PPAR, PPAR and LXR. Biochem Biophys Res Commun 314:151–158
Wolf AM, Wolf D, Rumpold H, Enrich B, Tilg H 2004 Adiponectin induces the anti-inflammatory cytokines IL-10 and IL-1RA in human leukocytes. Biochem Biophys Res Commun 15:630–635
Ouchi N, Kihara S, Arita Y, Maeda K, Kuriyama H, Okamoto Y, Hotta K, Nishida M, Takahashi M, Nakamura T, Yamashita S, Funahashi T, Matsuzawa Y 1999 Novel modulator for endothelial adhesion molecules. Adipocyte-derived plasma protein adiponectin. Circulation 100:2473–2476
Ouchi N, Kihara S, Arita Y, Okamoto Y, Maeda K, Kuriyama H, Hotta K, Nishida M, Takahashi M, Muraguchi M, Ohmoto Y, Nakamura T, Yamashita S, Funahashi T, Matsuzawa Y 2000 Adiponectin, an adipocyte-derived plasma protein, inhibits endothelial NF-B signaling through a cAMP-dependent pathway. Circulation 102:1296–1301
Rakatzi I, Mueller H, Ritzeler O, Tennagels N, Eckel J 2004 Adiponectin counteracts cytokine- and fatty acid-induced apoptosis in the pancreatic ?-cell line INS-1. Diabetologia 47:249–258
Lin LY, Lin CY, Su TC, Liau CS 2004 Angiotensin II-induced apoptosis in human endothelial cells is inhibited by adiponectin through restoration of the association between endothelial nitric oxide synthase and heat shock protein 90. FEBS Lett 574:106–110
Kobayashi H, Ouchi N, Kihara S, Walsh K, Kumada M, Abe Y, Funahashi T, Matsuzawa Y 2004 Selective suppression of endothelial cell apoptosis by the high molecular weight form of adiponectin. Circ Res 94:e27–e31
Masaki T, Chiba S, Tatsukawa T, Noguchi H, Seike M, Yoshimatsu H 2004 Adiponectin protects LPS-induced liver injury through modulation of TNF- in KK-Ay obese mice. Hepatology 40:177–184
Xu A, Wang Y, Keshaw H, Xu LY, Lam KSL, Cooper GJS 2003 The fat-derived hormone adiponectin alleviates alcoholic and nonalcoholic fatty liver diseases in mice. J Clin Invest 112:91–100
Kamada Y, Tamura S, Kiso S, Matsumoto H, Saji Y, Yoshida Y, Fukui K, Maeda N, Nishizawa H, Nagaretani H, Okamoto Y, Kihara S, Miyagawa J, Shinomura Y, Funahashi T, Matsuzawa Y 2003 Enhanced carbon tetrachloride-induced liver fibrosis in mice lacking adiponectin. Gastroenterology 125:1796–1807
Faggioni R, Jones-Carson J, Reed DA, Dinarello CA, Feingold KR, Grunfeld C, Fantuzzi G 2000 Leptin-deficient (ob/ob) mice are protected from T cell-mediated hepatotoxicity: role of tumor necrosis factor- and IL-18. Proc Natl Acad Sci USA 97:2367–2372
Matarese G, Di Giacomo A, Sanna V, Lord GM, Howard JK, Di Tuoro A, Bloom SR, Lechler RI, Zappacosta S, Fontana S 2001 Requirement for leptin in the induction of autoimmune encephalomyelitis. J Immunol 166:5909–5916
Busso N, So A, Chobaz-Peclat V, Morard C, Martinez-Soria E, Talabot-Ayer D, Gabay C 2002 Leptin signaling deficiency impairs humoral and cellular immune responses and attenuates experimental arthritis. J Immunol 168:875–882
Siegmund B, Lehr HA, Fantuzzi G 2002 Leptin: a pivotal mediator of intestinal inflammation. Gastroenterology 122:2011–2025
Siegmund B, Lear-Kaul KC, Faggioni R, Fantuzzi G 2002 Leptin deficiency, not obesity, protects mice from Con A-induced hepatitis. Eur J Immunol 32:552–560
Gantner F, Leist M, Lohse AW, Germann PG, Tiegs G 1995 Concanavalin A-induced T-cell-mediated hepatic injury in mice: the role of tumor necrosis factor. Hepatology 21:190–198
Ksontini R, Colagiovanni DB, Josephs MD, Edwards III CK, Tannahill CL, Solorzano CC, Norman J, Denham W, Clare-Salzler M, MacKay SL, Moldawer LL 1998 Disparate roles for TNF- and Fas ligand in concanavalin A-induced hepatitis. J Immunol 160:4082–4089
Takeda K, Hayakawa Y, Van Kaer L, Matsuda H, Yagita H, Okumura K 2000 Critical contribution of liver natural killer T cells to a murine model of hepatitis. Proc Natl Acad Sci USA 97:5498–5503
Shimomura I, Hammer RE, Richardson JA, Ikemoto S, Bashmakov Y, Goldstein JL, Brown MS 1998 Insulin resistance and diabetes mellitus in transgenic mice expressing nuclear SREBP-1c in adipose tissue: model for congenital generalized lipodystrophy. Genes Dev 12:3182–3194
Shimomura I, Hammer RE, Ikemoto S, Brown MS, Goldstein JL 1999 Leptin reverses insulin resistance and diabetes mellitus in mice with congenital lipodystrophy. Nature 401:73–76
Oral EA, Simha V, Ruiz E, Andewelt A, Premkumar A, Snell P, Wagner AJ, DePaoli AM, Reitman ML, Taylor SI, Gorden P, Garg A 2002 Leptin-replacement therapy for lipodystrophy. N Engl J Med 346:570–578
Montez JM, Soukas AA, Asilmaz E, Fayzikhodjaeva G, Fantuzzi G, Friedman JE Acute leptin deficiency, leptin resistance and the physiologic response to leptin withdrawal. Proc Nat Acad Sci USA, in press
Leist M, Gantner F, Jilg S, Wendel A 1995 Activation of the 55 kDa TNF receptor is necessary and sufficient for TNF-induced liver failure, hepatocyte apoptosis, and nitrite release. J Immunol 154:1307–1316
Haluzik M, Colombo C, Gavrilova O, Chua S, Wolf N, Chen M, Stannard B, Dietz KR, Le Roith D, Reitman ML 2004 Genetic background (C57BL/6J versus FVB/N) strongly influences the severity of diabetes and insulin resistance in ob/ob mice. Endocrinology 145:3258–3264
Asilmaz E, Cohen P, Miyazaki M, Dobrzyn P, Ueki K, Fayzikhodjaeva G, Soukas AA, Kahn CR, Ntambi JM, Socci ND, Friedman JM 2004 Site and mechanism of leptin action in a rodent form of congenital lipodystrophy. J Clin Invest 113:414–424
Tiegs G 1997 Experimental hepatitis and role of cytokines. Acta Gastroenterol Belg 60:176–179
Jaruga B, Hong F, Sun R, Radaeva S, Gao B 2003 Crucial role of IL-4/STAT6 in T cell-mediated hepatitis: up-regulating eotaxins and IL-5 and recruiting leukocytes. J Immunol 171:3233–3244
Cohen P, Zhao C, Cai X, Montez JM, Rohani SC, Feinstein P, Mombaerts P, Friedman JM 2001 Selective deletion of leptin receptor in neurons leads to obesity. J Clin Invest 108:1113–1121(Joseph A. Sennello, Raja )
Address all correspondence and requests for reprints to: Dr. Giamila Fantuzzi, Department of Human Nutrition, University of Illinois, 1919 West Taylor Street, M/C 517, Chicago, Illinois 60612. E-mail: giamila@uic.edu.
Abstract
Concanavalin A-induced hepatotoxicity was compared in lipodystrophic aP2-nSREBP-1c transgenic mice (LD mice) lacking adipose tissue, obese leptin-deficient ob/ob mice, and lean wild-type (WT) mice. Serum leptin and adiponectin were low in LD mice, whereas ob/ob mice had undetectable leptin, but high adiponectin. Protection from hepatotoxicity was observed in ob/ob, but not in LD mice, despite low cytokine levels and reduced T cell activation and hepatic natural killer T cells in both groups. Administration of adiponectin protected LD mice from hepatotoxicity without altering cytokine levels. In contrast, administration of leptin heightened disease susceptibility by restoring cytokine production. Neutralization of TNF protected LD mice from liver damage. Increased in vivo susceptibility to the hepatotoxic effect of TNF was observed in LD mice. In vitro, adiponectin protected primary hepatocytes from TNF-induced death, whereas leptin had no protective effect. In conclusion, although leptin increases susceptibility to hepatotoxicity by regulating cytokine production and T cell activation, adiponectin protects hepatocytes from TNF-induced death.
Introduction
UNTIL RECENTLY, THE white adipose tissue (WAT) had been considered an inert tissue, mainly devoted to energy storage. However, with the discovery of several adipocyte-derived factors, collectively known as adipokines, WAT is currently regarded as an active player in the regulation of metabolism (1). The observations that obesity is associated with systemic inflammation and that malnutrition is accompanied by immunodeficiency have pointed to a possible role of adipokines in the regulation of immunity and inflammation (2). Leptin and adiponectin, two of the most studied adipokines, play direct regulatory roles in a variety of inflammatory conditions, both in vivo and in vitro (3).
Adiponectin is the most abundant circulating adipokine in both mice and humans (4). Several cell types, including macrophages, express adiponectin receptors (5). Adiponectin directly affects the inflammatory response by regulating both the production and the activity of cytokines (6, 7, 8) and can also act as an antiapoptotic agent in a variety of cell types (9, 10, 11). In fact, recent data indicate that adiponectin plays an antiinflammatory role in both acute and chronic inflammatory liver disease in vivo in mice (12, 13, 14).
The role of leptin in the regulation of immunity and inflammation has been well described. Leptin-deficient ob/ob mice are protected from inflammation/autoimmunity in several experimental models (15, 16, 17, 18). In particular, ob/ob mice exhibit markedly reduced disease severity in the experimental model of fulminant autoimmune hepatitis induced by administration of the T cell mitogen Concanavalin A (Con A), whereas reconstitution with exogenous leptin restores their response (15, 19). Cytokines, particularly TNF, play a crucial role in Con A-induced liver damage (20, 21). CD4+ T lymphocytes as well as natural killer (NK) T cells also contribute to the disease (22). In ob/ob mice, a reduced response to Con A-induced hepatitis is associated with reduced levels of selected proinflammatory cytokines and a lower percentage of intrahepatic NK T cells (15, 19).
Despite increasing evidence that adipocyte-derived factors modulate inflammatory reactions, most reports to date have focused on dissecting the role of single adipokines. In the present study we evaluated the role of WAT itself in regulating inflammation. We used lipodystrophic aP2-nSREBP-1c transgenic mice (LD mice), in which WAT is virtually absent due to the lack of differentiated white adipocytes (23). Insulin resistance, diabetes mellitus, as well as pronounced hepatomegaly with steatosis are present in LD mice, highlighting the pivotal role of WAT-derived factors in regulating metabolic homeostasis. In particular, administration of exogenous leptin reverses the glycemic alterations of LD mice, a therapeutic effect also observed in patients with generalized lipodystrophy (24, 25).
In the current study we used the model of Con A-induced hepatitis in LD mice to evaluate the role of WAT in regulating cytokine production and liver damage after a stimulus directed to T lymphocytes. A second experimental model, depletion of WAT by chronic administration of leptin, was used to confirm results obtained in LD mice. Reconstitution studies indicated the presence of a coordinate interplay between leptin and adiponectin in the regulation of T cell-mediated hepatic inflammation. Moreover, the role of adiponectin in protecting hepatocytes from TNF-induced cell death was evaluated both in vivo and in vitro.
Materials and Methods
Mice
Animal protocols were approved by the animal studies committee of University of Colorado Health Sciences Center and the Rockefeller University. LD mice, generated as previously described (23), were purchased from the Jackson Laboratory (Bar Harbor, ME). LD males (C57BL/6JxSJL background) were crossed to C57BL/6J females. LD females and littermate controls [wild type (WT)] generated from this cross were used. Female leptin-deficient obese ob/ob mice (B6.V-Lepob/J) and their lean littermates (WT) on a C57BL/6J background were obtained from The Jackson Laboratory. Mice of each strain were 8–12 wk old at the time of the experiment.
Administration of Con A
Con A (type IV-S, Sigma-Aldrich Corp., St. Louis, MO) was injected iv in the tail vein at a dose of 200 μg/mouse, as previously described (15). Mice were killed by cervical dislocation under isoflurane anesthesia at various time points after the administration of Con A for evaluation of cytokine and alanine aminotransferase (ALT) levels as well as liver histology. For neutralization of TNF activity, soluble TNFRp55 (Amgen, Thousands Oaks, CA) was administered ip 1 h before Con A injection at a dose of 200 μg/mouse.
Administration of TNF and D-galactosamine (D-Gal)
WT and LD mice received an ip injection of 100 ng murine recombinant TNF (R&D Systems, Minneapolis, MN) together with 20 mg D-Gal. Serum ALT levels were measured 6 h later.
Administration of leptin and adiponectin
LD mice were treated with recombinant murine leptin (5 μg/d; Amgen) and/or adiponectin (30 μg/d; Alexis Corp., San Diego, CA) in PBS for 6 d using sc placed Alzet miniosmotic pumps (Durect Corp., Palo Alto, CA). Body weight was measured daily.
Depletion of WAT
Depletion of WAT was performed as previously described (26). Alzet pumps (Durect Corp.) with an exchange rate of 0.5 μl/h were filled aseptically with either sterile PBS solution or leptin (Amgen) at a concentration of 5 μg/μl and implanted sc into C57BL/6J mice. Leptin was withdrawn by removing the pumps after 8 d of leptin treatment. At this point, mice were fed for 48 h a diet normocaloric to their caloric intake during leptin administration.
Flow cytometry
Hepatic mononuclear cells were isolated from the liver as previously described (19). Flow cytometry followed routine procedures using 1 x 105 cells/sample. Staining with annexin V and propidium iodide (BD Pharmingen, San Diego, CA) was used for evaluation of the rate of apoptosis. Surface marker expression was evaluated using anti-CD3, anti-CD69, and anti-NK1.1 antibodies from Caltag Laboratories (Burlingame, CA). Analysis was conducted on a FACSCalibur (BD Pharmingen) using the CellQuest analysis program (BD Pharmingen).
Leptin, adiponectin, cytokine, and ALT measurements
Leptin levels were measured using a mouse leptin ELISA (R&D Systems). The sensitivity of the assay is 0.1 ng/ml. Adiponectin levels were measured using an RIA (Linco Research, Inc., St. Charles, MO; sensitivity, 1 ng/ml). TNF, IL-10, and IL-18 levels were measured using an electrochemiluminescence method as previously described (18). The sensitivity of these assays is 10 pg/ml for TNF and IL-10 and 30 pg/ml for IL-18. IL-4, IL-6, and IFN levels were measured using an ELISA (BD Pharmingen). The sensitivity of these assays is 20 pg/ml. Inter- and intraassay variabilities for each of the assays used were less than 20%. Time points for cytokine measurement were chosen according to previously published data and were as follows: TNF, IL-4, and IL-10, 2 h after Con A; IL-6, IL-18, and IFN, 6 h after Con A. ALT levels were measured by the clinical laboratory of University Hospital (Denver, CO).
TUNEL assay
Formalin-fixed sections of livers obtained 24 h after Con A were processed using a TUNEL kit from Promega Corp. (Madison, WI) according to the manufacturer’s instructions.
Hepatocyte isolation and culture
Primary hepatocytes were isolated from WT mice as previously described (27). Cells were incubated with murine recombinant adiponectin (30 μg/ml) and/or leptin (100 ng/ml) for 1 h before the addition of murine recombinant TNF (100 ng/ml) and actinomycin D (20 nM) (28). Cell death was evaluated after an overnight incubation using a lactate dehydrogenase kit from Promega Corp.
Statistical analysis
Data are expressed as the mean ± SEM. The statistical significance of differences between treatment and control groups was determined by factorial ANOVA. Statistical analyses were performed using XLStat software (Addinsoft, Brooklyn, NY).
Results
Leptin levels in LD mice
LD mice lack differentiated WAT, the primary source of circulating leptin (23). As expected, serum leptin levels were extremely low in LD mice compared with WT mice, whereas serum leptin was below the limit of detection (0.1 ng/ml) in leptin-deficient ob/ob mice (Fig. 1)
FIG. 1. Serum leptin in WT, LD, and ob/ob mice. Serum was obtained from WT, LD, and ob/ob mice for measurement of leptin. Data are expressed as nanograms per milliliter and are the mean ± SEM (n = 5 mice/group). a, P < 0.01 vs. WT, by ANOVA.
Pattern of cytokine production in LD and ob/ob mice after administration of Con A
We previously reported that administration of Con A to ob/ob mice leads to a significantly blunted cytokine response compared with that in WT mice (15). Evaluation of circulating TNF, IL-4, IL-6, IL-18, IL-10, and IFN levels after Con A administration revealed a comparable pattern of cytokine production in LD and ob/ob mice compared with WT mice. As shown in Fig. 2, both LD and ob/ob mice had significantly reduced serum TNF, IL-4, IL-6, and IL-18 levels compared with WT mice. On the contrary, serum IL-10 levels were higher in both LD and ob/ob mice compared with WT mice, whereas serum IFN levels were not significantly different among the three groups. Serum cytokine levels were below detection limit in vehicle-injected mice and are thus not shown in the figure. Therefore, LD and ob/ob have a similar response to Con A in terms of cytokine production.
FIG. 2. Serum cytokine levels in WT, LD, and ob/ob mice injected with Con A. Mice received an iv injection of Con A. Serum was obtained at 2 h for measurements of TNF, IL-4, and IL-10 and at 6 h for measurements of IFN, IL-6, and IL-18. Data are expressed as nanograms per milliliter and are the mean ± SEM (n = 5 mice/group). a, P < 0.05; b, P < 0.01 (vs. WT, by ANOVA).
Characterization of liver T lymphocytes in LD mice
In ob/ob mice, resistance to liver damage induced by Con A is associated with reduced numbers of hepatic NK T cells and diminished T cell activation (19). Similarly, as shown in Fig. 3A, LD mice had a low percentage of hepatic NK T cells compared with WT mice. Administration of Con A significantly reduced the percentage of NK T cells in the liver of WT, but not LD, mice, in agreement with previous observations in ob/ob mice (19). Furthermore, expression of the activation marker CD69 on hepatic CD3+ cells 24 h after Con A injection was significantly reduced in LD mice compared with WT mice (Fig. 3B), again confirming results previously obtained in ob/ob mice (19).
FIG. 3. Hepatic immune cell populations in WT and LD mice. Mice received an iv injection of Con A. Hepatic mononuclear cells were isolated 24 h later. Data are expressed as a percentage and are the mean ± SEM (n = 5 mice/group). a, P < 0.01; b, P < 0.05 (vs. WT vehicle, by ANOVA). c, P < 0.01 (vs. LD vehicle, by ANOVA). d, P < 0.01 (vs. WT Con A, by ANOVA).
Con A-induced hepatitis in LD mice: comparison with ob/ob mice
As previously reported (23), basal ALT levels were significantly elevated in LD and ob/ob mice compared with WT mice (87.25 ± 11.57, 109.50 ± 22.31, and 37.00 ±3.51 U/liter in LD, ob/ob, and WT mice, respectively; P < 0.05 for either LD or ob/ob vs. WT; n = 5).
As shown in Fig. 4, administration of Con A induced marked hepatotoxicity, as measured by a 136-fold increase in serum ALT levels in WT mice and an 85-fold increase in LD mice, but only a 2.6-fold increase in ob/ob mice. These data confirmed that leptin-deficient ob/ob mice are resistant to hepatitis induced by Con A administration (15). In contrast and unexpectedly, LD mice, despite having extremely low leptin levels, were as susceptible to Con A as WT mice.
FIG. 4. Serum ALT levels in WT, LD, and ob/ob mice injected with Con A. Mice received an iv injection of Con A. Serum was obtained 6 and 24 h later for ALT measurement. Data are the mean ± SEM (n = 10 mice/group). a, P < 0.001; b, P < 0.05 (vs. respective vehicle, by ANOVA). c, P < 0.001 (vs. WT or LD Con A, by ANOVA).
Reduced adiponectin levels in LD mice
The results reported above indicate that the patterns of cytokine production and lymphocyte activation after administration of Con A were similar in LD and ob/ob mice. Furthermore, both LD and ob/ob mice had virtually no circulating leptin. Yet, ob/ob mice were resistant to Con A-induced liver damage, whereas LD mice were not. The primary difference between LD and ob/ob mice is the lack of WAT in the former compared with the extreme abundance of WAT in the latter. Therefore, we reasoned that adipocyte-derived factors other than leptin might contribute to the observed differences in susceptibility to Con A. To this aim, we measured circulating levels of the most abundant WAT-derived factor, adiponectin. Serum adiponectin levels were markedly reduced in LD mice compared with either WT or ob/ob mice (Fig. 5). In contrast, a minor increase in serum adiponectin levels was observed in ob/ob compared with WT mice, in agreement with previous results (28).
FIG. 5. Serum adiponectin levels in WT, LD, and ob/ob mice. Serum was obtained from WT, LD, and ob/ob mice for measurement of adiponectin. Data are expressed as micrograms per milliliter and are the mean ± SEM (n = 5 mice/group). a, P < 0.01 vs. WT; b, P < 0.01 vs. LD (by ANOVA).
Con A-induced hepatitis in WAT-depleted mice
To investigate whether other circumstances associated with reduced WAT also modulate sensitivity to Con A in a manner comparable to that of congenital lipodystrophy, a model of WAT depletion induced by chronic administration of leptin was used (26). In this study, leptin is infused until adipose tissue in completely depleted. At this point, leptin is withdrawn while the animals’ food intake is maintained at the same level as before leptin withdrawal. With this model, reduced adipose tissue and leptin levels can be sustained for several weeks in the absence of any other nutritional alterations. As indicated in Table 1, circulating leptin and adiponectin levels were both extremely low in WAT-depleted mice compared with WT mice. When WAT-depleted mice were injected with Con A, a pattern similar to that in LD mice was observed. In fact, despite reduced production of TNF, Con A induced a sharp increase in serum ALT levels in both control and WAT-depleted mice. Therefore, the simultaneous reduction of leptin and adiponectin levels led to a similar response in two different models, LD mice and WAT-depleted mice.
TABLE 1. Adipokine levels and effect of Con-A in WAT-depleted mice
Effects of adiponectin and leptin reconstitution in LD mice
To evaluate how adiponectin and leptin regulate the response to Con A, LD mice received a chronic infusion of adiponectin, leptin, or both for 6 d. As expected, a significant reduction in body weight was observed with administration of leptin, with or without concomitant administration of adiponectin (Fig. 6A). Due to the absence of WAT, in LD mice the effect of leptin on body weight was restricted to a decrease in liver weight. In fact, average liver weight decreased from 3.56 ± 0.20 g in vehicle-treated LD mice to 1.98 ± 0.20 g in LD mice treated with leptin and 2.02 ± 0.21 g in LD mice treated with both adiponectin and leptin. The ratio of liver weight/body weight was reduced from 14.63 ± 0.78 in vehicle-treated LD mice to 9.06 ± 0.51 and 8.73 ± 0.72 in leptin- and adiponectin- plus leptin-treated LD mice, respectively. Administration of adiponectin alone did not significantly affect body or liver weight. Either leptin or adiponectin was capable of improving basal ALT levels in LD mice; the combination of adiponectin and leptin did not further decrease serum ALT (Fig. 6B). These data are in agreement with previous reports indicating that leptin administration improves liver function in LD mice (29).
FIG. 6. Effect of adiponectin and leptin administration in LD mice. Lipodystrophic mice received a chronic infusion of vehicle, adiponectin (ADP), leptin (LEP), or their combination (ADP + LEP) for 6 d. Untreated wild-type mice were used as controls. A, Body weight was recorded daily. B, Serum was obtained on d 7 for measurement of ALT levels. Data are the mean ± SEM (n = 7 mice/group). a, P < 0.05; b, P < 0.01; c, P < 0.001 (vs. WT, LD vehicle, or LD ADP, by ANOVA). d, P < 0.01 (vs. WT, by ANOVA). e, P < 0.05 (vs. LD vehicle, by ANOVA).
Effects of adiponectin and leptin administration on Con A- induced hepatitis in LD mice
Administration of Con A induced a marked and comparable increase in serum ALT levels in WT and vehicle-treated LD mice (see Fig. 7A; compare with levels shown in Fig. 6B). In contrast, Con A-induced ALT levels were significantly lower in adiponectin-treated compared with vehicle-treated LD mice, indicating that replacement of adiponectin protected LD mice from the hepatotoxic effect of Con A. In contrast, administration of leptin, with or without simultaneous adiponectin treatment, led to an exacerbation of Con A-induced liver damage.
FIG. 7. Effects of adiponectin and leptin administration on Con A-induced liver damage and cytokine production in LD mice. Lipodystrophic mice received a chronic infusion of vehicle, adiponectin (ADP), leptin (LEP), or their combination (ADP + LEP) for 6 d. Untreated wild-type mice were used as controls. On d 7, mice received an iv injection of Con A. A, Serum ALT levels were measured 24 h after Con A administration. B, Serum TNF was measured 2 h after Con A administration. C, Serum IL-4 was measured 2 h after Con A administration. Data are the mean ± SEM (n = 7 mice/group). a, P < 0.01 vs. WT; b, P < 0.01 vs. LD vehicle; c, P < 0.05 vs. WT, LD vehicle, or LD ADP (by ANOVA).
High levels of TNF and IL-4 were observed in the serum of Con A-injected WT mice (Fig. 7, B and C). In LD mice, administration of adiponectin did not affect serum TNF or IL-4 levels compared with vehicle administration, whereas treatment with leptin (with or without adiponectin) restored cytokines to levels observed in WT mice (Fig. 7, B and C). Serum cytokine levels were below the detection limit in each mouse that did not receive Con A (data not shown). These data are in agreement with the previously observed effect of leptin in modulating Con A-induced cytokine production in ob/ob mice (15).
The effects of adiponectin and leptin administration on liver histology and hepatocyte cell death were also evaluated (Fig. 8). As demonstrated by hematoxylin and eosin staining, Con A induced cell infiltration and necrosis to a similar degree in WT and LD mice. TUNEL staining also indicated a similar degree of cell death in the two groups. Administration of adiponectin protected LD mice from immune cell infiltration and liver necrosis and completely prevented Con A-induced cell death. In contrast, high levels of cell infiltration, tissue necrosis, and TUNEL positivity were observed in LD mice receiving leptin, with or without adiponectin.
FIG. 8. Effect of adiponectin and leptin administration on Con A-induced alterations in hepatic histology. Lipodystrophic mice received a chronic infusion of vehicle, adiponectin (ADP), leptin (LEP), or their combination (ADP + LEP) for 6 d. Untreated WT mice were used as controls. On d 7, mice received an iv injection of Con A. The liver was removed 24 h later for histological evaluation (left panels) and TUNEL staining (right panels). One representative staining of five performed in each group is shown.
Collectively, these data demonstrate that administration of adiponectin to LD mice induced a pattern of response to Con A similar to that present in ob/ob mice, i.e. protection from Con A-induced liver damage associated with low cytokine levels. In contrast, similar to what was observed in leptin-reconstituted ob/ob mice (15), administration of leptin to LD mice led to restoration of cytokine production to levels not significantly different from those observed in WT mice.
TNF neutralization reduces Con A-induced hepatitis in LD mice
The data presented above indicate that in LD mice, Con A can induce hepatitis even when serum TNF levels are significantly reduced compared with those in WT mice. To evaluate whether Con A-induced liver damage in LD mice was still dependent upon the presence of TNF, a TNF receptor, sTNFRp55, was administered to neutralize TNF activity. As shown in Fig. 9, pretreatment with sTNFRp55 significantly protected both WT and LD mice from the increase in serum ALT levels induced by Con A, indicating that TNF is indeed a crucial cytokine for Con A-mediated hepatitis in LD mice.
FIG. 9. Neutralization of TNF activity protects both WT and LD mice from Con A-induced hepatotoxicity. WT and LD mice received an injection of sTNRp55 1 h before iv administration of Con A. Serum was obtained 24 h later for evaluation of ALT levels. Data are the mean ± SEM (n = 5 mice/group). a, P < 0.01 vs. respective vehicle, by ANOVA.
Increased susceptibility to the hepatotoxic effect of TNF in LD mice
To verify whether LD mice have an increased susceptibility to the direct hepatotoxic effects of TNF, the model of TNF- plus D-Gal-induced liver damage was used. This is a commonly used model to evaluate the hepatotoxicity of TNF. Briefly, the administration of D-Gal sensitizes the hepatocytes to TNF’s activity, allowing for the use of low doses of this cytokine to study hepatocyte death. Six hours after the administration of TNF- and D-Gal, serum ALT levels were 131.33 ± 39.22 and 708.15 ± 302.00 U/liter in WT vs. LD mice, respectively (n = 15 mice/group; P < 0.01), representing 3.5- and 8-fold increases over basal ALT levels, respectively.
Adiponectin protects hepatocytes from TNF-induced cell death
We next verified whether adiponectin could protect hepatocytes from TNF-induced cell death. Cell death was induced in cultures of WT primary murine hepatocytes by overnight incubation with TNF and actinomycin D, and the effects of adiponectin and leptin were evaluated. Adiponectin significantly (P < 0.05) reduced TNF-induced cell death when used either alone or in the presence of leptin. In fact, TNF-induced cell death was 70.65 ± 1.47% in the absence of either leptin or adiponectin, 55.87 ± 1.35% in the presence of adiponectin alone, and 57.02 ± 2.00 in the presence of both adiponectin and leptin. Leptin alone did not significantly alter TNF-induced hepatocyte death (69.83 ± 3.67%).
Discussion
In the present report we show evidence that two adipocyte-derived factors, adiponectin and leptin, act in a coordinate fashion to regulate T cell-mediated liver damage in the model of Con A administration. These data extend previous studies from our group demonstrating that leptin deficiency is associated with protection from T cell-mediated hepatitis (15, 19). The present data suggest that adiponectin modulates the sensitivity of hepatocytes to immune-mediated damage. Hence, two of the most abundant adipokines act as critical mediators in regulating hepatic inflammation.
Based on previous observation that leptin-deficient ob/ob mice are protected from Con A-induced hepatitis, we were initially puzzled by results indicating a different pattern of response in LD and WAT-depleted mice, in which leptin is barely detectable. However, a major difference exists between ob/ob and LD or WAT-depleted mice, namely, the presence of high levels of WAT-derived adiponectin in ob/ob mice compared with the virtual absence of this adipokine in the other two strains. We therefore investigated whether adiponectin could be responsible for the observed results.
Adiponectin plays a protective role in both acute and chronic inflammatory liver disease in mice (12, 13, 14). In agreement with these data, we observed that administration of adiponectin protected LD mice from Con A-induced hepatitis. However, protection was only observed in mice in which leptin levels were low. In fact, the concomitant administration of adiponectin together with leptin did not result in a protective effect, suggesting a predominant role for leptin.
Several cytokines contribute to hepatotoxicity after administration of Con A; TNF probably being the most important (30). Activated T lymphocytes as well as NK T cells are also critical factors (30, 31). In leptin-deficient ob/ob mice, protection from Con A-induced hepatitis is associated with reduced production of proinflammatory cytokine, increased levels of the protective cytokine IL-10, as well as diminished T cell activation and low numbers of hepatic NK T cells (15, 19). The cytokine and lymphocyte patterns of response to Con A were similar in ob/ob and LD mice, suggesting that leptin is a critical mediator of cytokine production and T cell activation in response to Con A in both mouse models. These data were confirmed by the observation that administration of leptin restored TNF and IL-4 levels in LD mice.
However, it was quite remarkable that in LD mice, a full-blown response to Con A in terms of hepatocyte cell death and cell infiltration occurred even in the presence of a significantly blunted cytokine response, diminished lymphocyte activation, and reduced liver NK T cells. These data suggest that adiponectin modulates the sensitivity of hepatocytes to immune-mediated cell death. TNF is the most important mediator in Con A-induced hepatocyte death (30). In LD mice, liver damage occurred even in the presence of low levels of TNF. However, neutralization of TNF protected LD mice from hepatic necrosis, suggesting that the presence of low levels of adiponectin is associated with enhanced sensitivity to this cytokine. This interpretation is also supported by the observation that administration of adiponectin protected LD mice from Con A-induced hepatitis without altering cytokine levels. Furthermore, adiponectin protected hepatocytes from TNF-induced cell death in vitro. At variance with the in vivo results reported above, under in vitro conditions, the protective effect of adiponectin on hepatocyte survival was observed even in the presence of leptin. This discrepancy is probably due to the lack of a role for leptin in the direct regulation of TNF-induced hepatocyte death in vitro compared with its prominent effect in the modulation of cytokine production in vivo. This conclusion is consistent with previous reports suggesting that leptin’s effects on the liver are indirect (32). Our data on the role of adiponectin in protecting hepatocytes from TNF-induced death are in agreement with previous reports indicating that adiponectin inhibits TNF-induced signal transduction in endothelial cells (8). Furthermore, adiponectin protects both endothelial and pancreatic ?-cells from apoptosis (9, 10). A recent report also indicates that administration of adiponectin reduces hepatotoxicity in obese KK-Ay mice treated with either endotoxin or TNF and D-Gal, again pointing to a hepatoprotective role for adiponectin (12). Additional antiinflammatory activities of adiponectin, such as direct effects on T cell activation, could also be involved.
In summary, both adiponectin and leptin contribute to the regulation of T cell-mediated hepatic inflammation, albeit through separate actions and mechanisms, leading to functionally opposite effects. Leptin increases susceptibility to Con A by maintaining high numbers of hepatic NK T cells as well as mediating T cell activation and cytokine production. Adiponectin, in contrast, protects hepatocytes from cytokine-mediated cell death. Conditions associated with high levels of leptin and low levels of adiponectin, such as the obesity and the metabolic syndrome, would therefore put the liver at a heightened risk of immune-mediated damage. A proposed model for the concerted actions of leptin and adiponectin in the Con A model is presented in Fig. 10.
FIG. 10. A proposed model for the concerted role of leptin and adiponectin in the regulation of Con A-induced hepatotoxicity. Con A stimulates T lymphocytes to produce a variety of cytokines and to interact with macrophages and Kupffer cells, which are then also induced to produce cytokines. By acting on leptin receptors present on T lymphocytes and macrophages, leptin increases the production of IL-18, IL-6, TNF, and IL-4, while reducing IL-10 levels. In the Con A model, leptin does not modulate the production of the T cell-derived cytokine, IFN. Leptin also contributes to regulating the hepatotoxic effects of Con A by modulating number of intrahepatic NK T cells as well as T cell activation. T cell- and macrophage-derived TNF is the most critical mediator of the induction of hepatocyte cell death after the administration of Con A. Adiponectin acts on hepatocytes to protect them from TNF-induced death.
In conclusion, WAT-derived factors exert crucial roles in regulating T cell-mediated inflammation. However, different adipokines appear to act in opposite directions, by either up- or down-regulating inflammatory reactions. Although this study has focused only on leptin and adiponectin, it is likely that additional WAT-derived factors act together with cytokines, adipokines, and other mediators in controlling inflammation. The final overall effect of WAT in the regulation of inflammatory reactions will probably depend on the metabolic and immune status of the organism, the type of stimulus involved in the induction of inflammation, and the organ involved in the inflammatory reaction.
References
Fruhbeck G, Gomez-Ambrosi J, Muruzabal FJ, Burrell MA 2001 The adipocyte: a model for integration of endocrine and metabolic signaling in energy metabolism regulation. Am J Physiol 280:E827–E847
Fantuzzi G, Faggioni R 2000 Leptin in the regulation of immunity, inflammation and hematopoiesis. J Leukocyte Biol 68:437–446
Rajala MW, Scherer PE 2003 The adipocyte: at the crossroads of energy homeostasis, inflammation, and atherosclerosis. Endocrinology 144:3765–3773
Beltowski J 2003 Adiponectin and resistin: new hormones of white adipose tissue. Med Sci Monit 9:RA55–RA61
Chinetti G, Zawadski C, Fruchart JC, Staels B 2004 Expression of adiponectin receptors in human macrophages and regulation by agonists of the nuclear receptors PPAR, PPAR and LXR. Biochem Biophys Res Commun 314:151–158
Wolf AM, Wolf D, Rumpold H, Enrich B, Tilg H 2004 Adiponectin induces the anti-inflammatory cytokines IL-10 and IL-1RA in human leukocytes. Biochem Biophys Res Commun 15:630–635
Ouchi N, Kihara S, Arita Y, Maeda K, Kuriyama H, Okamoto Y, Hotta K, Nishida M, Takahashi M, Nakamura T, Yamashita S, Funahashi T, Matsuzawa Y 1999 Novel modulator for endothelial adhesion molecules. Adipocyte-derived plasma protein adiponectin. Circulation 100:2473–2476
Ouchi N, Kihara S, Arita Y, Okamoto Y, Maeda K, Kuriyama H, Hotta K, Nishida M, Takahashi M, Muraguchi M, Ohmoto Y, Nakamura T, Yamashita S, Funahashi T, Matsuzawa Y 2000 Adiponectin, an adipocyte-derived plasma protein, inhibits endothelial NF-B signaling through a cAMP-dependent pathway. Circulation 102:1296–1301
Rakatzi I, Mueller H, Ritzeler O, Tennagels N, Eckel J 2004 Adiponectin counteracts cytokine- and fatty acid-induced apoptosis in the pancreatic ?-cell line INS-1. Diabetologia 47:249–258
Lin LY, Lin CY, Su TC, Liau CS 2004 Angiotensin II-induced apoptosis in human endothelial cells is inhibited by adiponectin through restoration of the association between endothelial nitric oxide synthase and heat shock protein 90. FEBS Lett 574:106–110
Kobayashi H, Ouchi N, Kihara S, Walsh K, Kumada M, Abe Y, Funahashi T, Matsuzawa Y 2004 Selective suppression of endothelial cell apoptosis by the high molecular weight form of adiponectin. Circ Res 94:e27–e31
Masaki T, Chiba S, Tatsukawa T, Noguchi H, Seike M, Yoshimatsu H 2004 Adiponectin protects LPS-induced liver injury through modulation of TNF- in KK-Ay obese mice. Hepatology 40:177–184
Xu A, Wang Y, Keshaw H, Xu LY, Lam KSL, Cooper GJS 2003 The fat-derived hormone adiponectin alleviates alcoholic and nonalcoholic fatty liver diseases in mice. J Clin Invest 112:91–100
Kamada Y, Tamura S, Kiso S, Matsumoto H, Saji Y, Yoshida Y, Fukui K, Maeda N, Nishizawa H, Nagaretani H, Okamoto Y, Kihara S, Miyagawa J, Shinomura Y, Funahashi T, Matsuzawa Y 2003 Enhanced carbon tetrachloride-induced liver fibrosis in mice lacking adiponectin. Gastroenterology 125:1796–1807
Faggioni R, Jones-Carson J, Reed DA, Dinarello CA, Feingold KR, Grunfeld C, Fantuzzi G 2000 Leptin-deficient (ob/ob) mice are protected from T cell-mediated hepatotoxicity: role of tumor necrosis factor- and IL-18. Proc Natl Acad Sci USA 97:2367–2372
Matarese G, Di Giacomo A, Sanna V, Lord GM, Howard JK, Di Tuoro A, Bloom SR, Lechler RI, Zappacosta S, Fontana S 2001 Requirement for leptin in the induction of autoimmune encephalomyelitis. J Immunol 166:5909–5916
Busso N, So A, Chobaz-Peclat V, Morard C, Martinez-Soria E, Talabot-Ayer D, Gabay C 2002 Leptin signaling deficiency impairs humoral and cellular immune responses and attenuates experimental arthritis. J Immunol 168:875–882
Siegmund B, Lehr HA, Fantuzzi G 2002 Leptin: a pivotal mediator of intestinal inflammation. Gastroenterology 122:2011–2025
Siegmund B, Lear-Kaul KC, Faggioni R, Fantuzzi G 2002 Leptin deficiency, not obesity, protects mice from Con A-induced hepatitis. Eur J Immunol 32:552–560
Gantner F, Leist M, Lohse AW, Germann PG, Tiegs G 1995 Concanavalin A-induced T-cell-mediated hepatic injury in mice: the role of tumor necrosis factor. Hepatology 21:190–198
Ksontini R, Colagiovanni DB, Josephs MD, Edwards III CK, Tannahill CL, Solorzano CC, Norman J, Denham W, Clare-Salzler M, MacKay SL, Moldawer LL 1998 Disparate roles for TNF- and Fas ligand in concanavalin A-induced hepatitis. J Immunol 160:4082–4089
Takeda K, Hayakawa Y, Van Kaer L, Matsuda H, Yagita H, Okumura K 2000 Critical contribution of liver natural killer T cells to a murine model of hepatitis. Proc Natl Acad Sci USA 97:5498–5503
Shimomura I, Hammer RE, Richardson JA, Ikemoto S, Bashmakov Y, Goldstein JL, Brown MS 1998 Insulin resistance and diabetes mellitus in transgenic mice expressing nuclear SREBP-1c in adipose tissue: model for congenital generalized lipodystrophy. Genes Dev 12:3182–3194
Shimomura I, Hammer RE, Ikemoto S, Brown MS, Goldstein JL 1999 Leptin reverses insulin resistance and diabetes mellitus in mice with congenital lipodystrophy. Nature 401:73–76
Oral EA, Simha V, Ruiz E, Andewelt A, Premkumar A, Snell P, Wagner AJ, DePaoli AM, Reitman ML, Taylor SI, Gorden P, Garg A 2002 Leptin-replacement therapy for lipodystrophy. N Engl J Med 346:570–578
Montez JM, Soukas AA, Asilmaz E, Fayzikhodjaeva G, Fantuzzi G, Friedman JE Acute leptin deficiency, leptin resistance and the physiologic response to leptin withdrawal. Proc Nat Acad Sci USA, in press
Leist M, Gantner F, Jilg S, Wendel A 1995 Activation of the 55 kDa TNF receptor is necessary and sufficient for TNF-induced liver failure, hepatocyte apoptosis, and nitrite release. J Immunol 154:1307–1316
Haluzik M, Colombo C, Gavrilova O, Chua S, Wolf N, Chen M, Stannard B, Dietz KR, Le Roith D, Reitman ML 2004 Genetic background (C57BL/6J versus FVB/N) strongly influences the severity of diabetes and insulin resistance in ob/ob mice. Endocrinology 145:3258–3264
Asilmaz E, Cohen P, Miyazaki M, Dobrzyn P, Ueki K, Fayzikhodjaeva G, Soukas AA, Kahn CR, Ntambi JM, Socci ND, Friedman JM 2004 Site and mechanism of leptin action in a rodent form of congenital lipodystrophy. J Clin Invest 113:414–424
Tiegs G 1997 Experimental hepatitis and role of cytokines. Acta Gastroenterol Belg 60:176–179
Jaruga B, Hong F, Sun R, Radaeva S, Gao B 2003 Crucial role of IL-4/STAT6 in T cell-mediated hepatitis: up-regulating eotaxins and IL-5 and recruiting leukocytes. J Immunol 171:3233–3244
Cohen P, Zhao C, Cai X, Montez JM, Rohani SC, Feinstein P, Mombaerts P, Friedman JM 2001 Selective deletion of leptin receptor in neurons leads to obesity. J Clin Invest 108:1113–1121(Joseph A. Sennello, Raja )