Kupffer cells facilitate the acute effects of leptin on hepatic lipid metabolism

Abstract

Leptin has potent effects on lipid metabolism in a number of peripheral tissues. In liver, an acute leptin infusion (~120 min) stimulates hepatic fatty acid oxidation (~30%) and reduces triglycerides (TG, ~40%), effects that are dependent on phosphoinositol-3-kinase (PI3K) activity. In the current study we addressed the hypothesis that leptin actions on liver-resident immune cells are required for these metabolic effects. Myeloid cell-specific deletion of the leptin receptor (ObR) in mice or depletion of liver Kupffer cells (KC) in rats in vivo prevented the acute effects of leptin on liver lipid metabolism, while the metabolic effects of leptin were maintained in mice lacking ObR in hepatocytes. Notably, liver TG were elevated in both lean and obese myeloid cell ObR, but the degree of obesity and insulin resistance induced by a high-fat diet was similar to control mice. In isolated primary hepatocytes (HEP), leptin had no effects on HEP lipid metabolism and only weakly stimulated PI3K. However, the coculture of KC with HEP restored leptin action on HEP fatty acid metabolism and stimulation of HEP PI3K. Notably, leptin stimulated the release from KC of a number of cytokines. However, the exposure of HEP to these cytokines individually [granulocyte macrophage colony-stimulating factor, IL-1α, IL-1β, IL-6, IL-10, and IL-18] or in combination had no effects on HEP lipid metabolism. Together, these data demonstrate a role for liver mononuclear cells in the regulation of liver lipid metabolism by leptin.

chronic administration of the adipokine leptin to models of leptin deficiency such as the ob/ob mouse, lipodystrophic rodents, and lipodystrophic humans substantially improves the extreme generalized dyslipidemia and hepatic steatosis characteristic of these states (1, 7, 10, 28, 30, 31, 35, 36). Additionally, acute (~120-min) leptin administration reduces hepatic triglycerides and stimulates fatty acid oxidation (14, 15). While the chronic effects of leptin appear predominantly to be mediated by the central actions of leptin, the acute effects of leptin result from direct action on the liver and require the activation of phospoinositol 3-kinase (PI3K) and inactivation of acetyl-CoA carboxylase (ACC) (14–16). However, the liver is a heterogeneous tissue composed of a number of cells types, with parenchymal cells (hepatocytes) and immune cells predominating. Each of these cell types express leptin receptors and are responsive to leptin (2, 8, 24, 43). Additionally, there is now a well-established role for immune cells in the regulation of metabolism, and in liver a number of studies (17, 21, 26) have demonstrated a role for liver-specific macrophages (Kupffer cells) in the regulation of hepatocyte lipid metabolism. Together, these observations suggest leptin action at immune cells may impact liver metabolic function. The purpose of this study was to address this hypothesis.

MATERIALS AND METHODS

Animal care and maintenance.

Male Wister rats were purchased from Charles River (Madison, WI). After arrival, rats were maintained on a constant 12-h light-dark cycle with free access to water, ad libitum fed with a standard chow diet, and allowed to acclimate for at least a week before any experimental interventions. C57Bl/6J mice expressing Cre recombinase driven by 2.34 kb of the mouse albumin enhancer/promoter (AlbCRE) or by the Lyzs gene locus (LysCRE) were purchased from JAX Laboratories. Mice with a floxed leptin receptor (ObR) insert (LOXObR) were a kind gift from Dr. Jeffrey Friedman, Rockefeller University, New York, NY. All procedures were approved by the Institutional Animal Care and Use Committee of the University of Pittsburgh and were in accordance with the National Research Council's Guide for the Care and Use of Laboratory Animals.

Mouse breeding and genotyping.

LOXObR mice were backcrossed eight times onto a C57Bl/6J background (~99.5% genetically identical). Mice with deletions of the hepatocyte ObR (HEPObR) or myeloid cell ObR (MYObR) were obtained by crossing LOXObR with AlbCRE or LysCRE mice, respectively. Mice were genotyped by PCR of DNA obtained from tail samples. The extent of ObR deletion was confirmed in mice after death using qRT-PCR.

In vivo experimental design.

Implantation of chronic indwelling catheters in rats and mice was performed as previously described (5, 14, 15, 32). For rat studies the animals were divided into two subgroups: one subgroup received intracarotid injections of gadolinium chloride (GdCl₃, 10 mg/kg; Sigma, St. Louis, MO), and the other received injections of vehicle (saline). GdCl₃ is a selective toxicant for Kupffer cells (KC) and is commonly used to deplete the liver of these cells (11, 13, 17, 33). In previous studies, we have demonstrated that this intervention depletes KC by ~90% (17). Thus 7 and 4 days before studies, animals received injections of a sterile filtered GdCl₃ or saline solution into the carotid artery cannula. At day 0 (4 days after the last GdCl₃ injection), animals were fasted overnight. Subsequently, one-half of the GdCl₃ and saline groups received a leptin infusion (0.2 μg·kg⁻¹·min⁻¹ preceded by a 2-min priming dose of 2.0 μg·kg⁻¹·min⁻¹), and the other half of each group received a saline infusion for 120 min as previously described (14, 15). Subsequently, animals were killed, and liver and blood were isolated for analysis as described below.

For acute mouse studies, previously (~1 wk) cannulated HEPObR and MYObR mice received infusions of leptin (0.06 μg·kg⁻¹·min⁻¹), preceded by a bolus injection of 0.1 μg, or saline for 120 min. Subsequently, animals were killed, and liver and blood were isolated for analysis as described below. For chronic mouse studies, LOXObR or MYObR mice were exposed to a high-fat/high-sucrose diet (41% fat, 27% sucrose, per calories; no. 96001; Harlan Teklad, Madison, WI) or a standard chow diet (5% fat, 10% sucrose, per calories; no. 110340; Harlan Teklad) for 10 wk. Caloric intake and weight gain were monitored throughout this period. At the end of the diet exposure, body composition was assessed using dual-energy X-ray absorptiometry, and all mice underwent an intraperitoneal glucose tolerance test (1.5 g/kg glucose). Subsequently, mice were killed, and tissues were harvested and stored at −70°C until further analysis. In separate cohorts of mice, metabolic rate was assessed using a Comprehensive Laboratory Animal Monitoring System (CLAMS), and liver immunophenotype was assessed using flow cytometry, as previously described (39).

Isolation and culture of primary hepatocytes and KC.

The method for simultaneous isolation of hepatocytes and KC from the same rat was adapted from Smedsrod and colleagues (37, 38). Briefly, livers were perfused through the portal vein in situ with an oxygenenated modified Krebs-Henseleit (KH) buffer plus 2.5 mM ethylene glycol tetra-acetic acid (EGTA) at 37°C at a rate of 18 ml/min for 10 min. Subsequently, 100 ml of KH buffer containing 0.04% collagenase (type IV; Sigma) were recirculated through the liver for ~10 min. The resulting cell suspension was filtered through a sterile 150-mesh nylon screen and centrifuged 3 times at 50 g at 4°C for 2 min, and the cell pellet containing hepatocytes was further purified in a 30% Percoll/DMEM solution (GE Healthcare Bio-Sciences, Uppsala, Sweden) by centrifugation at 50 g for 10 min. The hepatocytes were plated at a density of 1.5 × 10⁶ cells/well in collagen-coated (Sigma) 6-well plates with complete DMEM containing 10% FBS, 25 mM glucose, 10 mM HEPES, 250 U/ml penicillin, and 250 µg/ml streptomycin and incubated for 2 h at 37°C in 5% CO_2. Nonattached cells were removed by washing with PBS after 1 h. For isolation of KC, the combined supernatant from the non-Percoll centrifugations above was centrifuged at 50 g for 2 min, and the resulting supernatant (enriched in nonparenchymal cells) was centrifuged at 350 g for 5 min. The cell pellet containing nonparenchymal cells was resuspended in 4-ml complete DMEM medium and mixed with 6-ml 30% (wt/vol) histodenz (Sigma) in PBS. The resulting suspension was layered under 10-ml DMEM containing 10% fetal bovine serum and centrifuged at 1,300 g for 20 min at 22°C with the brake off. The KC-enriched fraction was obtained and washed with 4°C culture medium. Subsequently, the cell pellet was suspended in DMEM at a density of ~1.0 × 10⁶ cells/ml. Approximately 1.0 × 10⁶ cells were then plated on transwell inserts (0.4-μm pore size membrane; Corning, Lowell, MA) in 2 ml to allow attachment of KC. After 1 h the medium containing non-KC was replaced with fresh medium. Cell viability as assessed by trypan blue exclusion was >95% for KC and >90% for hepatocytes.

In vitro experimental design.

Hepatocytes were incubated in the absence or presence of KC and/or leptin (10 ng/ml) and/or BSA-complexed 0.4 mM 9,10-[³H]-palmitate (5 µCi/ml) for the measurement of fatty acid oxidation and triglyceride accumulation as previously described (17). At the end of the leptin exposure (2 h), media were collected, and cells were washed 2 times in an excess volume of ice-cold PBS, briefly treated with 12.5% trypsin-EDTA, and collected by scraping into 2 ml of PBS, centrifugation, and retention of the pellet for analysis of lipids and proteins. For all conditions, triplicate (3-well) measurements were made in each experiment. For assessment of cytokine production by KC, isolated KC were exposed to leptin (10 ng/ml) or LPS (100 ng/ml) for 2 h. During this period, media were taken, and Luminex was used to determine cytokine content. To address the effects of cytokines on hepatic lipid metabolism, fatty acid oxidation was determined in hepatocytes that were exposed to a variety of cytokines (see Fig. 4 for cytokines and concentrations used) for 2 h. For analysis of PI3K activity, cells were serum deprived for 2 h after 24-h coculture and were subsequently stimulated with 10 ng/ml leptin for 30 min. After washing 2 times with PBS containing 2-mM protease inhibitors and phosphatase inhibitors, cells were scraped into PI3K assay buffer or cell lysis buffer for subsequent analysis.

Fig. 4.

Cytokine production of Kupffer cells cultured in the absence or presence of leptin or lipopolysaccharide (A) and fatty acid oxidation of hepatocytes exposed to a panel of leptin-responsive cytokines (B). Rat Kupffer cells were cultured in the presence of leptin (10 ng/ml), LPS (100 ng/ml), or vehicle (saline) for 2 h. Subsequently, media content of the indicated cytokines was assessed. Significant differences are indicated (*vs. effects of LPS, #vs. saline control). In separate experiments, cultured hepatocytes were exposed to the indicated cytokines for 2 h, and the effect of these exposures on fatty acid oxidation was assessed. The following concentrations of cytokines were used: for single experiments, 20 ng/ml for GM-CSF, IL-1α, IL-1β, IL-6, and IL-10 and 100 ng/ml for IL-18; for combined experiments, 4 ng/ml for all cytokines except IL-18 at 20 ng/ml. The fatty oxidation data are presented as means ± SE for four independent experiments performed in triplicate.

Tissue, cell, and plasma measurements.

Triglycerides and fatty acid oxidation in cell and tissues were determined as described previously (6, 14, 15, 40). For primary hepatocytes, media ³H₂O content derived from 9,10-[³H] palmitate was assessed to determine fatty acid oxidation, as previously described (6, 40). For liver, 4 pieces of tissue weighing ~200 mg each were isolated; each piece of tissue was then further divided into 4 pieces of ~50 mg and immediately incubated with 9,10-[³H] palmitate for the measurement of fatty acid oxidation (14, 15). Insulin receptor substrate-1 (IRS-1) associated PI3K activity, and ACC was determined as previously described (14–16). Flow cytometry was performed as previously described (34, 39). All cytokines were measured using Luminex and commercially available antibodies.

Statistics.

Data are expressed as means ± SE. Statistical significance was determined by t-test, and where appropriate, analysis of variance (1-way ANOVA; Bonferroni’s post hoc test) was performed using the Systat statistical program (Evanston, IL). Statistical significance was assumed at P < 0.05.

RESULTS

Deletion of ObR in myeloid cells, but not hepatocytes, blocks the acute metabolic actions of leptin in liver in vivo.

To address the contribution of leptin action at hepatocytes and leptin action at mononuclear cells to the acute metabolic action of leptin in liver, mouse models lacking leptin receptor (ObR) expression in either hepatocytes (HEPObR) or myeloid cells (MYObR) were generated. ObR expression in myeloid cells of MYObR was reduced by >90%, and by a similar level in hepatocytes of HEPObR mice, compared with wild-type mice (data not shown). In HEPObR mice, leptin infusion (plasma [leptin] ~18 ng/ml) reduced liver triglycerides by ~40% and fatty acid oxidation ~25% compared with saline-infused control mice (Fig. 1, A and B), data that are consistent with our previously reported effects of leptin in vivo and in the perfused liver (14, 15). In contrast, a similar leptin infusion regime had no effects on liver triglycerides in MYObR mice. Furthermore, while leptin stimulated liver fatty acid oxidation and ACC phosphorylation in HEPObR mice, no such effects were observed in MYObR mice (Fig. 1, B–D). Notably, there is an increase in the basal ACC/P-ACC ratio in MYObR compared with HEPObR, although the physiological significance of this observation is unknown, as basal fatty acid oxidation was similar in both lines of mice.

Fig. 1.

Effects of ObR deletion in hepatocytes or myeloid cells on the acute metabolic actions of leptin in liver. Leptin or saline was infused into chronically catheterized HEPObR or MYObR mice. Subsequently, the liver was isolated, and the triglyceride content (A), fatty acid oxidation (B), and total and phosphorylated acetyl-CoA carboxylase (T-ACC and P-ACC, respectively; C and D) were assessed. Data are presented as means ± SE for six (saline-HEPObR), five (leptin-HEPObR), six (saline-MYObR), and seven (leptin-MYObR) animals/group analyzed individually, except for fatty acid oxidation, where n = 4 for all groups. Pal, 9,10-[³H]-palmitate; AU, arbitrary units. *Significant difference.

Gadolinium chloride-mediated depletion of KC is a specific intervention (only phagocytic immune cells are targeted) previously used by a number of groups (11, 13, 33) and our group in studies of KC regulation of hepatic metabolism (17). Thus it provides an independent model that can address the effects of the deletion of a specific immune cell type and alleviates concerns related to the possible effects of germline deletion of ObR. As previously reported (14, 15), leptin infusion (plasma [leptin] ~15 ng/ml) into rats with an intact population of KC reduced liver triglycerides (TG) and increased fatty acid oxidation by ~30 and ~40%, respectively (Fig. 2, A and B). These effects were associated with an increase in ACC phosphorylation (Fig. 2, C and D). However, deletion of KC abolished the effects of leptin (Fig. 2, A–D).

Fig. 2.

Effects of Kupffer cell depletion on the acute metabolic actions of leptin in liver. Kupffer cells were depleted (GdCl₃ +) in chronically catheterized rats by administration of GdCl₃. Control rats received saline injections (GdCl₃ −). At the completion of the treatment regime, leptin or saline was infused. Subsequently, the liver was isolated, and the triglyceride content (A), fatty acid oxidation (B), and total and phosphorylated ACC (C and D) were assessed. Data are presented as means ± SE of four animals/group analyzed individually. *Significant difference.

In primary hepatocytes, leptin stimulation of fatty acid oxidation requires the presence of KC.

To further elucidate the role of KC in the acute effects of leptin directly and in a simpler model system, we isolated primary hepatocytes (HEP) and KC from rats and performed studies whereby KC and HEP were present in the same cell culture well but physically separated or were cultured separately from each other. Subsequently, the effects of leptin on fatty acid oxidation and triglyceride levels were assessed. Leptin exposure of KC and HEP cultured separately robustly activated STAT3 in KC but not in HEP (data not shown). HEP cultured in the absence of KC exhibited no differences in lipid metabolism following exposure to leptin. However, when cocultured with KC, HEP fatty acid oxidation was increased, and TG levels were reduced (Fig. 3, A and B) in response to leptin. Our previous work established that PI3K activity is required for the acute effects of leptin on liver lipid metabolism (14, 16). Thus we next determined the effects of PI3K inhibition on KC-mediated leptin activation of HEP fatty acid oxidation. In the absence of KC, leptin weakly stimulated HEP PI3K activity. However, when cocultured with KC, HEP PI3K activity was robustly stimulated by leptin (Fig. 3C). As expected, leptin increased HEP fatty acid oxidation in the presence, but not the absence, of KC. However, in the presence of the PI3K inhibitors Wortmannin or LY290024 the effects of leptin were abolished (Fig. 3D). Thus inhibition of PI3K prevents leptin effects on lipid metabolism in primary hepatocytes.

Fig. 3.

Effects of leptin on lipid metabolism in primary hepatocytes cultured in the absence or presence of Kupffer cells. Rat Kupffer cells and hepatocytes were isolated. Hepatocytes were cultured in the absence or presence of leptin and 9,10-[³H]-palmitate and in the absence or presence of KC. Fatty acid oxidation (A) was assessed as ³H₂O production in the media, and triglyceride content (B) of hepatocytes was assessed at the conclusion of the experiment. In separate experiments, leptin stimulation of hepatocyte PI3K activity was assessed in the absence or presence of KC (C), and leptin stimulation of fatty acid oxidation was assessed in presence of PI3K inhibitors [D, Wortmannin (Wort; 100 nM) or LY290024 (LY; 10 μM)]. Data are presented as means ± SE for four to five independent experiments. *Significant difference.

Secretion of a number of Th-1 and Th-2 cytokines from KC is stimulated by leptin.

In efforts to identify potential mediators of leptin stimulation of HEP fatty acid oxidation, KC were exposed to leptin or lipolysaccaride (LPS), and the cytokine profile was assessed [granulocyte macrophage colony-stimulating factor (GM-CSF), IL-1β, IL-6, IL-10, IL-18, monocyte chemoattractant protein-1 (MCP-1), TNFα, IL-1α, IL-4, and IFNγ]. While leptin and LPS stimulated the secretion of each of these cytokines except IL-4, five (GM-CSF, IL-1α, IL-1β, IL-6, and IL-18) were stimulated to a greater extent by leptin than by LPS (Fig. 4A). Thus each of these cytokines were tested for their capacity to stimulate fatty acid oxidation in HEP. However, neither individual cytokines nor a combination of the cytokines altered HEP fatty acid oxidation (Fig. 4B), suggesting that leptin-induced increases in KC cytokine alone are insufficient to mediate the effects of leptin on HEP metabolism.

Liver triglyceride accumulation is increased by the lack of ObR expression in myeloid cells, but obesity and glucose intolerance induced by a high-fat diet are unaffected.

We next assessed the basal metabolic characteristics of MYObR mice and the effects of a high-fat diet on the development of obesity and metabolic dysregulation in these animals. Deletion of ObR from myeloid cells had little impact on body composition. Notably, caloric intake and metabolic rate were similar in MYObR compared with control floxed mice (LOXObR; data not shown). Weight gain with age was somewhat elevated in MYObR (Table 1, compare weight gain prediet vs. postdiet in standard chow-fed groups), but this difference was not accounted for by increased adiposity. Rather, lean mass and liver weights were moderately elevated. Furthermore, liver TG was elevated in lean MYObR compared with lean LOXObR (Fig. 5B). When fed a high-fat diet, weight gain in MYObR mice was similar to that in LOXObR mice (Table 1). There was a trend toward decreased adiposity in MYObR, while differences in lean body mass were maintained and became statistically significant (P < 0.01). Basal glucose tolerance (i.e., on a standard chow diet) was similar in MYObR and LOXObR, and a high-fat diet induced a similar degree of glucose intolerance (Fig. 5A). As in lean cohorts, liver TG levels were elevated in obese MYObR mice compared with LOXObR mice (Fig. 5B). Finally, given the association between liver inflammation and metabolic abnormalities, including steatosis, flow cytometry was used to assess the liver for myeloid cells expressing the CD11b and CD11c markers (predominantly macrophages and dendritic cells), which are elevated in obesity (39). No differences in myeloid cell content or activity (as assessed by the expression of the activation marker CD86) was apparent in obese MYObR compared with LOXObR mice (Fig. 6).

Table 1.

Physical characteristics of lean and obese MYObR and LOXObR mice

	Lean			Obese
	LOX	MY	P	LOX	MY	P
Body weight prediet, g	29.9	31.1	0.26	28.8	29.2	0.75
Body weight postdiet, g	33.2	35.8	0.08	44.3	44.8	0.87
Weight gain, g	3.3	4.7	0.01	15.5	15.6	0.95
Fat mass, g	6.8	6.8	0.97	20.3	17.8	0.33
Fat mass, %body wt	22.1	20.7	0.48	46.8	41.3	0.09
Lean mass, g	23.9	26.0	0.07	22.5	25.2	0.01
Lean mass, %body wt	77.9	79.3	0.48	53.2	58.7	0.09
Liver weight, g	1.35	1.51	0.02	1.69	1.67	0.95

LOX, LOXObR mice; MY, MYObR mice.

Fig. 5.

Effects of myeloid cell ObR depletion on high-fat diet-induced obesity, liver triglycerides, and glucose tolerance. MYObR and LOXObR mice were exposed to an obesogenic high-fat/high-sucrose diet for 10 wk (M-Obese and L-Obese, respectively). Control mice were exposed to a standard chow diet (M-Lean and L-Lean). During this period, weight gain, caloric intake, and metabolic rate were assessed. Subsequently, all mice underwent an intraperitoneal glucose tolerance test (A). After 1 wk of recovery, mice were killed, and liver triglycerides were measured (B). Data are presented as means ± SE for six animals/group. *Significant difference (vs. corresponding LOXObR).

Fig. 6.

A–F: effects of myeloid cell ObR depletion on obesity-induced liver immune cell infiltration. Liver immune cells were isolated from the liver of obese MYObR or obese LOXObR mice, and the indicated cell populations were analyzed by flow cytometry. Data are presented as means ± SE for four animals/group. DC, dendritic cells.

DISCUSSION

The purpose of the current study was to address the potential role of immune cell leptin action in mediating leptin effects on lipid metabolism in the liver. The major findings demonstrate that 1) deletion of ObR in myeloid cells, but not hepatocytes, prevents the acute metabolic actions of leptin on lipid metabolism in liver, 2) liver KC depletion reproduces the effects of myeloid cell ObR deletion, 3) leptin stimulation of fatty acid oxidation in primary hepatocytes requires the presence of KC, 4) liver lipid accumulation is increased in lean and obese mice lacking ObR expression in myeloid cells, and 5) deletion of ObR from myeloid cells has only a minor impact on body composition and no effects on systemic insulin action.

We have previously reported that acute delivery of leptin, either infused peripherally in vivo or perfused through the isolated liver, at physiological concentrations, substantially reduces liver triglyceride levels and stimulates fatty acid oxidation. In tandem with these observations, we demonstrated that these effects are dependent on leptin stimulation of hepatic PI3K and are associated with inactivation of acetyl-CoA carboxylase, which would be expected to stimulate fatty acid oxidation by reducing malonyl-CoA levels. The current work extends these findings to demonstrate that the effects on hepatocyte lipid metabolism are indirect; that is, they are mediated by leptin action on resident liver macrophages. These observations fit well with the well-established literature on immune cell-mediated alterations in metabolism in a variety of tissues (9, 25, 27, 29). In liver, the prevention of the inflammatory response to obesity (34), the Th-2 polarization of macrophages (21, 26), and the deletion of liver KC (17) each prevent or reduce liver steatosis and insulin resistance in the context of obesity. Thus the obesity-induced Th-1 inflammatory response in liver, and the Th-1 activation of KC, are proposed to contribute to liver steatosis and insulin resistance, while Th-2 activation is considered protective against these metabolic abnormalities.

An inference to be drawn from our metabolic observations is that leptin acutely alters immune system function. Indeed, there is an extensive literature supporting direct leptin effects on the immune system, acting through the long form of the leptin receptor, ObRb (2, 4, 8, 12, 24). However, the reported effects are complex, since both anti-inflammatory (Th-2) and proinflammatory (Th-1) leptin influences have been reported (3, 18, 22, 23, 42). Thus leptin dampens the proinflammatory response when the innate immune system is activated, including in models of acute inflammation (3). Conversely, a number of studies have demonstrated a proinflammatory role for leptin in the adaptive and innate immune systems (42). Paradoxically, while leptin protects against TNFα-induced toxicity in ob/ob and db/db mice (41), leptin also stimulates TNFα and IFNγ production in monocytes and macrophages (8). Indeed, in the current study, leptin stimulated the production of both of these cytokines, in addition to a number of other prototypical Th-1 and Th-2 cytokines.

While there is much evidence supporting the capacity of macrophages to alter metabolism in tissues such as liver, adipose tissue, and skeletal muscle, the mechanisms of these effects have remained elusive. There is strong evidence that monocyte-derived TNFα induces insulin resistance, reduces fatty acid oxidation, and induces lipid accumulation, while TNFα deletion is protective against the development of insulin resistance. In liver, KC TNFα production is increased in response to a high-fat diet and is a prime candidate for mediating deleterious effects of KC on hepatocyte metabolism (17). However, in the current study, leptin was as effective as LPS in stimulating KC TNFα production (data not shown). Furthermore, the overall effects of leptin stimulation of KC cytokine production were in many respects similar to that observed when KC were stimulated with LPS. The complexity of the cytokine response, in combination with the similarity to the LPS response, poses challenges when attempting to identify the possible mediators of KC effects on hepatocyte fatty acid oxidation. In a number of cases (GM-CSF, IL-1α, IL-1β, IL-6, IL-10, and IL-18) the effects of leptin on cytokine production were greater than the effects of LPS, and two of these (IL-6 and IL-10) have been reported to stimulate fatty acid oxidation. However, none of these cytokines alone, or in combination, increased hepatocyte fatty acid oxidation in our hands. These data indicate that the metabolic effects of leptin on hepatocytes achieved through leptin action on myeloid cells are mediated by an as yet unidentified mechanism. Kupffer cells can produce numerous soluble factors in response to their activation, which then act in a paracrine fashion on liver cells in their vicinity. As addressed in this study, cytokines and chemokines are major players in this regard. However, KC also produce reactive oxygen species (e.g., hydrogen peroxide, superoxide anion, and nitric oxide), physiologically active lipid metabolites such as the prostanoids (e.g., prostaglandins and thromboxane, which are derived from the cyclooxygenase pathway) and fatty acid hydroperoxides (e.g., leukotrienes, which are derived from the lipoxygenase pathway), and growth factors (in the context of liver regeneration).

The outstanding pathophysiological phenotype arising from a deficit of leptin action, either through the loss of functional ObR [e.g., db/db mouse and Zucker diabetic fatty (ZDF) rat] or through a lack of leptin expression (e.g., ob/ob mouse), is hyperphagia-induced morbid obesity, accompanied by profound insulin resistance and dyslipidemia. A number of studies have demonstrated that this phenotype can be reproduced solely through the deletion of ObR expression in the CNS, where peripheral leptin action is presumably maintained intact, which categorically demonstrates the primacy of central leptin action in the regulation of appetite and body weight. Therefore it is not surprising that the deletion of ObR expression solely from myeloid cells had little or no impact on body composition or insulin sensitivity or that the physiological response to an obesogenic high-fat/high-sucrose diet (weight gain, fat accumulation, and development of insulin resistance) is broadly similar to that observed in wild-type mice. Given that, the physiological differences observed, while subtle, deserve comment. Thus the increased lean body mass in MYObR is notable, while differences in weight gain with age in MYObR, albeit subtle, speak to there being effects of altered immune system activity on body composition. However, for now these observations are descriptive, as mechanisms have not been identified, and speculative, since the effects of myeloid ObR deletion on overall immune system function are unknown. The data from the analysis of the lean and obese MYObR mice that are of most relevance to the acute component of the current study are the elevation in liver triglycerides in MYObR mice compared with controls. These data lend further weight to the hypothesis that leptin action at immune cells influences liver lipid homeostasis, which is further supported by observations that leptin administration to nonobese leptin-sensitive models (e.g., lean or lipodystrophic mice or humans) is invariably associated with lowering of tissue and circulating lipid levels. The data also could speak to a contribution of leptin resistance to steatosis/nonalcoholic fatty liver disease (NAFLD), i.e., by potentially increasing liver triglyceride content through loss of leptin action. However, it should be noted that other studies have implicated liver leptin action in the development of nonalcoholic steatohepatitis (19, 20), a condition that is preceded by NAFLD, which would argue against a role for hepatic leptin resistance in NAFLD.

GRANTS

This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grants 058855 and 072162 (both to R. M. O'Doherty).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

A.M., W.H., N.D., and I.S. performed experiments; A.M., W.H., M.S.-R., N.D., I.S., and R.M.O. analyzed data; A.M., W.H., M.S.-R., and R.M.O. interpreted results of experiments; A.M., W.H., and R.M.O. prepared figures; A.M., W.H., and R.M.O. drafted manuscript; A.M., W.H., M.S.-R., and R.M.O. edited and revised manuscript; A.M., W.H., and R.M.O. approved final version of manuscript.