Direct control of hepatic glucose production by interleukin-13 in mice

Hyperglycemia is a result of impaired insulin action on glucose production and disposal, and a major target of antidiabetic therapies. The study of insulin-independent regulatory mechanisms of glucose metabolism may identify new strategies to lower blood sugar levels. Here we demonstrate an unexpected metabolic function for IL-13 in the control of hepatic glucose production. IL-13 is a Th2 cytokine known to mediate macrophage alternative activation. Genetic ablation of Il-13 in mice (Il-13^–/–) resulted in hyperglycemia, which progressed to hepatic insulin resistance and systemic metabolic dysfunction. In Il-13^–/– mice, upregulation of enzymes involved in hepatic gluconeogenesis was a primary event leading to dysregulated glucose metabolism. IL-13 inhibited transcription of gluconeogenic genes by acting directly on hepatocytes through Stat3, a noncanonical downstream effector. Consequently, the ability of IL-13 to suppress glucose production was abolished in liver cells lacking Stat3 or IL-13 receptor α1 (Il-13rα1), which suggests that the IL-13Rα1/Stat3 axis directs IL-13 signaling toward metabolic responses. These findings extend the implication of a Th1/Th2 paradigm in metabolic homeostasis beyond inflammation to direct control of glucose metabolism and suggest that the IL-13/Stat3 pathway may serve as a therapeutic target for glycemic control in insulin resistance and type 2 diabetes.

Loss of postprandial glycemic control followed by fasting hyperglycemia is a hallmark of type 2 diabetes (1, 2). The resulting glucotoxicity contributes to metabolic complications, such as pancreatic β cell death and vascular dysfunction (3). Therefore, identification of therapeutic approaches to improve glycemic management will be beneficial for diabetes and related metabolic diseases.

Glucose homeostasis is achieved through tight regulation of its production and utilization. In healthy individuals, these 2 processes are balanced to maintain blood glucose within a narrow physiological range. The liver is the main tissue for glucose production, and the regulation of hepatic gluconeogenic genes, including pyruvate carboxylase (Pcx), phosphoenolpyruvate carboxykinase (Pepck), fructose-1,6-bisphosphatase (Fbp1), and glucose-6-phosphatase (G6p) has been extensively studied. During fasting, gluconeogenic genes are induced in the liver by several transcription factors, including forkhead box protein O1 (FOXO1), PPARγ coactivator–1α (PPARGC1α), and cAMP response element–binding protein (CREB) and its coactivators, CREB-binding protein/p300 (CBP/p300) and cAMP-regulated transcriptional coactivators (CRTCs) (4, 5). Conversely, in states of abundant glucose, such as after a meal, insulin suppresses hepatic glucose production through AKT kinases that trigger nuclear exclusion of FOXO1 and CRTC2 in a phosphorylation-dependent manner, thereby relieving transcriptional activities of FOXO1 and CREB (4–6). Transcriptional repression of gluconeogenesis is less understood, although STAT3 has been implicated in this process (7, 8). Interestingly, it has been shown that downregulation of hepatic gluconeogenic genes by feeding is maintained in mice lacking Akt1/2 and FoxO1 in the liver (9), indicative of the existence of additional, uncharacterized mechanisms controlling gluconeogenesis (6).

In addition to normal physiological regulation, recent evidence suggests that glucose homeostasis is also modulated by chronic inflammation associated with metabolic stress (10, 11). Genetic models in mice have demonstrated that deletion of key inflammatory mediators improves glucose tolerance in obesity-induced insulin resistance (12). The detrimental effects of proinflammatory pathways, most notably those triggered by Th1 cytokines (e.g., IFN-γ, TNF-α, and IL-1β), on glucose homeostasis are achieved partly through inhibitory serine phosphorylation of IRS1 by JNK. In turn, uncontrolled hyperglycemia could further contribute to chronic inflammation. For example, the activity of NF-κB is increased through O-GlcNAc modification under high-glucose conditions (13). NF-κB is a primary transcription factor downstream of Th1 cytokines, and its activation in the liver promotes systemic insulin resistance (14, 15). Consequently, increased activities of Th2 cytokines (e.g., IL-4 and IL-13) skew the immune response to a Th2 phenotype (16) and restore glucose homeostasis (17–21). Th2 cytokines activate several downstream effectors, including STAT6, PPARγ, and PPARδ, to induce macrophage alternative activation, which dampens inflammation. The Th1/Th2 (or M1/M2) paradigm is thought to play a major role in the progression of white adipose tissue (WAT) inflammation and dysfunction in obesity (22, 23). However, it is unclear whether Th2 cytokines directly interact with metabolic pathways to modulate systemic glucose and lipid homeostasis.

We previously reported that appreciable amounts of IL-13, but not IL-4, were detected in the liver and WAT (17). Several sources of tissue Th2 cytokines have since been identified, including CD4⁺ T lymphocytes, eosinophils, type I NKT cells (i.e., αGalCer loaded and CDd reactive), nonimmune cells, and a new class of innate lymphoid-2 cells (ILC2 cells; including nuocytes, Ih2 cells, and natural helper cells) and multipotent progenitor type 2 (MPP^type2) cells (17, 21, 24–29). In the current study, we sought to determine the role of IL-13 in metabolic regulation. Our data revealed a previously unrecognized IL-13/STAT3 pathway in the control of hepatic glucose metabolism.

C57BL/6 Il-13^–/– mice show glucose intolerance and insulin resistance on normal chow diet. IL-13 binds to the IL-13 receptor α1 (IL-13Rα1) subunit of the type II receptor complex consisting of an IL-13Rα1/IL-4Rα dimer (30). Both IL-13Rα1 and IL-4Rα were found to be expressed in nonimmune cells, such as hepatocytes (Supplemental Figure 1A; supplemental material available online with this article; doi: 10.1172/JCI64941DS1), indicative of potential metabolic functions for IL-13. In light of this observation, we used male Il-13^–/– mice on the C57BL/6 background as a genetic model to examine the effect of IL-13 depletion on metabolic homeostasis. On normal chow diet (9% fat), C57BL/6 Il-13^–/– mice progressively gained more weight over a 9-month period than did WT animals (Figure 1A), although fat mass (determined by dual-energy X-ray absorptiometry; DEXA) was not significantly different. Metabolic cage analyses performed on 4-month-old mice, prior to body weight divergence, revealed that C57BL/6 Il-13^–/– mice had reduced oxygen consumption during both day (light cycle; fasting state) and night (dark cycle; feeding state), with no difference in food intake and physical activity compared with control mice (Figure 1B). Interestingly, Il-13 gene deletion led to increased fasting glucose levels (6-hour fast) that were detected at 4 months and peaked at 7 months of age (Figure 1C). Blood chemistries at the latter time point were further analyzed at 10 am and 10 pm without disturbance of sleep/wake cycles (corresponding to the natural fasting and feeding states, respectively), which demonstrated that glucose concentrations of C57BL/6 Il-13^–/– mice were also elevated during the fed state, when lactate production was reduced (10 pm; Figure 1D). Insulin concentrations during the fasted state (10 am) were increased. In addition, C57BL/6 Il-13^–/– mice exhibited higher circulating TG levels, but normal FFA and cholesterol levels (Figure 1E).

Figure 1

Il-13^–/– mice show increased body weight and blood glucose levels. (A) Body weight in normal chow–fed WT and Il-13^–/– mice (C57BL/6 background; n = 7–10 per genotype). Inset: body composition at 7 months of age, determined by DEXA. (B) Reduced O₂ consumption, but normal food intake and activity, as determined by metabolic cages in 4-month-old Il-13^–/– mice. Average values from the light and dark cycles are shown. (C) Increased fasting glucose concentrations in Il-13^–/– mice. (D) Il-13^–/– mice exhibited dysregulated glucose metabolism. Blood glucose, serum insulin, and serum lactate levels of 7-month-old chow-fed WT and Il-13^–/– mice were measured at 10 am (fasted state) or 10 pm (fed state). (E) TG, FFA, and cholesterol levels of WT and Il-13^–/– mice were determined at 10 am and 10 pm. Data are mean ± SEM. *P < 0.05 vs. WT.

The increase in blood glucose and insulin levels indicated that Il-13 depletion causes dysregulated glucose metabolism. In fact, glucose tolerance test (GTT) and insulin tolerance test (ITT) showed that C57BL/6 Il-13^–/– mice were glucose intolerant and had reduced insulin sensitivity compared with control mice (Figure 2, A and B). Euglycemic-hyperinsulinemic clamp was performed to further characterize the defect in glucose homeostasis. C57BL/6 Il-13^–/– mice required less exogenous glucose to maintain euglycemia (as demonstrated by a reduced glucose infusion rate), due to their increased hepatic glucose production — both at the basal level and during the clamp — and reduced glucose disposal (Figure 2C). Furthermore, glucose uptake was decreased in both muscle and WAT (Figure 2D). Tissue-specific insulin signaling was assessed by insulin-stimulated AKT phosphorylation. C57BL/6 Il-13^–/– muscle and WAT showed decreased levels of phosphorylated AKT (p-AKT) compared with control animals (Figure 2E). The strongest reduction in p-AKT was observed in C57BL/6 Il-13^–/– livers, where glycogen content and insulin-induced tyrosine phosphorylation of IRS1 were also lower, while fat accumulation was elevated (Figure 2, F and G). These observations suggest that C57BL/6 Il-13^–/– mice exhibit systemic metabolic dysregulation with severe hepatic insulin resistance.

Figure 2

Il-13^–/– mice develop hepatic insulin resistance and systemic metabolic dysfunction on normal chow. (A and B) GTT (A) and ITT (B) of 7- to 8-month-old chow-fed WT and Il-13^–/– mice (C57BL/6 background; n = 7–10 per genotype). (C) Euglycemic-hyperinsulinemic clamp demonstrated that Il-13^–/– mice were more insulin resistant than WT controls. GIR, glucose infusion rate; HGP, hepatic glucose production (basal and during clamp); GDR, glucose disposal rate. (D) Following the clamp, WT and Il-13^–/– mice were given 2-deoxy-d-[1-¹⁴C]glucose to determine tissue glucose uptake. (E) Immunoblotting of tissue insulin signaling, assessed by insulin-stimulated AKT phosphorylation. Liver, muscle, and WAT were collected following the clamp. Representative tissue samples from 4 individual mice are shown. Samples of the same tissue were run on the same gel but were noncontiguous (white lines). Quantification of immunoblot is also shown. t-AKT, total AKT. (F) Immunoblotting showed insulin-stimulated IRS1 tyrosine phosphorylation in the liver. Quantification of hepatic glycogen content is also shown. (G) Liver histology was performed to assess fat accumulation; shown are sections from 2 representative mice. Scale bar: 100 μm. Quantification of hepatic TG content is also shown. Data are mean ± SEM. *P < 0.05 vs. WT.

BALB/c Il-13^–/– mice develop hyperglycemia on high-fat diet. We next sought to determine the effect of Il-13 deficiency in the BALB/c genetic background. BALB/c mice exhibit a much stronger Th2 response (accompanied by a weak Th1 response) and are less prone to develop metabolic diseases than are C57BL/6 mice (31, 32). Chow-fed BALB/c Il-13^–/– mice had normal body weight and glucose metabolism up to 6 months of age (Supplemental Figure 1, B and C). On high-fat diet (35% fat), BALB/c Il-13^–/– mice had elevated concentrations of fasting glucose and TG, whereas weight gain and fasting insulin remained similar to those of control animals (Figure 3A and Supplemental Table 1). Elevated glucose concentrations were evident throughout the 2-hour period of both GTT and ITT (Figure 3, B and C). Pyruvate tolerance test (PTT) demonstrated that BALB/c Il-13^–/– mice also exhibited increased hepatic glucose production (Figure 3D). Consistent with the dysregulated glucose metabolism, glycogen content was decreased, whereas TG content was increased, in both liver and muscle of BALB/c Il-13^–/– mice (Figure 3, E and F). Notably, tissue insulin signaling, as assessed by insulin-stimulated p-AKT, appeared to be similar between high-fat diet–fed control and Il-13^–/– mice (Supplemental Figure 1D). Collectively, our results from Il-13^–/– mice in both genetic backgrounds implicated a critical role for IL-13 in maintaining glucose homeostasis.

Figure 3

Hyperglycemia and increased hepatic glucose production in high-fat diet–fed BALB/c Il-13^–/– mice. (A) WT and Il-13^–/– mice (BALB/c background; n = 8 per genotype) were fed high-fat diet for 6 months. Body weight and blood glucose (inset) were measured at the indicated time points. (B) GTT. (C) ITT. (D) PTT showed increased glucose production in Il-13^–/– mice. (E and F) Decreased glycogen and increased TG content in liver (E) and muscle (F) of Il-13^–/– mice. Data are mean ± SEM. *P < 0.05 vs. WT.

Il-13 regulates hepatic gluconeogenesis. Insulin is known to suppress glucose output in the liver. The increased glucose production in C57BL/6 Il-13^–/– mice (Figure 2) could be due to hepatic insulin resistance. To assess the physiological function of IL-13, pathways involved in glucose synthesis and production were examined in chow-fed C57BL/6 WT and Il-13^–/– mice at 2 months of age, when insulin responsiveness (as determined by ITT) and circulating concentrations of fasting glucose, insulin, lactate, and lipids were similar between genotypes (data not shown). Liver samples were collected at 10 am and 10 pm to examine fasting and feeding metabolic responses. Gluconeogenic gene expression at 10 am was similar between the 2 C57BL/6 genotypes. As expected, mRNA levels of hepatic gluconeogenic genes (e.g., Pcx, Fbp1, and G6p) were decreased during the fed state (10 pm) compared with the fasted state (10 am) in WT mice (Figure 4A). However, this downregulation was blunted in C57BL/6 Il-13^–/– mice, concomitant with an increase in feeding blood glucose levels (WT, 176.07 ± 2.40 mg/dl glucose, Il-13^–/–, 188.66 ± 4.57 mg/dl glucose, P < 0.05; WT, 1.62 ± 0.37 ng/ml insulin, Il-13^–/–, 1.70 ± 0.44 ng/ml insulin). Induction of de novo lipogenic genes (e.g., Srebp-1c and acetyl-CoA carboxylase 1 and 2 [Acc1/2]) by feeding was not affected, consistent with the notion that C57BL/6 Il-13^–/– mice at this age show a normal hepatic insulin response. Dysregulated glucose metabolism at the fed state worsened by 7 months of age, as demonstrated by increased expression of several gluconeogenic genes, including Pcx, Pepck, Fbp1, G6p, and Ppargc1α, in C57BL/6 Il-13^–/– mice (Figure 4B). Subsequently, recombinant IL-13 (rIL-13; 1 μg every other day, total 3 doses) was used to identify primary metabolic targets. rIL-13 treatment normalized gluconeogenic gene expression, but did not affect fatty acid metabolism genes, in C57BL/6 Il-13^–/– mice (Figure 4B). rIL-13 also restored glucose and lactate to control levels (Figure 4C), suggesting a direct role for IL-13 in the control of hepatic glucose metabolism. Indeed, primary hepatocytes isolated from C57BL/6 Il-13^–/– mice showed increased glucose production and gluconeogenic gene expression, a defect that was rescued by rIL-13 treatment for 6 hours (Figure 4D). Similar results were observed in livers/hepatocytes derived from high-fat diet–fed BALB/c Il-13^–/– mice (Supplemental Figure 2, A and B). These data suggest that IL-13 depletion results in loss of postprandial glycemic control, leading to hepatic insulin resistance and fasting hyperglycemia.

Figure 4

Dysregulation of hepatic gluconeogenic genes in Il-13^–/– mice. (A) The effect of feeding-induced suppression on gluconeogenic gene expression was blunted in Il-13^–/– mice. Livers from 2-month-old chow-fed WT and Il-13^–/– mice (C57BL/6 background; n = 6 per genotype) were collected at 10 am and 10 pm, and gene expression was examined by quantitative real-time PCR. (B) By 7 months of age, expression of the gluconeogenic pathway was further elevated in liver of chow-fed Il-13^–/– mice (C57BL/6 background), which was normalized by rIL-13 injection (1 μg every other day, total 3 doses). Livers were collected at 10 pm. (C) Acute rIL-13 treatment rescued the glucose phenotype of Il-13^–/– mice (cohort as in B). Blood glucose and lactate were measured at 10 pm. (D) Glucose production and gluconeogenic gene expression in primary hepatocytes derived from 4- to 5-month-old Il-13^–/– mice. rIL-13 (10 ng/ml) was given to hepatocytes for 2 hours followed by a 4-hour glucose production assay in the presence of rIL-13. RNA was isolated 6 hours after rIL-13 treatment. Data are mean ± SEM. *P < 0.05.

Th2 cytokines have been shown to modulate high-fat diet–induced tissue inflammation (10) and brown adipose tissue (BAT) thermogenesis (33), both of which contribute to metabolic homeostasis. On normal chow, the expression of proinflammatory (M1) and antiinflammatory (M2) markers in the liver and WAT was similar between WT and Il-13^–/– mice on the C57BL/6 background (Figure 5A). FACS and gene expression analyses showed same percentage of F4/80⁺ cells (e.g., macrophages) with similar M1/M2 expression patterns in the liver and WAT of C57BL/6 Il-13^–/– and WT mice (Supplemental Figure 3, A and B). The expression of 2 M2 genes, Mgl1 and Mrc1, was lower by 20% in F4/80⁺ cells from WAT of C57BL/6 Il-13^–/– mice, but the protein levels were not significantly altered (Supplemental Figure 3, A and B). There was also no difference in concentrations of circulating cytokines/chemokines (TNF-α, IL-1β, MCP-1, and MIP-3α) between genotypes (Supplemental Figure 3C). rIL-13 treatment was able to induce M2 genes (Figure 5A). These results were consistent with the notion that tissue inflammation is more relevant under conditions such as high-fat feeding and obesity (34–36). In fact, immune cell infiltration, typically identified by crown-like structures, was not evident based on WAT histology in these chow-fed C57BL/6 mice (Supplemental Figure 3D). In BAT, the mRNA levels of genes involved in mitochondrial oxidative metabolism and thermogenesis were not affected by IL-13 status (Figure 5B). Il-13 gene deletion in the BALB/c background on high-fat diet led to downregulation of M2 markers, particularly in WAT (Supplemental Figure 4, A and B). Notably, proinflammatory genes were unchanged. There was no difference in BAT metabolic gene expression (Supplemental Figure 4C). Collectively, these results indicate that the defect in glucose metabolism in Il-13^–/– mice under these experimental settings is mediated primarily by the function of IL-13 in hepatic gluconeogenic gene expression.

Figure 5

Minimal contribution of tissue inflammation to the metabolic phenotype of chow-fed Il-13^–/– mice. (A) Expression profiling of inflammatory genes in liver and WAT from 7-month-old chow-fed WT, Il-13^–/–, and rIL-13–injected Il-13^–/– mice (C57BL/6 background; cohort as in Figure 4B). F4/80 is a macrophage marker. M1 and M2 markers represent classic (proinflammatory) and alternative (antiinflammatory) activation of macrophages, respectively. (B) Expression profiling of oxidative metabolism and thermogenic genes in BAT using real-time PCR. Data are mean ± SEM. *P < 0.05.

Stat3 mediates the suppressive effect on gluconeogenic gene expression by Il-13. Th2 signaling is known to induce phosphorylation and nuclear translocation of STAT6 to control transcription of M2 genes in immune cells (16). However, STAT3, another member of the STAT family known to suppress gluconeogenic genes (7, 8), has also been shown to interact with IL-13Rα1 (37, 38). Additionally, feeding is thought to increase p-STAT3 (39). We found that expression of Il-13, but not Il-4, was increased in the liver from fed compared with fasted mice (Supplemental Figure 5A). As reported above, p-STAT3 was readily detectable in WT liver at the fed state (10 pm). In contrast, levels of p-STAT3 were reduced in the liver of C57BL/6 Il-13^–/– mice, which could be restored by rIL-13 treatment (Figure 6A). The p-STAT6 signal was relatively low in both genotypes, but inducible by rIL-13. To demonstrate a causal relationship between IL-13 and glucose production and identify the downstream mediators, we conducted functional studies in isolated hepatocytes using various genetic models. In primary hepatocytes, rIL-13 treatment led to rapid STAT3 phosphorylation (Figure 6B), which suggests that STAT3 may serve as an effector of IL-13 in regulating gluconeogenesis. In fact, the inhibitory effect of IL-13 on basal glucose production and gluconeogenic gene expression was abolished in Stat3^–/– hepatocytes (with the exception of Ppargc1α expression), but was preserved in WT and Stat6^–/– cells (Figure 6, C and D). In addition, rIL-13 was unable to suppress the activity of a reporter driven by a 1.7-kb PEPCK promoter element in Stat3^–/– hepatocytes (Figure 6E). Reexpression of STAT3 in these cells reduced the promoter activity by 50%, which was further decreased with rIL-13 treatment. ChIP assays further demonstrated that IL-13 increased STAT3 occupancy on Pepck and G6p gene promoters (binding sites located at approximately –1 kb and at –5 kb, respectively; Figure 6F). siRNA knockdown of Il-13rα1 (or Stat3) also abrogated the ability of IL-13 to increase STAT3 phosphorylation, reduce glucose production, and suppress gluconeogenic gene expression (Figure 6, G and H). Knockout or knockdown of Stat3, Stat6, or Il-13rα1 was shown to be specific (Supplemental Figure 5, B and C). These findings demonstrated that IL-13 controls glucose homeostasis through the IL-13Rα1/STAT3 signaling pathway in the liver.

Figure 6

IL-13 suppresses gluconeogenic gene expression through STAT3. (A) Immunoblotting showing reduced p-STAT3 levels in Il-13^–/– liver. Liver samples were collected at 10 pm (2 representative mice per group; cohort as in Figure 4B). (B) Immunoblotting showed that IL-13 induced STAT3/STAT6 phosphorylation in primary hepatocytes from WT, liver-specific Stat3^–/–, and Stat6^–/– mice. (C) IL-13 suppressed glucose production in hepatocytes in a STAT3-dependent manner. (D) IL-13 decreased expression of gluconeogenic genes through STAT3. Samples were collected 6 hours after rIL-13 treatment for gene expression analyses by quantitative real-time PCR. (E) Suppression of PEPCK promoter activity by IL-13 was mediated by STAT3. Stat3^–/– hepatocytes were transfected with a luciferase reporter driven by PEPCK promoter, STAT3 expression vector, and/or rIL-13. RLU, relative luciferase unit. (F) ChIP in WT and Stat3^–/– hepatocytes showed STAT3 occupancy on Pepck and G6p promoters induced by IL-13. ND, not detected. (G) IL-13Rα1 was required for IL-13–dependent inhibition of glucose production. Hepatocytes were isolated from WT mice and transfected with siRNAs targeting either Il-13rα1 (siIL-13Rα1) or Stat3 (siSTAT3). siControl, control siRNA. Immunoblotting of p-STAT3 and total STAT3 in hepatocytes transfected with control or IL-13Rα1 siRNA is also shown. Samples were run on the same gel but were noncontiguous (white lines). (H) IL-13Rα1 mediated IL-13–dependent inhibition on gluconeogenic gene expression, as assessed by quantitative real-time PCR. Glucose production and gene expression are presented as fold change relative to control. Data are mean ± SEM. *P < 0.05 vs. vehicle, or as indicated by brackets.

As mentioned above, many immune cells are capable of producing IL-13. To identify potential IL-13–producing cells within the liver, we first used Il-13^eGFP reporter mice, which contained a GFP knockin within the Il-13 locus (25). T cells, type I NKT cells, B cells, and macrophages were readily detected in the liver by FACS, whereas the presence of eosinophils and Lin^–ICOS⁺Sca-1⁺CD25⁺IL-7Rα⁺ nuocytes (an ILC2 cell type) was not evident (Figure 7A and data not shown). There was an increase in the percentage of CD4⁺ T cells and a decrease in CD19⁺ B cells at the fed state. However, the main IL-13⁺ population identified using Il-13^eGFP/+ mice appeared to be in CD4⁺ NKT cells (Figure 7A), at both fed (10 pm) and fasted (10 am) states. We also determined IL-13 expression in CD45⁺ liver immune cells. Similarly, a higher percentage of type I NKT cells was found to be IL-13⁺. Of note, in CD45⁺CD4^–CD1d^– (non-T, non-NKT) cells, the IL-13⁺ population from the fed state was higher than that from the fasted state (Figure 7B). In contrast, there was no difference in IL-4⁺ population (Figure 7C). These results suggest that although NKT cells represent a major source of IL-13 in the liver, certain non-T, non-NKT cells may also contribute to IL-13 release in response to feeding.

Figure 7

Assessment of IL-13–producing cells in the liver. (A) Immune cell populations in the liver collected at 10 am and 10 pm, determined by FACS. mac, macrophages. Detection of IL-13⁺ NKT cells using Il-13^eGFP is also shown. Dotted line indicates the background signal in WT BALB/c cells (n = 3–4 per time point). (B) IL-13 and (C) IL-4 expression in CD1d⁺ NKT cells, CD4⁺ T cells, and CD1d^–CD4^– cells, determined by FACS. Liver CD45⁺ immune cells were isolated from mice (n = 4 per group) that were fasted overnight or refed for 2 hours. Data are mean ± SEM. *P < 0.05.

The metabolic benefit of Th2 cytokines is believed to resolve, at least in part, the proinflammatory response elicited by Th1 signaling that has been shown to inhibit insulin sensitivity. Using experimental conditions that minimize contributions from metabolic stress–induced inflammation, the current study identified a key function for IL-13 in hepatic glucose production. Depletion of Il-13 by gene targeting resulted in hyperglycemia accompanied by elevated expression of genes required for glucose synthesis and production in the liver, while rIL-13 treatment reversed the dysregulated glucose metabolism. In the C57BL/6 background, the hyperglycemic phenotype of chow fed Il-13^–/– mice worsened with age, leading to hepatic and systemic insulin resistance. BALB/c Il-13^–/– mice also showed elevated glucose levels with high-fat diet feeding. IL-13 suppressed transcription of gluconeogenic genes through a noncanonical downstream transcription factor, STAT3, and the effect of IL-13 on inhibition of glucose production was lost in Stat3^–/– hepatocytes. These results suggest that IL-13 exhibits metabolic functions beyond its role in modulating inflammation.

In response to endotoxin, the body exhibits transient systemic insulin resistance, in an attempt to spare glucose for utilization by immune cells (40–42). It is therefore conceivable that in addition to resolving inflammation, one of the main functions of IL-13 is to restore glucose homeostasis that is disrupted by proinflammatory actions. Along this line, nutrient intake may trigger low-grade inflammation through mechanisms including O-GlcNAcylation–mediated activation of NF-κB, as described previously (13). In healthy individuals, insulin is necessary and sufficient to quickly restore glucose levels after feeding. IL-13 is likely required when insulin signaling is compromised or overwhelmed, such as in aged individuals or in cases of overnutrition. Although the glucose-lowering effect of IL-13 may help keep metabolic inflammation under control, this activity is independent of its function in macrophage polarization. Increased circulating glucose levels were observed in C57BL/6 Il-13^–/– mice on normal chow, which did not elicit strong tissue inflammation. In the BALB/c background, a strain resistant to metabolic diseases (32), hyperglycemia in Il-13^–/– mice was brought about by high-fat diet feeding without altering insulin sensitivity. In line with IL-13’s role in macrophage alternative activation, the expression of WAT M2 genes was reduced. However, proinflammatory markers were not altered. This is not unexpected, because BALB/c mice are known to have a weaker M1 response (31). Our results from these 2 mouse models indicated that the M1/M2 paradigm could be extended to direct metabolic control, in which M1 causes dysregulated glucose metabolism by inhibiting insulin signaling, while M2 (notably IL-13) maintains glucose homeostasis by suppressing hepatic glucose production.

STAT6 is known as the main effector of Th2 cytokines in immune cells (16, 43). In fact, IL-4/STAT6 signaling is essential for Th2 cell differentiation (44–46). STAT6 has also been shown to mediate certain physiological functions of IL-4, most notably adaptive thermogenesis in BAT (33). In contrast, IL-13 was not required for Th2 differentiation nor thermogenic gene expression (Figure 5B and Supplemental Figure 4C). Interestingly, IL-4 did not appear to control hepatic gluconeogenesis (Supplemental Figure 5D). It is unclear how the differential activities of Th2 cytokines are regulated, as both IL-4 and IL-13 can signal through the type II receptor IL-13Rα1/IL-4Rα heterodimer (30). Structure-based studies suggest that the relative expression of IL-4Rα and IL-13Rα1 may determine the outputs of IL-4 versus IL-13 signaling (30). Although early studies established an interaction between STAT3 and the IL-13Rα1 subunit (37, 38), Th2 cytokines have been shown to weakly activate STAT3 in immune cell lines (30, 38). The ability of IL-13 to induce STAT3 phosphorylation in primary hepatocytes appeared to be rapid and robust (Figure 6), consistent with relatively high expression of IL-13Rα1 in hepatocytes compared with adipocytes and T cells (Supplemental Figure 1A). Levels of p-STAT3 in the liver have been shown to be increased by feeding, although the upstream signaling is not clear (39). Together, these observations suggest that the IL-13/STAT3 axis directs Th2 signaling toward metabolic responses, such as feeding, whereas IL-4 and STAT6 may play roles in adaptive responses. This notion is supported by recent work that identified widely dispersed IL-13–producing cells characterized as non-T, non-B, lineage-negative ILC2 cells that did not express IL-4 (24–26, 29). These cells provide early sources of IL-13 during helminth infection, whereas T cell–derived Th2 cytokines are thought to be important for the development of adaptive immunity (47). NKT cells represent another innate cell type capable of producing IL-13. Our data suggest that type I NKT cells are a main source of IL-13 in the liver. Although this cell population is not altered by feeding, a recent study demonstrated that NKT cell activation can be modulated by dietary fats (28). In addition to NKT cells, a population of non-T, non-NKT cells isolated from livers of fed mice expressed higher IL-13 levels than did the same population in fasted animals. These cells were also low in IL-4 and thus appeared to exhibit certain characteristics of ILC2 cells. We did not detect nuocytes, but other ILC2 populations may be involved, such as Ih2 and natural helper cells expressing different markers or MPP^type2 cells (48, 49). Collectively, these observations suggest that metabolic stress is sensed by multiple innate immune cells, which may help maintain metabolic homeostasis via release of IL-13.

STAT3 has previously been shown to be recruited to promoters of G6p and Pepck in response to IL-6 (8), but the mechanism by which STAT3 actively represses transcription remains unclear. The role of IL-6 in glucose metabolism and metabolic homeostasis is complex. As a result, genetic models of Il-6 deletion have yielded conflicting metabolic phenotypes (50–53). Our current results demonstrated that IL-13 is a physiological signal that activates hepatic STAT3 to suppress glucose production. Our data also showed that depletion of IL-13 leads to hyperglycemia, which progresses to hepatic and systemic insulin resistance. Importantly, the ability of IL-13 to reduce glucose production appears to be independent of insulin signaling. Although the data presented here indicate the liver as a major site of IL-13 action (in addition to macrophages), we could not rule out the possibility that IL-13 targets other metabolic tissues. In addition, Il-13^–/– mice in both genetic backgrounds had higher hepatic fat accumulation than did control animals. IL-13 does not affect the expression of hepatic genes involved in fatty acid synthesis and catabolism. In primary hepatocytes, rIL-13 treatment reduced lipogenesis and increased fatty acid oxidation in Il-13^–/–, but not WT, cells (Supplemental Figure 5E). It is possible that the hepatic steatosis phenotype of Il-13^–/– mice is secondary to the lower metabolic rate and metabolic dysregulation. Nevertheless, the IL-13/STAT3 pathway in the liver may provide new therapeutic opportunities to control hyperglycemia associated with insulin resistance.

View Supplemental data

The authors thank S.M. Reilly and U. Ünlütürk for technical assistance, S. Akira and K. Takeda (Osaka University, Osaka, Japan) for providing floxed Stat3 mice, and G.S. Hotamisligil for help with metabolic cage and DEXA studies. P. Bhargava and K. Stanya were supported by NIH training grant T32ES016645. This work was supported by the American Heart Association, by the American Diabetes Association, and by NIH grant R01DK075046 (to C.H. Lee).

Address correspondence to: Chih-Hao Lee, Department of Genetics and Complex Diseases, Harvard School of Public Health, 665 Huntington Ave., Bldg. 1, Rm. 207, Boston, Massachusetts 02115, USA. Phone: 617.432.5778; Fax: 617.432.5236; E-mail: clee@hsph.harvard.edu.

Conflict of interest: The authors have declared that no conflict of interest exists.

Reference information: J Clin Invest. 2013;123(1):261–271. doi:10.1172/JCI64941.

Hang Shi’s present address is: Department of Biology, Georgia State University, Atlanta, Georgia, USA.

See the related article at The benefits of restraint: a pivotal role for IL-13 in hepatic glucose homeostasis.

1. Monnier L, Colette C, Dunseath GJ, Owens DR. The loss of postprandial glycemic control precedes stepwise deterioration of fasting with worsening diabetes. Diabetes Care. 2007;30(2):263–269.
   View this article via: PubMed CrossRef Google Scholar
2. Tahrani AA, Bailey CJ, Del Prato S, Barnett AH. Management of type 2 diabetes: new and future developments in treatment. Lancet. 2011;378(9786):182–197.
   View this article via: PubMed CrossRef Google Scholar
3. Chang-Chen KJ, Mullur R, Bernal-Mizrachi E. Beta-cell failure as a complication of diabetes. Rev Endocr Metab Disord. 2008;9(4):329–343.
   View this article via: PubMed CrossRef Google Scholar
4. Altarejos JY, Montminy M. CREB and the CRTC co-activators: sensors for hormonal and metabolic signals. Nat Rev Mol Cell Biol. 2011;12(3):141–151.
   View this article via: PubMed CrossRef Google Scholar
5. Lin HV, Accili D. Hormonal regulation of hepatic glucose production in health and disease. Cell Metab. 2011;14(1):9–19.
   View this article via: PubMed CrossRef Google Scholar
6. Cheng Z, White MF. The AKTion in non-canonical insulin signaling. Nat Med. 2012;18(3):351–353.
   View this article via: PubMed CrossRef Google Scholar
7. Inoue H, et al. Role of STAT-3 in regulation of hepatic gluconeogenic genes and carbohydrate metabolism in vivo. Nat Med. 2004;10(2):168–174.
   View this article via: PubMed Google Scholar
8. Ramadoss P, Unger-Smith NE, Lam FS, Hollenberg AN. STAT3 targets the regulatory regions of gluconeogenic genes in vivo. Mol Endocrinol. 2009;23(6):827–837.
   View this article via: PubMed CrossRef Google Scholar
9. Lu M, et al. Insulin regulates liver metabolism in vivo in the absence of hepatic Akt and Foxo1. Nat Med. 2012;18(3):388–395.
   View this article via: PubMed CrossRef Google Scholar
10. Bhargava P, Lee CH. Role and function of macrophages in the metabolic syndrome. Biochem J. 2012;442(2):253–262.
    View this article via: PubMed CrossRef Google Scholar
11. Hotamisligil GS. Inflammation and metabolic disorders. Nature. 2006;444(7121):860–867.
    View this article via: PubMed CrossRef Google Scholar
12. Osborn O, Olefsky JM. The cellular and signaling networks linking the immune system and metabolism in disease. Nat Med. 2012;18(3):363–374.
    View this article via: PubMed CrossRef Google Scholar
13. Issad T, Kuo M. O-GlcNAc modification of transcription factors, glucose sensing and glucotoxicity. Trends Endocrinol Metab. 2008;19(10):380–389.
    View this article via: PubMed Google Scholar
14. Arkan MC, et al. IKK-beta links inflammation to obesity-induced insulin resistance. Nat Med. 2005;11(2):191–198.
    View this article via: PubMed Google Scholar
15. Cai D, et al. Local and systemic insulin resistance resulting from hepatic activation of IKK-beta and NF-kappaB. Nat Med. 2005;11(2):183–190.
    View this article via: PubMed CrossRef Google Scholar
16. Martinez FO, Helming L, Gordon S. Alternative activation of macrophages: an immunologic functional perspective. Annu Rev Immunol. 2009;27:451–483.
    View this article via: PubMed CrossRef Google Scholar
17. Kang K, et al. Adipocyte-derived Th2 cytokines and myeloid PPARdelta regulate macrophage polarization and insulin sensitivity. Cell Metab. 2008;7(6):485–495.
    View this article via: PubMed CrossRef Google Scholar
18. Odegaard JI, et al. Macrophage-specific PPARgamma controls alternative activation and improves insulin resistance. Nature. 2007;447(7148):1116–1120.
    View this article via: PubMed CrossRef Google Scholar
19. Odegaard JI, et al. Alternative M2 activation of Kupffer cells by PPARdelta ameliorates obesity-induced insulin resistance. Cell Metab. 2008;7(6):496–507.
    View this article via: PubMed CrossRef Google Scholar
20. Ricardo-Gonzalez RR, et al. IL-4/STAT6 immune axis regulates peripheral nutrient metabolism and insulin sensitivity. Proc Natl Acad Sci U S A. 2010;107(52):22617–22622.
    View this article via: PubMed CrossRef Google Scholar
21. Wu D, et al. Eosinophils sustain adipose alternatively activated macrophages associated with glucose homeostasis. Science. 2011;332(6026):243–247.
    View this article via: PubMed CrossRef Google Scholar
22. Chawla A, Nguyen KD, Goh YP. Macrophage-mediated inflammation in metabolic disease. Nat Rev Immunol. 2011;11(11):738–749.
    View this article via: PubMed CrossRef Google Scholar
23. Sun S, Ji Y, Kersten S, Qi L. Mechanisms of inflammatory responses in obese adipose tissue. Annu Rev Nutr. 2012;32:261–286.
    View this article via: PubMed CrossRef Google Scholar
24. Moro K, et al. Innate production of T(H)2 cytokines by adipose tissue-associated c-Kit(+)Sca-1(+) lymphoid cells. Nature. 2010;463(7280):540–544.
    View this article via: PubMed CrossRef Google Scholar
25. Neill DR, et al. Nuocytes represent a new innate effector leukocyte that mediates type-2 immunity. Nature. 2010;464(7293):1367–1370.
    View this article via: PubMed CrossRef Google Scholar
26. Price AE, et al. Systemically dispersed innate IL-13-expressing cells in type 2 immunity. Proc Natl Acad Sci U S A. 2010;107(25):11489–11494.
    View this article via: PubMed CrossRef Google Scholar
27. Winer S, et al. Normalization of obesity-associated insulin resistance through immunotherapy. Nat Med. 2009;15(8):921–929.
    View this article via: PubMed CrossRef Google Scholar
28. Ji Y, et al. Activation of natural killer T cells promotes M2 Macrophage polarization in adipose tissue and improves systemic glucose tolerance via interleukin-4 (IL-4)/STAT6 protein signaling axis in obesity. J Biol Chem. 2012;287(17):13561–13571.
    View this article via: PubMed CrossRef Google Scholar
29. Saenz SA, et al. IL25 elicits a multipotent progenitor cell population that promotes T(H)2 cytokine responses. Nature. 2010;464(7293):1362–1366.
    View this article via: PubMed CrossRef Google Scholar
30. LaPorte SL, et al. Molecular and structural basis of cytokine receptor pleiotropy in the interleukin-4/13 system. Cell. 2008;132(2):259–272.
    View this article via: PubMed CrossRef Google Scholar
31. Mills CD, Kincaid K, Alt JM, Heilman MJ, Hill AM. M-1/M-2 macrophages and the Th1/Th2 paradigm. J Immunol. 2000;164(12):6166–6173.
    View this article via: PubMed Google Scholar
32. Nishikawa S, Yasoshima A, Doi K, Nakayama H, Uetsuka K. Involvement of sex, strain and age factors in high fat diet-induced obesity in C57BL/6J and BALB/cA mice. Exp Anim. 2007;56(4):263–272.
    View this article via: PubMed CrossRef Google Scholar
33. Nguyen KD, et al. Alternatively activated macrophages produce catecholamines to sustain adaptive thermogenesis. Nature. 2011;480(7375):104–108.
    View this article via: PubMed CrossRef Google Scholar
34. Lumeng CN, DelProposto JB, Westcott DJ, Saltiel AR. Phenotypic switching of adipose tissue macrophages with obesity is generated by spatiotemporal differences in macrophage subtypes. Diabetes. 2008;57(12):3239–3246.
    View this article via: PubMed CrossRef Google Scholar
35. Weisberg SP, McCann D, Desai M, Rosenbaum M, Leibel RL, Ferrante AW. Obesity is associated with macrophage accumulation in adipose tissue. J Clin Invest. 2003;112(12):1796–1808.
    View this article via: JCI PubMed Google Scholar
36. Xu H, et al. Chronic inflammation in fat plays a crucial role in the development of obesity-related insulin resistance. J Clin Invest. 2003;112(12):1821–1830.
    View this article via: JCI PubMed Google Scholar
37. Orchansky PL, Kwan R, Lee F, Schrader JW. Characterization of the cytoplasmic domain of interleukin-13 receptor-alpha. J Biol Chem. 1999;274(30):20818–20825.
    View this article via: PubMed CrossRef Google Scholar
38. Umeshita-Suyama R, et al. Characterization of IL-4 and IL-13 signals dependent on the human IL-13 receptor alpha chain 1: redundancy of requirement of tyrosine residue for STAT3 activation. Int Immunol. 2000;12(11):1499–1509.
    View this article via: PubMed CrossRef Google Scholar
39. Nie Y, et al. STAT3 inhibition of gluconeogenesis is downregulated by SirT1. Nat Cell Biol. 2009;11(4):492–500.
    View this article via: PubMed CrossRef Google Scholar
40. Agwunobi AO, Reid C, Maycock P, Little RA, Carlson GL. Insulin resistance and substrate utilization in human endotoxemia. J Clin Endocrinol Metab. 2000;85(10):3770–3778.
    View this article via: PubMed CrossRef Google Scholar
41. Virkamaki A, Yki-Jarvinen H. Mechanisms of insulin resistance during acute endotoxemia. Endocrinology. 1994;134(5):2072–2078.
    View this article via: PubMed CrossRef Google Scholar
42. Yki-Jarvinen H, Sammalkorpi K, Koivisto VA, Nikkila EA. Severity, duration, and mechanisms of insulin resistance during acute infections. J Clin Endocrinol Metab. 1989;69(2):317–323.
    View this article via: PubMed CrossRef Google Scholar
43. Gordon S. Alternative activation of macrophages. Nat Rev Immunol. 2003;3(1):23–35.
    View this article via: PubMed CrossRef Google Scholar
44. Kaplan MH, Schindler U, Smiley ST, Grusby MJ. Stat6 is required for mediating responses to IL-4 and for development of Th2 cells. Immunity. 1996;4(3):313–319.
    View this article via: PubMed CrossRef Google Scholar
45. Shimoda K, et al. Lack of IL-4-induced Th2 response and IgE class switching in mice with disrupted Stat6 gene. Nature. 1996;380(6575):630–633.
    View this article via: PubMed CrossRef Google Scholar
46. Takeda K, et al. Essential role of Stat6 in IL-4 signalling. Nature. 1996;380(6575):627–630.
    View this article via: PubMed CrossRef Google Scholar
47. Liang HE, Reinhardt RL, Bando JK, Sullivan BM, Ho IC, Locksley RM. Divergent expression patterns of IL-4 and IL-13 define unique functions in allergic immunity. Nat Immunol. 2011;13(1):58–66.
    View this article via: PubMed CrossRef Google Scholar
48. Neill DR, McKenzie AN. Nuocytes and beyond: new insights into helminth expulsion. Trends Parasitol. 2011;27(5):214–221.
    View this article via: PubMed CrossRef Google Scholar
49. Saenz SA, Noti M, Artis D. Innate immune cell populations function as initiators and effectors in Th2 cytokine responses. Trends Immunol. 2010;31(11):407–413.
    View this article via: PubMed CrossRef Google Scholar
50. Clementi AH, Gaudy AM, Zimmers TA, Koniaris LG, Mooney RA. Deletion of interleukin-6 improves pyruvate tolerance without altering hepatic insulin signaling in the leptin receptor-deficient mouse. Metabolism. 2011;60(11):1610–1619.
    View this article via: PubMed CrossRef Google Scholar
51. Di Gregorio GB, Hensley L, Lu T, Ranganathan G, Kern PA. Lipid and carbohydrate metabolism in mice with a targeted mutation in the IL-6 gene: absence of development of age-related obesity. Am J Physiol Endocrinol Metab. 2004;287(1):E182–E187.
    View this article via: PubMed CrossRef Google Scholar
52. Fritsche L, et al. IL-6 deficiency in mice neither impairs induction of metabolic genes in the liver nor affects blood glucose levels during fasting and moderately intense exercise. Diabetologia. 2010;53(8):1732–1742.
    View this article via: PubMed CrossRef Google Scholar
53. Wallenius V, et al. Interleukin-6-deficient mice develop mature-onset obesity. Nat Med. 2002;8(1):75–79.
    View this article via: PubMed CrossRef Google Scholar
54. McKenzie GJ, Bancroft A, Grencis RK, McKenzie AN. A distinct role for interleukin-13 in Th2-cell-mediated immune responses. Curr Biol. 1998;8(6):339–342.
    View this article via: PubMed CrossRef Google Scholar
55. Reilly SM, et al. Nuclear receptor corepressor SMRT regulates mitochondrial oxidative metabolism and mediates aging-related metabolic deterioration. Cell Metab. 2010;12(6):643–653.
    View this article via: PubMed CrossRef Google Scholar
56. Liu S, et al. Role of peroxisome proliferator-activated receptor {delta}/{beta} in hepatic metabolic regulation. J Biol Chem. 2011;286(2):1237–1247.
    View this article via: PubMed CrossRef Google Scholar
57. Wang X, Yang Z, Xue B, Shi H. Activation of the cholinergic antiinflammatory pathway ameliorates obesity-induced inflammation and insulin resistance. Endocrinology. 2011;152(3):836–846.
    View this article via: PubMed CrossRef Google Scholar