The Acute-Phase Protein Orosomucoid Regulates Food Intake and Energy Homeostasis via Leptin Receptor Signaling Pathway

Obesity is a condition marked by excess accumulation of body fat that results from an imbalance between calorie intake and energy expenditure. Energy homeostasis in the body is maintained by the integrated actions of multiple factors (1,2), including adipose hormones (such as leptin and adiponectin), gastrointestinal hormones (such as insulin, ghrelin, and cholecystokinin), and nutrient-related signals (such as free fatty acids). In addition to acting on peripheral tissues, these actions can also influence central circuits in the hypothalamus, brainstem, and limbic system to modulate food intake and energy expenditure (1,3). Notably, the adipose tissue–produced leptin is a major regulator of fat, and the level of leptin in circulation is proportional to body fat (4) and is a reflection of long-term nutrition status as well as acute energy balance. Furthermore, leptin deficiency or leptin receptor (LepR) mutation leads to hyperphagia, obesity, and insulin resistance (5), whereas administration of leptin causes weight loss and improved insulin resistance and hyperglycemia in type 2 diabetes mice (6,7). Patients with leptin deficiency or LepR mutation also develop severe obesity (8,9). It is evident that hypothalamic LepR (10,11) is critical for leptin-mediated regulation of energy metabolism, as impairment of LepR signaling in the hypothalamus selectively results in hyperphagia and adiposity (1,12,13).

Orosomucoid (ORM), also known as α1-acid glycoprotein (AGP), is one of the acute-phase proteins. There are two isoforms of ORM in human (ORM1 and ORM2), one isoform in rat (ORM), and three isoforms in mouse (ORM1, ORM2, and ORM3) (14). These genes have an identical structure with six exons and five introns. Both in humans and mice, constitutive level of ORM1 is much higher (fivefold) than ORM2, and only ORM1 can be induced by acute-phase stimuli (15). Although it is mainly synthesized by the liver, many extrahepatic tissues, including adipocytes, heart, and brain, are capable of producing ORM under myriad physiological and pathological conditions (16–19). A variety of activities have been attributed to ORM, which include modulating immunity, carrying drugs, maintaining the capillary barrier, and mediating sphingolipid metabolism (14,20–23). It has been reported that the effects of ORM on macrophages, neutrophils, and liver parenchymal cells are mediated by membrane receptor CCR5, Siglect-5, and hemoglobin β-chain, respectively (24–26). Interestingly, increase of serum ORM level has been observed in obese humans, mice, and Ossabaw pigs (17,27–29). The increased level is correlated with BMI, body fat mass, serum leptin, and fasting plasma glucose level in human (27,30). In addition, adipose ORM level is correlated with adiponectin that regulates glucose level and fatty acid breakdown and is regulated by insulin, high glucose, and free fatty acid in differentiated adipocytes (17,27). These results suggested that ORM might participate in the regulation of energy balance.

In this study, we found alterations of energy homeostasis in mice deficient of ORM1, which accounts for the majority of serum ORM as well as most of the changes induced by acute-phase stimuli (15,31). The aberrant energy homeostasis is characterized by significant elevation in body weight and fat mass, increased serum total cholesterol (TC), fatty liver, and insulin and leptin resistance. We also found that ORM derived from adipose and liver tissues is regulated by short- or long-term nutrition signals and administration of ORM affects feeding behavior. Furthermore, we demonstrate that ORM binds to LepR and activates the Janus kinase (JAK)2–signal transducer and activator of transcription (STAT)3 pathway in hypothalamus. Thus, ORM could function as an agonist for LepR and is an important regulator in food intake and energy homeostasis.

ORM was purchased from Sigma-Aldrich (St. Louis, MO). BSA was obtained from Boguang Biological Technology (Shanghai, China). IgG was from Beyotime Institute of Biotechnology (Shanghai, China). Fluorescein isothiocyanate (FITC)-labeled ORM and BSA were made by Youke Biological Technology (Shanghai, China). Antibodies against ORM (rat) and LepR were purchased from Abcam (Cambridge, U.K.). Antibodies against JAK, phosphorylated (p)-JAK, STAT3, and p-STAT3 were from Cell Signaling Technology (Danvers, MA). Antibody against ORM (mouse) was obtained from Genway (San Diego, CA). Antibodies against GAPDH and tubulin were from Beyotime Institute of Biotechnology. Secondary antibodies conjugated with IRDye 800CW were from Rockland Immunochemicals, Inc. (Limerick, PA). The LepR small interfering RNA and its control small interfering RNA were from Santa Cruz Biotechnology (Dallas, TX). Lentivirus carrying full-length ORM1 or LepR short hairpin (sh)RNA was constructed by Shanghai GenePharma Co., Ltd. (Shanghai, China). The sequence used for LepR shRNA is 5′-GCTGAAATTGTCTCAGCTACA-3′. The 60% high-fat diet (HFD) and standard chow were purchased from Shanghai SLAC Laboratory Animal Co., Ltd. (Shanghai, China).

Mouse hypothalamic GT1-7 cells were generously provided by Professor Xiao-Ying Li from the Shanghai Clinical Center for Endocrine and Metabolic Diseases at Shanghai Jiaotong University School of Medicine (Shanghai, China). C2C12 cells (mouse muscle myoblasts) were obtained from Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences. These cells were cultured in DMEM (Gibco) supplemented with 10% FBS (Gibco). All cells were incubated at 37°C in a 5% CO₂ incubator. For knockdown studies, these cells were transfected with Lipofectamine 2000 (Invitrogen, Carlsbad, CA) according to the manufacturer’s instructions.

Eight-week-old male db/db and ob/ob mice were purchased from Shanghai SLAC Laboratory Animal Co., Ltd. Male C57BL/6 mice (18–22 g) and Sprague-Dawley rats (180–200 g) were purchased from Sino-British SIPPR/BK Laboratory Animals (Shanghai, China). ORM1 knockout mice were generated as previously described (32) and were backcrossed 10 times with C57BL/6 mice before they were characterized. All animal experiments were undertaken in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and with the approval of the Scientific Investigation Board of the Second Military Medical University.

Total RNA was extracted with TRIzol reagent (Invitrogen) following manufacturer’s instructions. Real-time quantitative RT-PCR analysis was performed using the SYBR RT-PCR kits (Takara, Otsu, Japan). Primer sequences were shown in Supplementary Table 1.

Serum levels of ORM, leptin, and insulin were detected by ELISA kit according to the manufacturer’s instruction. A rat ORM ELISA kit was obtained from Abcam. A leptin ELISA kit was bought from R&D Systems (Minneapolis, MN). An insulin ELISA kit was bought from Millipore (Billerica, MA). Total plasma cholesterol and triglyceride (TG) were measured by the Clinical Biochemical Laboratory in Changhai Hospital (Shanghai, China).

The effect of ORM on eating behavior was evaluated by the amount of food intake. For fasting-induced food intake, mice were starved overnight (no drinking limited), and the weight of consumed food at 2 h, 8 h, and 24 h was recorded 30 min after tail vein injection of vehicle or ORM (100 mg/kg). For spontaneous food intake, mice were intraperitoneally injected with vehicle or ORM (50 mg/kg/day) for 4 or 7 consecutive days as indicated, and the weight of daily consumed food was recorded.

Frozen 20-μm-thick mouse liver sections were fixed in 4% paraformaldehyde. After removal of formalin and washing again with PBS, slides were incubated with Oil Red O working solution (Sigma-Aldrich) at room temperature for 10 min. Then, slides were differentiated in 60% ethanol solution and rinsed three times with distilled water, stained in hematoxylin for 30 s, and washed thoroughly. Photos were taken under the optical microscope (Olympus, Tokyo, Japan).

GT1-7 cells or C2C12 cells were treated with ORM (10 μg/mL) or vehicle for 3 h and fixed with 4% paraformaldehyde. Cells were incubated with primary antibodies against ORM (1:300; rabbit, Genway) and LepR (1:50; goat, Santa Cruz Biotechnology, Santa Cruz, CA) overnight. Proximity ligation assay (PLA) was performed using the Duolink in situ PLA kit (Olink Bioscience, Uppsala, Sweden) with PLA PLUS or MINUS probes for rabbit or goat antiserum. The nuclei of cells were stained using DAPI (Olink Bioscience).

Cells and tissues were lysed with Cell & Tissue Protein Extraction Reagent (Kangchen, Shanghai, China) supplemented with a protease inhibitor mixture (Kangchen). For immunoblotting, cell and tissue lysates (30–50 μg protein) in the supernatant were separated by SDS-PAGE and transferred to nitrocellulose membranes. For immunoprecipitation, cell or tissue lysates (100–200 μg) were incubated with 2 μg either polyclonal antibody against ORM (Abcam) or polyclonal LepR antibody (Abcam) coupled to protein G–sepharose (Invitrogen) overnight at 4°C. Immunoprecipitates were washed and separated by SDS-PAGE and transferred to nitrocellulose membranes. The membranes were then probed for specific proteins.

Immunohistochemistry was carried out according to the standard protocol using the slides from brain tissues. After overnight incubation at 4°C with the anti-ORM antibody (1:50; rat, MyBioSource) or anti-LepR antibody (1:50; goat, Santa Cruz Biotechnology), the slides were washed and incubated with Cy3-conjugated anti-rat IgG or Alexa Fluor 488–conjugated anti-goat IgG (Jackson ImmunoResearch) for 2 h at room temperature. After extensive washing, the nuclei were stained with DAPI. Negative controls were run concurrently, except that the antibody dilution buffer was used to substitute the primary antibody.

The interaction of ORM with the surface of GT1-7 cells was analyzed using flow cytometry. Briefly, cells were blocked with 5% BSA and incubated with FITC-conjugated BSA, FITC-conjugated ORM (10 μg/mL), FITC-conjugated ORM in combination with LepR antibody (mouse-blocking peptide, 10 μg/mL) (Alpha Diagnostic International, San Antonio, TX), or FITC-conjugated ORM in combination with leptin (10 μg/mL; Abcam) for 1 h and washed with PBS three times and then analyzed by flow cytometry.

Male C57BL/6 mice (25–30 g) were fixed in an ALC-H motorized digital stereotaxic instrument (Shanghai Alcott Biotech Co., Ltd., Shanghai, China). Lentiviruses expressing ORM1, shRNA silencing LepR, or their corresponding controls were injected into arcuate nucleus (ARC) (bilaterally, 0.5 μL/side; 10⁹ transduction units/μL) with microliter syringes (Hamilton Company, Reno, NV). The coordinates for lateral ventricular injection were as follows: anterior-posterior (from bregma), –2 mm; dorsal-ventral (from skull surface), –6 mm; and medial-lateral, 0.24 mm. All the mice with ARC injection were examined via immunofluorescence microscopy to verify the correct placement after experiments were completed. Virus in ARC occasionally spread over other nucleus (usually ventromedial nucleus) in the hypothalamus. Data from mice with proven injection in ARC were collected.

Intraperitoneal glucose tolerance testing was performed according to the protocol recommended by the Animal Models of Diabetic Complications Consortium. Mice were fasted for 6 h by removal to a clean cage without food at the end of their dark (feeding) cycle. After 6 h of fasting, the basal blood glucose derived from tail venous blood was detected. Then 2 mg/g body wt i.p. glucose was injected. Blood samples were collected at 15, 30, 60, and 120 min postinjection to measure blood glucose levels using an OneTouch Ultra glucometer (LifeScan, Milpitas, CA).

The model of the leptin-binding domain of mouse LepR (LepR-LBD) was built by the homology modeling protocol in Discovery Studio, version 3.5 (Accelrys, San Diego, CA). The X-ray crystal structure of human LepR-LBD (33) was obtained from the Protein Data Bank (34) (PDB) (PDB entry code 3V6O) and used as the template for model building. They share ∼75.7% of sequence identity, and good superimposition was obtained between the crystal structure and the constructed model (Supplementary Fig. 1A and B). In total, 20 models were generated. The best model verified by MODELER and Profiles-3D (Supplementary Fig. 1C) was chosen as the mouse LepR-LBD structure. The X-ray crystal structure of human ORM (i.e., AGP) was taken from PDB (PDB entry code 3KQ0) (35). Both structures were then prepared with the Protein Preparation Wizard workflow in Maestro, version 9.3 (Schrödinger, L.L.C., New York, NY). Docking of LepR-LBD with ORM was performed using ZDOCK, version 2.3 (36). Two thousand outputs of the ORM poses were generated for each docking run. All poses were clustered based on an all-against-all root mean square deviation matrix.

Data are presented as means ± SEM. Statistical analyses were performed using GraphPad Prism. Two-tailed Student t tests were used to compare two distinct groups. One-way ANOVA followed by Bonferroni test was used to compare more than two groups. Interactions were analyzed using two-way ANOVA followed by Bonferroni test. P < 0.05 was considered statistically significant.

In previous study, we generated a strain of ORM1-deficient mice to explore the biological function of ORM1 (Supplementary Fig. 2A–D). Interestingly, they had a significantly increased serum insulin level at the age of 4 weeks compared with wild-type littermates, although these mice had similar serum leptin, TC, TG, body weight, fat mass, and glucose tolerance (Supplementary Fig. 2E–K). When examined at the age of 24 weeks, ORM1-deficient mice showed markedly elevated serum levels of insulin and leptin (Fig. 1A and B), indicatives of insulin and leptin resistance. TC was also increased in these mice, whereas TG levels were not significantly different between wild-type and ORM1-deficient mice (Fig. 1C and D). Of note, compared with wild-type mice, ORM1-deficient mice had a significant increase in fat mass (1.32 ± 0.20 g vs. 2.27 ± 0.27 g, P < 0.05) and a mild increase in body weight (31.3 ± 0.75 g vs. 34.6 ± 1.05 g, P < 0.05) (Fig. 1E and F and Supplementary Fig. 2L and M). While there were no increases in the weights of heart, kidney, pancreas, and spleen (Supplementary Fig. 2L), the weight of liver was markedly raised, accompanied by increased adipose deposition (Fig. 1G). In addition, ORM1-deficient mice showed impaired glucose tolerance (Fig. 1H). Therefore, deficiency of ORM1 in mice impaired energy homeostasis, which prompted us to further investigate the mechanisms and potential physiological and pathological significance of the ORM1 action.

To further clarify the relationship between ORM and obesity, we examined ORM expression in mice with HFD-induced obesity and in LepR-defected db/db obese mice. As shown in Fig. 2, these mice had significantly elevated serum levels of leptin and insulin (Fig. 2A–D) as well as increased ORM levels in serum, liver, and subcutaneous adipose tissues (Fig. 2E–J). It is worth noting that the anti-ORM antibody we used for immunoblotting cannot distinguish the three murine ORM isoforms (ORM1 and ORM2 were single polypeptide chains of 207 amino acids each, and ORM3 contains 206 amino acids). Real-time PCR was then used to determine the level of each isoform. As shown in Fig. 2K and L, ORM1 is the most abundant isoform that is likely responsible for the majority of ORM increase in the liver and subcutaneous adipose tissues (Fig. 2M–P). Interestingly, it was shown that HFD feeding caused significant increases in both ORM1 and -2 mRNA levels in fat tissue without significant changes in hepatic ORM1 and -2 mRNA levels (17). The differences in HFD feeding period (7 days in the report vs. 7 weeks in our study) and fat tissue source (epididymal fat tissue in the report vs. subcutaneous fat tissue in our study) might be responsible for the inconsistency. Furthermore, we found that ORM was widely expressed in the brain including hypothalamic area as detected by an antibody that recognizes all forms of ORM (Supplementary Fig. 3A). Isoform-specific quantitative PCR revealed that ORM2 is the major isoform (Supplementary Fig. 3B) in hypothalamus, and the expression of ORMs in the hypothalamus tissue did not change in db/db mice or in response to HFD (Supplementary Fig. 3C and D).

Short-term acute nutritional changes (24-h fasting and 2-h refeeding) are known to affect serum leptin and insulin levels independent of adiposity (37–39), which illustrates how the body maintains energy homeostasis. We used this model to explore the potential role of ORM in the process. Similar to leptin and insulin, serum ORM level significantly decreased in fasted animals, and it increased markedly after refeeding in both mice and rats (Fig. 3A–C and Supplementary Fig. 4). The level of ORM was further assessed in tissues. As shown in Fig. 3D, while ORM expression in liver showed a pattern of change similar to that of serum, the amount of ORM in subcutaneous adipose tissue and hypothalamus tissue (Fig. 3E and Supplementary Fig. 3E) did not have such a pattern. These data support the notion that liver is responsible for producing ORM under an acute nutritional alteration condition.

The close association of ORM and energy homeostasis prompted us to ask whether ORM is involved the regulation of feeding behavior. We examined the effect of purified ORM on food intake in C57BL/6 mice. Tail vein injection of ORM significantly decreased fasting-induced food intake at 2 h, 8 h, and 24 h (Fig. 4A). Intraperitoneal injection of ORM for consecutive 4 days also markedly reduced spontaneous daily food consumption (Fig. 4B). Interestingly, these effects were still found in HFD-induced obese mice and ob/ob obese mice that are insulin and/or leptin resistant (Fig. 4C–F), whereas they were not present in LepR-defected db/db mice (Fig. 4G and H), indicating the involvement of LepR in the process. The ARC of the hypothalamus plays a pivotal role in the integration of signals regulating appetite. The ARC lies in close proximity to the median eminence that lacks a complete blood brain barrier (40) and thus is uniquely placed to respond to circulating hormonal signals. It has been shown that the ARC is essential for the regulation of energy balance through LepR (1,41). We therefore asked whether blockade of LepR signaling in hypothalamic ARC affects the action of ORM on energy homeostasis. Lentivirus encoding shRNA targeting LepR (shLepR) or control was injected into hypothalamic ARC stereotaxically (Supplementary Fig. 5A–C). Seven days later, as expected, LepR knockdown in ARC resulted in a significant increase in fasting-induced food intake and spontaneous daily food intake (Fig. 4J and K). While ORM administration still reduced fasting-induced food intake significantly at indicated times in mice injected with control lentivirus, its effect was significantly attenuated in mice treated with shLepR (Fig. 4I). ORM-induced reduction in spontaneous daily food intake and serum insulin level was also reversed with LepR knockdown in ARC (Fig. 4J and K). Additionally, injection of lentiviral vector LV-ORM1 into ARC to overexpress ORM1 (Supplementary Fig. 5D and E) resulted in a significant decrease in food intake compared with control mice (Fig. 4L), and this effect lasted up to 6 weeks. Meanwhile, body weight of these mice and their serum insulin level began to decrease 2 weeks after LV-ORM1 injection (Fig. 4M and N). These data indicated that ORM acts on hypothalamus to regulate food intake in a LepR-dependent manner. Despite its effect in reducing food intake, 4 days of ORM treatment did not affect body weight, serum insulin, TC, or TG levels in HDF or ob/ob mice (Supplementary Fig. 6). Since these two obese animal models have insulin and/or leptin resistance, it is likely that longer time of ORM administration is needed to further significantly change energy homeostasis.

We then examined whether there exists a direct interaction between ORM and LepR. From rat hypothalamus tissue, an ORM/LepR complex could be coimmunoprecipitated, especially after ORM tail vein injection (Fig. 5A). Use was also made of PLAs, which showed that the ORM and LepR interact on membrane of GT1-7 cells (Fig. 5B). Furthermore, flow cytometry analysis found that FITC-labeled ORM bound to GT1-7 cell membrane and the binding was completely inhibited by a LepR-blocking peptide (Fig. 5C). Taken together, these data demonstrated that ORM interacts directly with LepR. Intriguingly, this binding was not competitively blocked by leptin (Fig. 5D). Structural studies have indicated that leptin binds to LepR-LBD, likely in the middle of the LBD’s outer side (33). Based on molecular docking, we predicted that the binding site of ORM is in the middle of the LBD’s inner side (Fig. 5E), which may explain the lack of competition between ORM and leptin. Four clusters were further identified as the most possible binding poses out of a total 2,000 output from the docking (Supplementary Table 2). Among them, clusters 1 and 7 provided us with the detailed interactions between ORM and LepR-LBD (Supplementary Fig. 7A and B). They will be used as basic models for further mutational analysis.

In mice and humans, only the long isoform of LepR (LepR isoform b) has an elongated intracellular domain coupled to downstream signaling cascades (42). Although LepR isoform b is primarily expressed in hypothalamic regions and ARC (11), it can also be detected in a broad range of other cell types in line with the pleiotropic effects of leptin in peripheral tissues (43). We further observed the interaction of ORM and LepR in rat muscle and mouse myoblast C2C12 cells and achieved similar results (Supplementary Fig. 7C and D).

It has been shown that binding of leptin to LepR activates the associated JAK2 tyrosine kinase that phosphorylates LepR on tyrosine residues (44,45), which in turn leads to the recruitment and phosphorylation of STAT3. Disruption of the STAT3-binding site on LepR resulted in hyperphagia and obesity (46). Therefore, activation of STAT3 is a critical event for LepR-mediated signaling and is a commonly used marker. As shown in Fig. 6, ORM treatment induced a significant increase in the phosphorylation of JAK2 and STAT3 in the hypothalamus of C57BL/6, HFD, and ob/ob mice (Fig. 6A and B), which was largely attenuated in the db/db mice (Fig. 6C) and in the mice with ARC LepR knockdown (Fig. 6D). In vitro, ORM treatment induced dose- and time-dependent JAK2/STAT3 phosphorylation in GT1-7 cells (Fig. 6E and F), and the effects were inhibited by LepR knockdown (Fig. 6G and Supplementary Fig. 8A). Similarly, the dose- and time-dependent increase of JAK2/STAT3 phosphorylation induced by ORM in myoblast C2C12 cells was largely blocked when LepR was knocked down (Supplementary Fig. 8B–E).

A variety of biological activities have been attributed to proteins of the ORM family. It has been shown that cell membrane proteins CCR5, Siglect-5, and HBB can bind with ORM and mediate its action in vitro (24–26). However, the signal transduction pathways for these interactions and their biological significance in vivo have not been further explored. In the current study, we showed that ORM interacts with LepR specifically both in central and peripheral tissues and activates its downstream JAK-STAT3 signal pathway. We also found that ORM1-deficiency resulted in changes of energy homeostasis, characterized by adiposity, hyperleptinemia, hyperinsulinemia, hypercholesterolemia, and impaired glucose tolerance, whereas administration of exogenous ORM, both centrally and peripherally, reduced food intake and body weight and improved insulin resistance in a LepR-dependent manner. These data indicated that ORM could act as an agonist for LepR. Similar to leptin, ORM activates LepR to regulate food intake and energy homeostasis. Interestingly, the data also raised the question of why ORM binds to multiple cell membrane receptors. As suggested by the finding that the interaction with CCR5 was partially due to its glycosylation (24), it is likely that the heavy glycosylation of ORM might also contribute to its binding with LepR and other receptors.

The “negative feedback” model of energy homeostasis has been well established. Circulating signals inform the brain of changes in body fat mass. In response to this input, the brain mounts adaptive adjustments of energy balance to stabilize fat storage. Leptin and insulin are two well-known negative feedback signals that circulate at levels proportional to body fat content and can act on the brain to promote weight loss (1). Our data showed that ORM is regulated by nutritional conditions and activates the LepR pathway in the hypothalamus to affect food intake, indicating that it could be another negative feedback signal and could play an important role in energy homeostasis (Fig. 7). Because of the existence of hypothalamic ORM, the question of whether central ORM is also important in the regulation of metabolism was raised. We found that ORM2, not ORM1, is the major isoform of ORM in hypothalamus and that there was no response of central ORM to short- or long-term nutritional signals, indicating that it may not be responsible for ORM-induced change of energy homeostasis under these conditions.

In the present study, we found that in addition to the fat tissue, liver is also the main organ sensing long-term nutritional state, such as HFD and obese, to produce ORM. ORM is a dual signal both from adipose tissue and from gastrointestinal tract. Interestingly, it is the liver, not the fat tissue, that sensed a short-term nutritional signal, with ORM decreased after the fasting and increased after refeeding. The ORM derived from gastrointestinal tract is more important for the regulation of food intake in response to acute nutritional signals. This may be related to the specific regulation of liver in protein production. The liver can respond to various stresses and induce the quick production of acute-phase proteins, whereas ORM is one of the acute-phase proteins mainly produced by the liver (47). In addition, liver hepatic nuclear bile acid receptor farnesoid X receptor (FXR) may be involved in the quick regulation of ORM in response to an acute nutritional signal. FXR regulates bile acid homeostasis, while bile acid secretion is affected by immediate food intake (48). ORM is reported to be a direct target gene of hepatic FXR, and hepatic FXR deletion in mice affects the expression of ORM (48). Fasting and refeeding may promptly affect bile acid secretion and therefore decrease or increase ORM expression in liver via FXR. Interestingly, vertical sleeve gastrectomy (VSG) is at present the most effective therapy for the treatment of obesity, while the mechanism remains largely unclear. Recently, bile acids and FXR signaling have been reported to be important molecular targets for the effects of VSG. In the absence of FXR, the ability of VSG to reduce body weight and improve glucose tolerance is substantially reduced (49). Whether ORM, the downstream target of FXR, is involved in the role of VSG is worthy of being explored.

It should be noted that although ORM-induced phosphorylation of STAT3 was largely impaired when LepR deficient or knockdown, there still exists the activation of the STAT3 pathway, suggesting the involvement of other unknown receptor(s). Evidences showed that ORM might bind to the macrophages via CCR5 (24), which is also expressed in the hypothalamus and skeletal muscle. As a member of the cytokine receptors, CCR5 activation also results in STAT3 phosphorylation in T cells (50). In addition, CCR5 deficiency suppressed lung tumor development through the inhibition of nuclear factor-κB/STAT3 pathways (51). Therefore, the remaining phosphorylation of STAT3 triggered by ORM might be the result of CCR5 signaling. Interestingly, while we found that the effect of ORM on decreased serum insulin levels (most likely the result of reducing food intake and weight loss) is LepR dependent, Lee et al. (17) reported that ORM improved glucose tolerance (oral glucose tolerance test) and increased the sensitivity to insulin in decreasing blood glucose (insulin tolerance test) in db/db mice. These results indicate that ORM affects insulin action at least at two levels, controlling food intake and body weight with subsequent improvement of insulin resistance via the central LepR pathway and increasing peripheral insulin sensitivity in a LepR-independent pathway. Additional receptors such as CCR5 may also be involved in this process, since it was reported that CCR5 activation enhanced glucose uptake in activated T cells, and our previous studies found that ORM increased muscle glycogen content via the CCR5 pathway (32,52). One of our ongoing efforts is analyzing the ORM1 and LepR interaction through site-directed mutagenesis.

ORM also showed a variety of functions in peripheral tissues. It has been reported that ORM1 increases glucose uptake activity in 3T3-L1 adipocytes and relieves hyperglycemia-induced insulin resistance as well as tumor necrosis factor-α–mediated lipolysis in adipocytes (17). Accordingly, ORM improved glucose and insulin tolerance in obese and diabetic db/db mice and could protect adipose tissue from excessive inflammation and subsequent metabolic dysfunction (17). We further found that ORM also interacted with LepR in peripheral skeletal muscle and activated the JAK2-STAT3 signaling. Skeletal muscle, liver, and adipose tissue are the tissues with great metabolic activity and may constitute important targets for ORM in peripheral tissue and be involved in the regulation of insulin sensitivity as well as glucose and lipid metabolism. The function of ORM in peripheral tissues needs further investigation. Additionally, it has been shown that the level of ORM2 is increased under certain stress conditions (19). Therefore, the function of central ORM and other isoforms of ORM, which are not found to be affected by nutritional signals, remains to be further investigated.

In summary, we found that the expression of ORM, one of the acute-phase proteins, is regulated by long- or short-term nutritional signals and that ORM affects food intake and energy homeostasis through activating the LepR pathway in the hypothalamus. Thus, it is a negative feedback molecule in energy homeostasis and a novel target for the management of obesity and related metabolic disorders.

Acknowledgments. The authors thank Professor Xiao-Ying Li of Shanghai Clinical Center for Endocrine and Metabolic Diseases, Shanghai Jiaotong University School of Medicine (Shanghai, China) for the hypothalamic GT1-7 cells.

Funding. This study was supported by grants from the National Natural Science Foundation of China (81230083 to D.-F.S. and 81273606 and 81473259 to X.L.).

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Author Contributions. Y.S. wrote the manuscript. Y.Y., D.-F.S., and X.L. revised the manuscript. Y.S., Y.Y., D.-F.S., and X.L. designed the study. D.-F.S. and X.L. supervised the experiments. Y.S., Z.Q., and X.G. carried out most of the studies. J.C. and Y.T. carried out molecular modelling and docking analysis. X.G. carried out lentivirus injection experiments. Y.S. and X.G. bred and maintained ORM1 knockout mice. J.W. helped with some of the experiments. All authors analyzed and interpreted experimental data. X.L. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.