• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Apr 26, 2005; 102(17): 6086–6091.
Published online Apr 18, 2005. doi:  10.1073/pnas.0408452102
PMCID: PMC1087912
Medical Sciences

Angiopoietin-like protein 4 decreases blood glucose and improves glucose tolerance but induces hyperlipidemia and hepatic steatosis in mice


Angiopoietin-like protein 4 (ANGPTL4) is a circulating protein predominantly expressed in adipose tissue and liver. Several recent studies demonstrated that ANGPTL4 is the target gene of peroxisome proliferation activators, the agonists of which are widely used as the antidiabetic and lipid-lowering drugs. Here we provide evidence that ANGPTL4 is a blood-borne hormone directly involved in regulating glucose homeostasis, lipid metabolism, and insulin sensitivity. Adenovirus-mediated expression of ANGPTL4 potently decreased blood glucose and improved glucose tolerance, whereas it induced hyperlipidemia, fatty liver, and hepatomegaly in C57 mice. In db/db diabetic mice, ANGPTL4 treatment reduced hyperglycemia to a normal level, and markedly alleviated glucose intolerance and hyperinsulinemia. Ex vivo studies on primary rat hepatocytes revealed that ANGPTL4 significantly decreased hepatic glucose production and enhanced insulin-mediated inhibition of gluconeogenesis. Serum levels of ANGPTL4 in human subjects inversely correlated with plasma glucose concentrations and HOMA IR, the homeostasis model assessment of insulin resistance. In patients with type 2 diabetes, serum levels of ANGPTL4 were significantly lower than those in healthy subjects, suggesting that the decreased ANGPTL4 could be a causative factor of this disease. These results collectively indicate that ANGPTL4 exerts distinct effects on glucose and lipid metabolism, and that its beneficial effect on glucose homeostasis might be useful for the treatment of diabetes.

Keywords: adipokine, diabetes, fatty liver, metabolism

Adipose tissue is now recognized to be an important endocrine organ that secretes a variety of bioactive peptides, known as adipokines (or adipocytokines). Growing evidence suggests that adipokines are critically involved in regulating energy metabolism, systemic insulin sensitivity, cardiovascular tone, and immune response (1, 2). Several adipokines, such as TNF-α, resistin, and IL6, play causative roles in the pathogenesis of insulin resistance, type 2 diabetes, and thrombotic diseases (1). On the other hand, leptin and adiponectin possess many beneficial functions on energy metabolism and insulin sensitivity. Leptin has long been viewed as an antiobesity hormone (3), whereas adiponectin is an insulin-sensitizing adipokine with direct antidiabetic, antiatherogenic, and antiinflammatory functions (4).

Angiopoietin-like protein 4 (ANGPTL4), also known as peroxisome proliferator-activated receptor γ (PPARγ) angiopoietinrelated protein, fasting-induced adipose factor, or hepatic fibrinogen/angiopoietin-r elated protein, is a recently identified adipokine that is predominantly expressed in adipose tissue and liver (5-7). Mouse ANGPTL4 is composed of an NH2-terminal coiled-coil domain and a carboxyl fibronectin-like motif, a structural organization conserved in both angiopoietins and angiopoietin-like proteins (5). ANGPTL4 was originally identified as the target gene of PPAR (5, 6). The agonists of both PPARγ and PPARα could enhance ANGPTL4 expression and also elevate the circulating levels of this protein in human subjects and rodents (8). In addition, the expression of ANGPTL4 is under nutritional control, with its plasma concentration being increased by fasting and decreased by high fat feeding (6).

The metabolic functions of ANGPTL4 are still poorly understood. It has recently been shown that ANGPTL4 treatment acutely increases plasma triglycerides in mice, suggesting that it is a modulator of lipid metabolism (9, 10). However, these studies showed that the effect of ANGPTL4 on plasma triglycerides is transient, and the long-term effects of this protein on energy metabolism and insulin sensitivity remain to be established.

In this study, we used an adenovirus-mediated expression system to investigate the metabolic effects of ANGPTL4 in mice. Our results demonstrated that ANGPTL4 is an important regulator of glucose homeostasis, lipid metabolism, and insulin sensitivity. In both C57 mice and db/db diabetic mice, ANGPTL4 markedly improved glucose tolerance and decreased blood glucose, possibly by inhibition of hepatic glucose production. Moreover, our clinical study demonstrated that serum ANGPTL4 levels are inversely correlated with plasma glucose concentrations in human subjects, and are significantly decreased in patients with type 2 diabetes mellitus (T2DM).


Animals. Male C57BL/6J and C57BKS db/db diabetic mice (The Jackson Laboratory) between 8 and 10 weeks old were used for this study. The mice were housed in a room under controlled temperature (23 ± 1°C), with free access to water and standard mouse chow. All of the experiments were conducted under our institutional guidelines for the humane treatment of laboratory animals.

Human Clinical Study Protocol. A total of 42 lean healthy subjects, 46 patients with T2DM, and 22 obese individuals without T2DM were recruited for this study. The clinical characteristics of the subject are given in Table 1, which is published as supporting information on the PNAS web site. Fasting (12 h overnight) blood was taken for measurement of fasting plasma glucose (FPG), insulin, and total cholesterol and triglyceride as described (11). The homeostasis model assessment of insulin resistance (HOMA IR), a simple assessment of insulin sensitivity, was calculated by using the following formula: FPG (mmol/liter) × fasting insulin (microunits/ml)/22.5.

For the Avandia intervention study, 24 T2DM patients were treated with rosiglitazone (4 mg twice daily) for 8 weeks. The clinical parameters of these patients at baseline and at 2 months after rosiglitazone treatment is described in Table 2, which is published as supporting information on the PNAS web site. The study protocol was approved by the Ethics Committee of the Medical Faculty (University of Hong Kong).

Construction of Adenoviral Vector for Expression of ANGPTL4 and Production of Adenoviruses. The method for cloning of the mouse ANGPTL4 gene and construction of its mammalian expression vector is described in Supporting Text, which is published as supporting information on the PNAS web site. The adenovirus expression vector that encodes FLAG-tagged ANGPTL4 was generated by using the Adeno-X Expression System (BD Biosciences Clontech). The recombinant virus was packaged and amplified in HEK293 cells and purified by CsCl density gradient centrifugation. The recombinant adenovirus that encodes luciferase was kindly provided by Christopher Rhodes (University of Washington, Seattle) (12).

Development of a Sandwich Enzyme-Linked Immunoassay for Measurement of Human ANGPTL4. The methods for expression of a recombinant human ANGPTL4 fragment and production of monoclonal and polyclonal antibodies against human ANGPTL4 are described in Supporting Text. The antihuman ANGPTL4 polyclonal antibody was biotinylated with a kit from Pierce and used as the detection antibody. The monoclonal antibody ED12B9 (5 μg/ml) was used for coating a 96-well microtiter plate overnight at 4°C. Human serum was diluted (1:50), and 100 μl of the diluted samples or standards were applied to each well, incubated at 37°C for 1 h, washed three times, then incubated with 100 μl of the detection antibody for another 2 h. After three washes, the wells were incubated with streptavidin-conjugated horseradish peroxidase for 1 h and subsequently reacted with tetramethyl-benzidine reagent for 15 min. A total of 100 μl of 2 M H2SO4 was added to each well to stop the reaction, and the absorbance at 450 nm was measured. The intra- and interassay coefficients of variance were determined by measuring five serum samples from healthy subjects in a total of six independent assays with duplicate determinations.

Glucose Tolerance Test. Mice were placed in clean cages with no food but with free access to water at ≈9:00 a.m. After a 6-h starvation, mice were weighed, and the tip of the tail was clipped to obtain blood for glucose measurement. Mice were injected i.p. with glucose (1 g per kg of body weight). Blood (≈5 μl) was taken from the tail tip at various time points for measurement of glucose concentration by using a Glucometer Elite (Bayer).

Analysis of Lipid Profiles, Insulin, and Adiponectin Levels in Mouse Serum. The levels of serum triglycerides (TGs) and free fatty acids (FFAs) were determined by using the TG glycerol phosphate oxidase reagent (Pointe Scientific, Lincoln Park, MI) and a Roche Diagnostics FFA kit, respectively. The serum levels of total cholesterol were measured by using a kit from Wako. Plasma insulin levels were quantified by using the commercial ELISA kits from Mercodia AB (Uppsala, Sweden). Circulating adiponectin was determined by using an in-house ELISA established in our laboratory (13).

Oil Red O Staining of Liver Sections and Quantification of Hepatic Glycogen Contents. Oil Red O staining of lipid droplets in liver sections was performed as described (14). The glycogen content in the liver extracts of control and transgenic mice was determined with amyloglucosidase according to the method of Keppler and Decker (15).

Isolation of Primary Rat Hepatocytes and Measurement of Hepatic Glucose Output. Primary hepatocytes were prepared from male Wistar rats (≈200 g) as described (16). Cells were plated on collagen type I-coated 12-well plates in DMEM with 10% FBS, 2 mM l-glutamine, 10 μM dexamethasone, and 10 μg/ml insulin at a density of 5 × 105 cells per well. The cells were allowed to adhere to the cell culture dishes for 8 h and then infected with Adv-ANGPTL4 or Adv-Luc at the concentrations of 50 plaque-forming units (pfu) per cell. At 24 h after infection, the cells were stimulated without or with different concentration of insulin for another 24 h. The medium was then replaced with 0.5 ml of glucose-free DMEM without phenol red and supplemented with 5 mM each alanine, valine, glycine, pyruvate, and lactate. After incubation for another 6 h, the glucose level in the medium was measured as described (16).

Statistical Analysis. Experiments were performed routinely with five to six mice per group with values presented as means ± SE. Statistical significance was determined by one-way ANOVA. In all statistical comparisons, a P value < 0.05 was used to indicate a significant difference.


Adenovirus-Mediated ANGPTL4 Expression Induces Hyperlipidemia, Hepatomegaly, and Fatty Liver in C57 Mice. To investigate the metabolic functions of ANGPTL4 in vivo, we generated the recombinant adenovirus that encodes the mouse full-length ANGPTL4 tagged with the FLAG-epitope to facilitate the detection of the expressed protein. A total of 5 × 109 pfu of the recombinant adenovirus that expresses ANGPTL4 (Adv-ANGPTL4) or luciferase (Adv-Luc as a control) was introduced into C57 mice through tail-vein injection. Expression of FLAG-tagged ANGPTL4 protein was detected in the circulation at day 1, peaked at day 4, and subsequently attenuated (Fig. 1). A trace amount of ANGPTL4 expression was still detectable at 2 weeks after injection.

Fig. 1.
Expression of ANGPTL4 protein in C57 mice following tail-vein injection with Adv-ANGPTL4. (A) The schematic diagram of mouse ANGPTL4 protein structure. Note that cysteines 76 and 80 are responsible for oligomerization (27). Asp-181, -232, and -320 are ...

ANGPTL4 overexpression did not significantly affect food intake or body weight at 2 weeks after treatment. On the other hand, this treatment caused a marked elevation of triglycerides and a relatively modest increase of total cholesterol in the circulation. Plasma levels of triglycerides rose to peak levels at day ≈4 after injection, attenuated sharply afterward, and returned to nearly normal levels within 2 weeks (data not shown).

The livers of ANGPTL4-treated mice were whitish and markedly enlarged compared with those of control mice treated with Adv-Luc (Fig. 2A). Liver weight in Adv-ANGPTL4 injected mice is ≈1.8-fold greater than that of the control mice injected with Adv-Luc (Fig. 2B). The hepatic TG levels in mice treated with Adv-ANGPTL4 were substantially higher than those in mice treated with Adv-Luc (Fig. 2C). Oil Red O staining showed that fine to medium-sized lipid droplets accumulated diffusely in the liver of Adv-ANGPTL4-treated mice (Fig. 2D). These results suggest that ANGPTL4 overexpression induces fatty liver and hepatomegaly.

Fig. 2.
Overexpression of ANGPTL4 induces hepatomegaly and fatty liver in mice. C57 mice were treated with 5 × 109 pfu of Adv-ANGPTL4 or Adv-Luc for a period of 2 weeks. (A) Liver images of mice treated with Adv-ANGPTL4 or Adv-Luc. (B and C) The liver ...

ANGPTL4 Overexpression Potently Decreases Blood Glucose Levels and Improves Glucose Tolerance in C57 Mice. In sharp contrast with its hyperlipidemia effect, ANGPTL4 overexpression markedly decreased blood glucose levels in both fasted and ad libitum states (Fig. 3A and Table 3, which is published as supporting information on PNAS web site). The glucose-lowering effect of ANGPTL4 was observed at day 2 after injection with Adv-ANGPTL4, and sustained throughout the treatment period. Two weeks after injection, blood glucose levels in mice injected with Adv-ANGPTL4 were still significantly lower than that in control mice injected with Adv-Luc. Hepatic glycogen content was significantly higher in the Adv-ANGPTL4-treated group than that in controls, indicating that ANGPTL4 might enhance insulin sensitivity in the liver tissue. Notably, when challenged with an i.p. glucose load, the mice treated with Adv-ANGPTL4 showed a much lower peak glucose concentration at 15 min and a faster decline of blood glucose levels throughout the glucose tolerance curve, suggesting that ANGPTL4 might also increase peripheral glucose disposal (Fig. 3B). ANGPTL4 treatment did not significantly affect the plasma concentrations of insulin or adiponectin in C57 mice.

Fig. 3.
Overexpression of ANGPTL4 causes a sustained decrease of blood glucose levels and improves glucose tolerance in C57 mice. (A) Blood glucose levels of mice at various days after injection without (control) or with 5 × 109 pfu of Adv-ANGPTL4 or ...

ANGPTL4 Overexpression Markedly Alleviates Hyperglycemia, Hyperinsulinemia, and Glucose Intolerance Associated with db/db Diabetic Mice. We next investigated the effect of ANGPTL4 on energy metabolism and insulin sensitivity in C57BLKS db/db mice, a genetically inherited diabetic mouse model that is characterized by severe hyperglycemia, hyperinsulinemia, and glucose intolerance. Overexpression of ANGPTL4 significantly elevated serum levels of triglycerides, FFA, and cholesterol, and increased the liver weights and liver TG contents in db/db diabetic mice (Table 4, which is published as supporting information on the PNAS web site). On the other hand, hyperglycemia in both fasted and ad libitum states was sharply decreased to a normal level in db/db diabetic mice treated with Adv-ANGPTL4 (Fig. 4A). This potent glucose-lowering effect was observed at day 2 after injection with Adv-ANGPTL4 and sustained for at least 2 weeks. In addition, ANGPTL4 treatment significantly decreased hyperinsulinemia and markedly improved glucose intolerance associated with this diabetic mouse model (Fig. 4 B and C).

Fig. 4.
The potent therapeutic effects of ANGPTL4 on hyperglycemia, hyperinsulinemia, and glucose intolerance associated with db/db diabetic mice. (A) Fasted and fed blood glucose levels of db/db diabetic mice at 1 and 2 weeks after injected without (control) ...

Quantitative real-time PCR analysis revealed that ANGPTL4 overexpression resulted in a significant down-regulation of phosphoenolpyruvate carboxykinase (PEPCK) and glucose-6-phosphatase (G6Pase), the two key enzymes involved in gluconeogenesis (Fig. 7, which is published as supporting information on the PNAS web site). On the other hand, ANGPTL4 increased the expression of several key lipogenic enzymes, including fatty acid synthase (FAS) and stearoyl-CoA desaturease (SCD1). Taken together, ANGPTL4-mediated changes in expression of FAS, SCD1, PEPCK, and G6PD might divert substrate from gluconeogenesis to lipogenesis, which in turn leads to decreased blood glucose levels and increased liver TG contents.

ANGPTL4 Inhibits Glucose Production in Rat Primary Hepatocytes. Our finding that ANGPTL4 can decrease blood glucose and increase hepatic glycogen contents suggests that the liver might be the target tissue of this protein. We next investigated the direct effect of ANGPTL4 on glucose output of rat primary hepatocytes by infecting these cells with Adv-ANGPTL4 or Adv-luc. ANGPTL4 protein in the culture media of rat primary hepatocytes was detected as the full-length as well as the cleaved form (Fig. 5A). Notably, the basal hepatic glucose output (HGO), as determined by measurement of glucose contents released into culture media, was significantly decreased in hepatocytes treated with Adv-ANGPTL4 (Fig. 5B). In addition, ANGPTL4 treatment enhanced the sensitivity of insulin to inhibit hepatic glucose output in this system (Fig. 5C), suggesting that inhibition of HGO might represent a potential mechanism underlying the glucose-lowering effect of ANGPTL4.

Fig. 5.
ANGPTL4 exerts direct inhibitory effects on glucose production in primary rat hepatocytes. (A) Western analysis of ANGPTL4 secreted from primary rat hepatocytes. Cells grown in a 12-well plate were infected with Adv-ANGPTL4 or Adv-Luc (50 pfu per cell) ...

Serum Concentrations of ANGPTL4 Are Decreased in Patients with T2DM and Are Inversely Correlated with Plasma Glucose Levels and HOMA IR. To validate the role of ANGPTL4 as a circulating hormone in humans, we established a sandwich ELISA method for measurement of this protein in human plasma. To this end, we generated both the monoclonal and polyclonal antibodies against human ANGPTL4, using a recombinant human ANGPTL4 fragment as the antigen (Fig. 8, which is published as supporting information on the PNAS web site.). The specificity of the antihuman ANGPTL4 monoclonal and polyclonal antibodies was validated by the fact that both of these antibodies can specifically immunoprecipitate the recombinant human ANGPTL4 from bacterial lysates (data not shown). The sandwich ELISA standard curve using the human recombinant ANGPTL4 fragments yielded a consistent r2 value >0.985.

Serum ANGPTL4 concentrations of healthy individuals were 345.04 ± 10.83 ng/ml in both genders. Notably, its serum concentrations in patients with T2DM were substantially lower than those in nondiabetic subjects, and this difference was significant in both gender groups (Fig. 6A). On the other hand, serum concentrations of ANGPTL4 in obese individuals without T2DM were similar to those in nonobese individuals (Fig. 6B). Treatment of T2DM patients with rosiglitazone for 8 weeks significantly elevated circulating concentrations of ANGPTL4 (243.5 ± 12.6 at baseline versus 309.7 ± 14.8 ng/ml after treatment, P < 0.01), suggesting that the PPARγ agonist induces ANGPTL4 production in humans.

Fig. 6.
Serum levels of ANGPTL4 are decreased in patients with T2DM but not in obese individuals without T2DM. (A) Comparison of serum ANGPTL4 concentrations between T2DM patients and age-, BMI-, and sex-matched healthy individuals. (B) Serum ANGPTL4 levels in ...

Bivariate correlation analysis showed a significant inverse relationship between the serum levels of ANGPTL4 and plasma glucose concentrations as well as HOMA IR (Fig. 9, which is published as supporting information on the PNAS web site), suggesting that ANGPTL4 might also act as a glucose-lowering and insulin-sensitizing hormone in humans. On the other hand, there were no significant correlation between serum ANGPTL4 concentrations and serum levels of triglycerides and total cholesterol.

A recent report demonstrated that human plasma ANGPTL4 is present as the full-length as well as the cleaved fragment (8). We next investigated the percentage compositions of these two forms in human plasma by Western blot analysis. In line with those observed by Mandard et al. (8), ANGPTL4 in human plasma was detected as two distinct bands (Fig. 10, which is published as supporting information on the PNAS web site), which might represent the full-length and the NH2-terminal domain of ANGPTL4. Quantitative analysis revealed that the NH2-terminal domain and the full-length ANGPTL4 account for 79.2 ± 6.3% and 21.8 ± 1.9% of the total serum ANGPTL4 in healthy subjects. The ratio between the NH2-terminal domain and the full-length ANGPTL4 was similar between the two genders. There were no significant differences in this ratio among healthy subjects, obese individuals, and patients with T2DM.


ANGPTL4 is a downstream target gene of both PPARγ and PPARα, the agonists of which are widely used for the treatment of type 2 diabetes, insulin resistance, and dyslipidemia. In this study, we provided direct evidence that ANGPTL4 is a blood-borne hormone involved in regulating glucose homeostasis, lipid metabolism, and systemic insulin sensitivity. Overexpression of ANGPTL4 in mice potently decreased blood glucose levels and improved glucose tolerance. In db/db diabetic mice, ANGPTL4 treatment reduced hyperglycemia to a normal level, and markedly alleviated glucose intolerance and hyperinsulinemia. Notably, chronic treatment of db/db mice with PPARγ agonists also produced a similar insulin-sensitizing and glucose-lowering effect (17, 18), suggesting that the glucose-lowering effects of the PPARγ agonists might be partly mediated by induction of ANGPTL4 production.

Our results suggest that the glucose-lowering effect of ANGPTL4 could be due to its direct actions on hepatocytes. This conclusion is supported by the fact that ANGPTL4 suppresses basal glucose output and enhances the sensitivity of insulin to inhibit glucose production in primary rat hepatocytes. The inhibitory effect of ANGPTL4 on hepatic glucose production is reminiscent of adiponectin (16, 19), another adipokine induced by the PPARγ agonists. However, unlike ANGPTL4, adiponectin has no effect on basal glucose output. Adiponectin exerts its hepatic actions by activation of AMP-activated protein kinase (20), whereas we found that ANGPTL4 had no effect on this kinase in mice (data not shown), suggesting that these two adipokines might act through distinct pathways. The detailed metabolic pathways and signal transduction events that underlie the hepatic action of ANGPTL4 on glucose metabolism remain to be clarified.

It is known that PPARα agonists have potent lipid-lowering function, mainly by enhancing fatty acid oxidation and inhibiting lipogenesis in the liver tissue (21). PPARγ agonists have also been shown to decrease circulating TG levels and hepatic lipid contents in humans and rodents (22, 23). Given that the agonists of both PPARα and PPARγ induce expression of ANGPTL4, this adipokine would be expected to exert beneficiary effect on lipid metabolism. Unexpectedly, we found that ANGPTL4 overexpression in mice results in a hyperlipidemia phenotype with severe hepatomegaly and liver steatosis, suggesting that ANGPTL4 is not involved in mediating the beneficial effects of these agonists on lipid metabolism and hepatic steatosis. An earlier study from Yoshida et al. (9) also demonstrated that injection of recombinant ANGPTL4 acutely induces hypertriglyceridemia in mice possibly by inhibition of lipoprotein lipase activity. More recently, Ge et al. (10) reported that adenovirus-mediated expression of ANGPTL4 raises plasma triglyceride levels by suppressing the clearance of very-low-density lipoprotein from the circulation.

Our finding that the beneficial effects of ANGPTL4 on glucose homeostasis are associated with hyperlipidemia and fatty liver is somewhat surprising. Hyperlipidemia and fatty liver are thought to be part of the metabolic syndrome, and are clinically associated with hyperglycemia, hyperinsulinemia, and insulin resistance. Nevertheless, several recent studies on animal models suggest that certain signaling pathways that reduce blood glucose and improve insulin sensitivity can simultaneously induce hyperlipidemia and fatty liver. Ono et al. (24) showed that hepatic activation of Akt, a key signaling molecule that mediates the metabolic actions of insulin, results in a markedly hypoglycemic, hypoinsulinemic, and hypertriglyceridemic phenotype with fatty liver and hepatomegaly. Similarly, liver-specific depletion of PTEN, a negative regulator of the phosphatidylinositol 3-kinase/Akt pathway, causes enhanced liver insulin action with improved systemic glucose tolerance, but concurrently induces hypertriglyceridemia and liver steatosis (25). A similar phenotype was also observed in PGC-1 (peroxisome proliferator-activated PPARγ coactivator-1) knockdown mice (26). Further studies are needed to investigate whether ANGPTL4 exerts its liver actions through these pathways.

It is worth noting that ANGPTL4-mediated hyperlipidemia is only of a transient nature. Ge et al. (10) showed that, although ANGPTL4 expression remained robust for >2 weeks, plasma triglycerides rose to peak levels at the first several days and then returned to near-normal levels within 2 weeks. In this study, we observed that ANGPTL4-mediated elevation of both plasma triglycerides and total cholesterol peaked between 2 and 4 days, and these effects were markedly attenuated afterward. Notably, the sharp attenuation of its lipid-elevating effect coincides with the decreased ANGPTL4 expression in the circulation. On the other hand, the glucose-lowering effect of ANGPTL4 remained robust throughout the treatment period in both C57 lean mice and db/db diabetic mice, despite the marked reduction of ANGPTL4 expression at the later stage. These results suggest that the lipid-elevating effect of ANGPTL4 requires substantially higher concentrations of this protein than those needed for its glucose-lowering action.

The role of ANGPTL4 as a potential glucose-lowering hormone was also supported by our finding that serum levels of ANGPTL4 are inversely correlated with plasma glucose concentrations in human subjects. Furthermore, the serum ANGPTL4 levels in patients with T2DM, but not in obese subjects without diabetes, are substantially decreased, suggesting that decreased ANGPTL4 could be a causative factor of hyperglycemia. It is interesting to note that adiponectin, another fat-derived hormone with direct insulin-sensitizing and glucose-lowering functions, is also decreased in T2DM patients (4). However, unlike adiponectin, serum ANGPTL4 levels lack obvious correlation with adiposity, and the circulating concentrations of triglycerides and total cholesterol, suggesting that lipid-elevating effects of ANGPTL4 observed in mice might not be physiologically relevant.

In plasma, ANGPTL4 is present as the full-length oligomerized protein as well as the proteolytically cleaved form (8, 27). The processing of ANGPTL4 oligomerization and proteolysis is similar with that of adiponectin. In the case of adiponectin, its proteolytic fragment is functionally different from the full-length adiponectin, with the former enhancing muscular fatty acid oxidation (28) and the later inhibiting hepatic glucose production (16, 19). Different oligomeric forms of adiponectin have also been reported to act on different target tissues and activate distinct signaling pathways (29). Given the similarities between adiponectin and ANGPTL4, it is tempting to speculate that the multiple metabolic effects of ANGPTL4 on glucose and lipid metabolism might be differentially mediated by its distinct oligomeric complexes or proteolytic fragments.

In summary, our present study provided both clinical and experimental evidence supporting the role of ANGPTL4 as a blood-borne hormone involved in maintaining glucose homeostasis. The glucose-lowering effect of ANGPTL4 might be partly mediated by its direct inhibition on hepatic glucose output. However, the beneficial effects of ANGPTL4 on glucose homeostasis are associated with undesirable hyperlipidemia and fatty liver in mice. We believe that further elucidation of molecular and structural mechanism that underlies the multiple metabolic effects of ANGPTL4 might help to develop novel therapeutic strategies for T2DM and other obesity-related metabolic disorders.

Supplementary Material

Supporting Information:


This study was supported by a seeding fund for basic research, the University of Hong Kong, and the Research Grant Council of Hong Kong (RGC HKU7404/04).


Author contributions: A.X. and K.S.L.L. designed research; M.C.L., K.W.C., Y.W., J.Z., R.L.C.H., B.C., W.-S.C., and A.W.K.T. performed research; K.S.L.L. contributed new reagents/analytic tools; A.X., M.C.L., K.W.C., Y.W., J.Z., R.L.C.H., J.Y.X., and K.S.L.L. analyzed data; and A.X. wrote the paper.

This paper was submitted directly (Track II) to the PNAS office.

Abbreviations: ANGPTL4, angiopoietin-like protein 4; PPAR, peroxisome proliferator-activated receptor; T2DM, type 2 diabetes mellitus; TG, triglyceride; FFA, free fatty acid; pfu, plaque-forming unit.


1. Kershaw, E. E. & Flier, J. S. (2004) J. Clin. Endocrinol. Metab. 89, 2548-2556. [PubMed]
2. Lyon, C. J., Law, R. E., Hsueh, W. A., Chaldakov, G. N., Stankulov, I. S., Hristova, M. & Ghenev, P. I. (2003) Endocrinology 144, 2195-2200. [PubMed]
3. Friedman, J. M. & Halaas, J. L. (1998) Nature 395, 763-770. [PubMed]
4. Matsuzawa, Y., Funahashi, T., Kihara, S. & Shimomura, I. (2004) Arterioscler. Thromb. Vasc. Biol. 24, 29-33. [PubMed]
5. Yoon, J. C., Chickering, T. W., Rosen, E. D., Dussault, B., Qin, Y., Soukas, A., Friedman, J. M., Holmes, W. E. & Spiegelman, B. M. (2000) Mol. Cell. Biol. 20, 5343-5349. [PMC free article] [PubMed]
6. Kersten, S., Mandard, S., Tan, N. S., Escher, P., Metzger, D., Chambon, P., Gonzalez, F. J., Desvergne, B. & Wahli, W. (2000) J. Biol. Chem. 275, 28488-28493. [PubMed]
7. Kim, I., Kim, H. G., Kim, H., Kim, H. H., Park, S. K., Uhm, C. S., Lee, Z. H. & Koh, G. Y. (2000) Biochem. J. 346, 603-610. [PMC free article] [PubMed]
8. Mandard, S., Zandbergen, F., Tan, N. S., Escher, P., Patsouris, D., Koenig, W., Kleemann, R., Bakker, A., Veenman, F., Wahli, W., et al. (2004) J. Biol. Chem. 279, 34411-34420. [PubMed]
9. Yoshida, K., Shimizugawa, T., Ono, M. & Furukawa, H. (2002) J. Lipid Res. 43, 1770-1772. [PubMed]
10. Ge, H., Yang, G., Yu, X., Pourbahrami, T. & Li, C. (2004) J. Lipid Res. 45, 2071-2079. [PubMed]
11. Wat, N. M., Lam, T. H., Janus, E. D. & Lam, K. S. (2001) Int. J. Obes. Relat. Metab. Disord. 25, 1789-1793. [PubMed]
12. Wrede, C. E., Dickson, L. M., Lingohr, M. K., Briaud, I. & Rhodes, C. J. (2002) J. Biol. Chem. 277, 49676-49684. [PubMed]
13. Xu, A., Yin, S., Wong, L., Chan, K. W. & Lam, K. S. (2004) Endocrinology 145, 487-494. [PubMed]
14. Xu, A., Wang, Y., Keshaw, H., Xu, L. Y., Lam, K. S. & Cooper, G. J. (2003) J. Clin. Invest. 112, 91-100. [PMC free article] [PubMed]
15. Keppler, D. & Decker, K. (1974) in Methods of Enzymatic Analysis, ed. Bergmeyer, N. U. (VCH, Weinheim, Germany), Vol. 6, pp. 2225-2228.
16. Wang, Y., Xu, A., Knight, C., Xu, L. Y. & Cooper, G. J. (2002) J. Biol. Chem. 277, 19521-19529. [PubMed]
17. Berger, J., Leibowitz, M. D., Doebber, T. W., Elbrecht, A., Zhang, B., Zhou, G., Biswas, C., Cullinan, C. A., Hayes, N. S., Li, Y., et al. (1999) J. Biol. Chem. 274, 6718-6725. [PubMed]
18. Berger, J., Bailey, P., Biswas, C., Cullinan, C. A., Doebber, T. W., Hayes, N. S., Saperstein, R., Smith, R. G. & Leibowitz, M. D. (1996) Endocrinology 137, 4189-4195. [PubMed]
19. Berg, A. H., Combs, T. P., Du, X., Brownlee, M. & Scherer, P. E. (2001) Nat. Med. 7, 947-953. [PubMed]
20. Yamauchi, T., Kamon, J., Minokoshi, Y., Ito, Y., Waki, H., Uchida, S., Yamashita, S., Noda, M., Kita, S., Ueki, K., et al. (2002) Nat. Med. 8, 1288-1295. [PubMed]
21. Lee, C. H., Olson, P. & Evans, R. M. (2003) Endocrinology 144, 2201-2207. [PubMed]
22. Schoonjans, K. & Auwerx, J. (2000) Lancet 355, 1008-1010. [PubMed]
23. Kawaguchi, K., Sakaida, I., Tsuchiya, M., Omori, K., Takami, T. & Okita, K. (2004) Biochem. Biophys. Res. Commun. 315, 187-195. [PubMed]
24. Ono, H., Shimano, H., Katagiri, H., Yahagi, N., Sakoda, H., Onishi, Y., Anai, M., Ogihara, T., Fujishiro, M., Viana, A. Y., et al. (2003) Diabetes 52, 2905-2913. [PubMed]
25. Stiles, B., Wang, Y., Stahl, A., Bassilian, S., Lee, W. P., Kim, Y. J., Sherwin, R., Devaskar, S., Lesche, R., Magnuson, M. A. & Wu, H. (2004) Proc. Natl. Acad. Sci. USA 101, 2082-2087. [PMC free article] [PubMed]
26. Koo, S. H., Satoh, H., Herzig, S., Lee, C. H., Hedrick, S., Kulkarni, R., Evans, R. M., Olefsky, J. & Montminy, M. (2004) Nat. Med. 10, 530-534. [PubMed]
27. Ge, H., Yang, G., Huang, L., Motola, D. L., Pourbahrami, T. & Li, C. (2004) J. Biol. Chem. 279, 2038-2045. [PubMed]
28. Fruebis, J., Tsao, T. S., Javorschi, S., Ebbets-Reed, D., Erickson, M. R., Yen, F. T., Bihain, B. E. & Lodish, H. F. (2001) Proc. Natl. Acad. Sci. USA 98, 2005-2010. [PMC free article] [PubMed]
29. Tsao, T. S., Tomas, E., Murrey, H. E., Hug, C., Lee, D. H., Ruderman, N. B., Heuser, J. E. & Lodish, H. F. (2003) J. Biol. Chem. 278, 50810-50817. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...