|Home | About | Journals | Submit | Contact Us | Français|
Leptin regulates energy homeostasis through central activation of multiple signaling pathways mediated by Ob-Rb, the long form of leptin receptor. Leptin resistance underlies the pathogenic development of obesity, which is closely associated with environmental factors. To further understand the physiological function of leptin signaling mechanisms, we generated a knock-in line of mice (Y985F) expressing a mutant Ob-Rb with a phenylalanine substitution for Tyr985, one of the three intracellular tyrosines that mediate leptin's signaling actions. Surprisingly, whereas young homozygous Y985F animals were slightly leaner, they exhibit adult-onset or diet-induced obesity. Importantly, both age-dependent and diet-induced deterioration of energy balance was paralleled with pronounced leptin resistance, which was largely attributable to attenuation of leptin-responsive hypothalamic STAT3 activation as well as prominently elevated expression of hypothalamic SOCS3, a key negative regulator of leptin signaling. Thus, these results unmask distinct binary roles for Try985-mediated signaling in energy metabolism, acting as an age/diet-dependent regulatory switch to counteract age-associated or diet-induced obesity.
As a component of the metabolic syndrome, obesity is closely associated with increased risk for the development of type 2 diabetes and cardiovascular disorders (16). Arising from a chronic imbalance between energy intake and expenditure, the pathogenic progression of obesity is attributable to the complex interactions between genetic factors and environmental influences. In mammals, energy balance is maintained through multiple homeostatic mechanisms that operate coordinately in response to hormonal and nutritional cues. Leptin is an adipose-secreted hormone (43) that plays a pivotal role in the regulation of energy metabolism. Acting through its active-form receptor Ob-Rb in distinct classes of leptin-responsive neurons (11, 14, 34), leptin activates multiple signaling pathways in the hypothalamus to regulate food intake and energy expenditure. Mice with deficiency in leptin (ob/ob) or its functional receptor (db/db) develop morbid obesity, hyperphagia, and diabetes (20). Impaired leptin responsiveness, i.e., leptin resistance (33), is a key characteristic of the metabolic defects that are responsible for disrupted energy control, presumably underlying the pathogenic development of human obesity (29). Although diminished leptin signaling has been found to occur in association with aging (21, 39) or feeding of a high-fat diet (HFD) (17, 18), the exact physiological mechanisms linking the environmental factors to the impairment in leptin-mediated regulation of energy metabolism remain largely elusive.
Leptin binds to Ob-Rb and elicits an array of subsequent intracellular signaling cascades (7, 22) via Jak2 phosphorylation. The mouse Ob-Rb comprises three cytoplasmic tyrosine residues, Tyr985, Tyr1077, and Tyr1138, which are known to be phosphorylated and mediate leptin's physiological functions (22, 26). The phosphorylated Tyr1138 is thought to recruit STAT3 (1), thereby activating the JAK2-STAT3 pathway, which has been shown to play an important role in the control of energy balance (2), whereas our recent investigation has demonstrated crucial actions in vivo for Ob-Rb tyrosine-dependent as well as tyrosine-independent mechanisms in the regulation of energy and glucose homeostasis (26). Our earlier studies in vitro as well as observations from other laboratories have also documented that phosphorylation at Tyr985 leads to recruitment of SH2-containing protein tyrosine phosphatase 2 (SHP2) (28, 41) and activation of extracellular signal-regulated kinase (ERK) (9). On the other hand, phosphorylated Tyr985 has been postulated to serve as a docking site for SOCS3, thereby exerting an antagonizing effect on Tyr1138-mediated STAT3 activation (8). Consistent with this, a leptin-activated autoinhibitory action in vivo has recently been suggested for Tyr985 in the l/l mice expressing a mutant leptin receptor where Tyr985 was replaced with leucine (10). However, whereas elevated hypothalamic expression of SOCS3 has been reported to occur in aged rodents (36) or in mice with diet-induced obesity (18, 42), it has yet to be understood whether Ob-Rb Tyr985-mediated mechanisms are physiologically connected to altered SOCS3 expression, particularly in the face of aging or high dietary fat intake. Moreover, direct in vivo evidence also has been lacking with respect to whether there exist potential interplays between Ob-Rb Tyr985 signaling and other Ob-Rb tyrosine-dependent mechanisms, which act to influence the homeostatic control of energy balance.
To gain further insight into the roles of Ob-Rb intracellular tyrosine phosphorylation in mediating leptin's physiological functions in vivo, we previously generated two lines of mice expressing mutant leptin receptors with phenylalanine substitution for all three tyrosines or for Tyr1138 alone, revealing the metabolic contribution of both tyrosine-dependent and -independent actions in energy homeostasis (26). Here we investigated the physiological consequences of abrogation of signaling through Ob-Rb Tyr985 via characterization of the knock-in mice generated by introducing a phenylalanine substitution mutation at this site. We examined the impact of deficiency in Tyr985-mediated signaling upon the susceptibility of mice to age-associated and diet-induced energy imbalance, attempting to explore the potential mechanistic links between leptin resistance and environmental influences such as aging and overnutrition.
The gene-targeting strategy for generation of the knock-in mice (Y985F) expressing the mutant leptin receptor Ob-RbF985 was carried out at the Transgenic Technology Center of UTSW as previously described in detail (26). Briefly, the targeting vector was constructed by introducing Lox-Neo-Lox (LNL), the lox-flanked neomycin resistance gene, downstream of the mutant exon 18 of the Ob-Rb gene, which contains the targeted substitution mutation at Tyr985 (Fig. (Fig.1A).1A). Stable homologous recombinant clones were selected with G418 from 129/Sv embryonic stem (ES) cells transfected with the targeting DNA construct, followed by confirmation of the Tyr-to-Phe substitution via sequencing analysis of the PCR-amplified exon 18-spanning DNA fragments. Blastocysts injected with the confirmed ES clones were subsequently transferred into pseudopregnant females to produce chimeric agouti mice, and successful gene targeting was further confirmed by Southern blotting. Germ line-transmitted knock-in mice were crossed with protamine-Cre mice, removing the LNL selection cassette from the male germ line only.
After deletion of LNL, heterozygous Y985F animals in the mixed 129Sv/C57BL/6 background were backcrossed to wild-type (WT) C57BL/6 mice (from Shanghai Laboratory Animals Co., Shanghai, China) for six to seven generations to yield mice >99% in the C57BL/6 background, which were then intercrossed to produce homozygous mice and WT littermates. The mutation was confirmed again by direct sequencing of Ob-Rb PCR products derived from tail genomic DNA. Mice were maintained under a 12-hour dark/light cycle (lights on at 6:30 a.m.) at a temperature of 22 ± 3°C in accredited animal facilities of the Shanghai Institute for Biological Sciences, CAS, and UT Southwestern Medical Center, with ad libitum access to standard chow and water. For HFD-induced obesity studies, mice at 5 weeks or 15 weeks of age were fed a diet containing 60% kcal fat (D12330; Research Diets Inc.). All experimental procedures and protocols were approved by the Institutional Animal Care and Use Committees at the Institute for Nutritional Sciences, SIBS, CAS, and UTSW.
Total body fat was measured by nuclear magnetic resonance (NMR) using a Minispec Mq7.5 Analyzer (Bruker, Germany). Food consumption was determined from mice individually caged by weighing food daily before the dark cycle for 5 to 10 days. Fast (6-hour fasting, 8:30 a.m. to 14:30 p.m.) blood was collected from the tail vein, and glucose was measured using a glucometer (FreeStyle, Alameda, CA). Levels of insulin and leptin were analyzed with the Mouse Serum Adipokine LINCOplex Kit (Linco Research, St. Charles, MO) using the Bio-Plex System (Bio-Rad, Hercules, California) according to the manufacturer's instructions.
Oxygen consumption and physical activity were determined for animals fed ad libitum using the comprehensive laboratory animal monitoring system (CLAMS; Columbus Instruments, Columbus, OH) according to the manufacturer's instructions. Mice were allowed to acclimate to the system for 16 to 20 h, and oxygen uptake (VO2) was measured for the following 24 h. Voluntary activity was monitored from the x axis beam breaks collected every 15 min.
Male mice at 7 or 40 weeks of age were injected intraperitoneally (i.p.) twice daily (at 8:30 a.m. and 6:00 p.m.) with phosphate-buffered saline (PBS) or recombinant mouse leptin (National Hormone and Peptide Program, UCLA, Torrance, CA) (1 mg/kg/day for 7-week-old mice and 2.5 mg/kg/day for 40-week-old mice). Mice were first injected with PBS for 3 days, followed by leptin injection for 3 to 4 days as indicated and then PBS injection for another 2 days. Body weight and food consumption were monitored daily during the treatment.
After fasting for 24 or 45 h, mice were anesthetized with sodium pentobarbital and then subjected to treatment with vehicle (PBS) or leptin (at 1 or 2 mg/kg) for 45 min by tail vein injection. For immunoblotting, dissected hypothalami were quickly frozen in liquid nitrogen, and protein extracts were prepared by homogenizing with radioimmunoprecipitation assay (RIPA) lysis buffer supplemented with 1% protease inhibitor cocktail (Sigma, St. Louis, MO). Protein extracts (~50 μg) were subjected to SDS-polyacrylamide gel electrophoresis, and the blots were probed first with the antibody against phospho-STAT3 (pTyr705) (Cell Signaling, Boston, MA) and then with the anti-STAT3 antibody (Cell Signaling, Boston, MA) after stripping. For immunohistochemistry (IHC), mice were sacrificed by cardiac perfusion via the ascending aorta with 10 ml of saline, followed by perfusion with 100 ml of 4% paraformaldehyde (Sigma Chemical Co., St. Louis, MO) in 0.1 M sodium phosphate buffer (pH 7.4). Brains were removed, incubated in the same fixative overnight, and then dehydrated in 30% sucrose. The brain block containing the hypothalamus was dissected on a Rodent Brain Matrix (ASI Instruments, Warren, MI) and then coronally sectioned (25 μm) with a sliding microtome (Leica Microsystems, GmbH, Wetzlar, Germany). Sections containing the arcuate nucleus of hypothalamus (ARC), ventromedial hypothalamic nucleus (VMH), and dorsomedial hypothalamic nucleus (DMH) regions were collected, and every fifth section was used for subsequent staining as described previously (18, 27). Free-floating sections were incubated at −20°C for 10 min in 1% NaOH-0.3% H2O2 in methanol and then sequentially in 0.3% glycine and 0.03% SDS for 10 min each. After being washed with PBS three times, sections were blocked with 3% BSA in PBS-0.25% Triton X-100-0.02% NaN3 for 1 h, followed by incubation with the antibody against phospho-STAT3 (pTyr705) overnight at 4°C. PBS-washed sections were incubated with Biotin-SP-conjugated goat anti-rabbit IgG (1:200; Invitrogen, Zymed) and then with streptavidin conjugated to horseradish peroxidase (1:1,000; Calbiochem). PBS-rinsed sections were subsequently developed in 0.05% 3,3′-diaminobenzidine (DAB) (Sigma-Aldrich, St. Louis, MO)-0.03% H2O2 in PBS, and images were recorded using an Olympus microscope (BX61) equipped with a DP70 digital camera. Sections ranging from Bregma −1.34 mm to −2.06 mm, which contain the ARC, VMH, and DMH nuclei, were chosen for quantitative measurement of STAT3 phosphorylation levels, with the third ventricle used as the landmark. The intensity of pSTAT3-positive signals was evaluated for each nucleus using the Image-Pro Plus software program (Media Cybernetics, Inc.).
Total hypothalamic RNA was isolated using TRIzol reagent (Invitrogen, Carlsbad, CA). After treatment with RNase-free DNase I (Roche Applied Science, Penzberg, Germany), first-strand cDNA was synthesized with Moloney murine leukemia virus (M-MLV) reverse transcriptase and random hexamer primers (Invitrogen). Real-time quantitative PCR was conducted with the ABI Prism 7500 sequence detection system, following the manufacturer's recommendations (Applied Biosystems, Foster City, CA). The oligonucleotide primers used for each target gene analyzed were as follows: for Ob-Rb, 5′-CCTCCAGGAGAGATGCTCACAC-3′ and 5′-TGACTGTGCGTGGAACAGGT-3′; for cyclophilin, 5′-ATGGCAAATGCTGGACCAAA-3′ and 5′-CATGCCTTCTTTCACCTTCCC-3′; for neuropeptide Y (NPY), 5′-TCCGCTCTGCGACACTACAT-3′ and 5′-GGCGTTTTCTGTGCTTTCCT-3′; for agouti-related protein (AgRP), 5′-GCTGTGTAAGGCTGCACGAG-3′ and 5′-TCCATTGGCTAGGTGCGACT-3′; for pro-opiomelanocortin (POMC), 5′-AAGATGCCGAGATTCTGCTACA-3′ and 5′-GGGCTGTTCATCTCCGTTG-3′; for SOCS3, 5′-GCGAGAAGATTCCGCTGGTA-3′ and 5′-TACTGATCCAGGAACTCCCGA-3′; for SHP2, 5′-CAAGTGCAACAATTCCAAACC-3′ and 5′-AACACGCATGACCCCGTA-3′; for PTP1b, 5′-TGGCCACAGCAAGAAGAAAA-3′ and 5′-GGAAAGGCAGGATCTCTCGA-3′; and for SH2B, 5′-GAAGCAGCCACAGGATCGTT-3′ and 5′-CCTTGCCCTGGAAGTTGA-3′.
All data are presented as means ± standard errors of the means (SEM). Statistical analysis of differences was done via the unpaired two-tailed Student's t test, one-way analysis of variance (ANOVA), or ANOVA with repeated measures as noted, with a P value of <0.05 considered statistically significant.
To minimize the possible disruption of the leptin receptor's structural functionality, we replaced Tyr985 with a phenylalanine within the exon 18 of Ob-Rb through homologous gene targeting (Fig. (Fig.1A),1A), which was confirmed by Southern blotting (Fig. (Fig.1B).1B). The knock-in mice, denoted Y985F, were subsequently obtained after backcrossing for more than six generations onto the C57BL/6 background. In contrast to their wild-type (WT) littermates, both male and female homozygous Y985F animals exhibited accelerated body weight gain with more rapid body fat accumulation, developing marked adult-onset obesity (Fig. (Fig.1C)1C) with >10% higher body fat mass after the age of 35 weeks for males and 30 weeks for females (Fig. 1D and E). Similar to the case for the previously reported l/l mice (10), however, Y985F homozygous animals at younger ages were slightly leaner than their WT counterparts until reaching comparable body weight and adiposity levels at the age of around 15 weeks (Fig. 1D and E). Notably, both male and female heterozygous +/Y985 mice displayed the same tendency of age-dependent increases in body fat mass, with intermediate degrees of adiposity observed at 51 weeks of age compared to WT and homozygous animals (Fig. 2A and B). Arguing against possible age-related differences in Ob-Rb expression levels, comparable hypothalamic Ob-Rb mRNA abundance was observed in homozygous Y985F males at 7 versus 55 weeks of age relative to WT littermates (Fig. (Fig.2C).2C). In parallel with the age-dependent changes in their adiposity levels, male homozygous Y985F mice exhibited hyperleptinemia (~3-fold increases in serum leptin levels) at the age of 39 weeks but displayed hypoleptinemia (~60% lower circulating leptin levels) at 7 weeks in comparison with age-matched WT animals (Table (Table1).1). Moreover, in association with their adiposity degrees, male Y985F mice at the age of 39 weeks had moderately but significantly increased serum levels of glucose and insulin in the fasted state in comparison with WT littermates (Table (Table1).1). Together, these results demonstrate a hitherto-unrecognized age-dependent role of Tyr985-mediated mechanisms in the control of body weight and fat mass.
We next investigated the impact of Tyr985 mutation within Ob-Rb upon feeding and energy expenditure. Compared to age-matched WT littermates, both male and female homozygous Y985F mice at 48 weeks of age exhibited considerably increased food intake (~8% for males and ~12% for females); whereas 8-week-old Y985F males consumed significantly less food (by ~17%) (Fig. (Fig.3A).3A). To compare the metabolic rates measured by the comprehensive laboratory animal monitoring system (CLAMS), we utilized the “metabolic mass” (25), BW0.75, of each animal for normalization to minimize the confounding effects that may arise from the different body sizes of obese mice. In comparison with WT animals, male Y985F mice, through a 12-hour light/dark cycle, showed markedly reduced oxygen consumption rates (11% and ~17% over the light and dark cycles, respectively) at 48 weeks of age and, conversely, significantly higher oxygen consumption rates (~10%) at 7 weeks of age (Fig. (Fig.3B).3B). Moreover, Y985F males at the age of 48 weeks exhibited a tendency of decreased activity, whereas at the age of 7 weeks they showed an ~2-fold increase in physical activity through the dark cycle (Fig. (Fig.3C).3C). Female Y985F mice at 50 weeks of age also displayed significantly lower oxygen consumption rates (~15%) as well as dramatically reduced activities (~50%), but they had comparable metabolic rates and physical activity at 7 weeks of age (data not shown). Thus, disruption of Tyr985 signaling resulted in age-dependent opposite alterations in food intake and energy expenditure, suggesting the existence of opposing binary modes of actions by Ob-Rb Tyr985-mediated mechanisms that operate as an age-responsive regulatory switch in the control of energy balance.
To determine whether the adult-onset obesity is accompanied by diminished leptin sensitivity, we measured the direct responses of homozygous Y985F animals to the body weight-reducing and appetite-suppressing effects of exogenous leptin administration. In comparison to WT animals, repeated intraperitoneal injection of leptin resulted in significantly less body weight reduction in 40-week-old male Y985F mice (Fig. (Fig.4A),4A), paralleled by insignificant appetite suppression (Fig. (Fig.4B),4B), whereas 7-week-old Y985F animals exhibited appreciably greater body weight reductions with similar decreases in food intake upon leptin treatment (Fig. 4A and B). Therefore, deficiency in Tyr985-mediated actions conferred adult-onset resistance to leptin with accompanying obesity and hyperleptinemia but increased leptin sensitivity at young ages with concurrent hypoleptinemia (Table (Table1),1), further supporting our hypothesis that Tyr985 signaling exerts age-dependent binary actions in mediating leptin's metabolic functions.
Because mutation of Tyr985 was reported to affect STAT3 signaling at longer times of stimulation in cultured cells expressing a chimeric fusion of erythropoietin receptor and Ob-Rb (8), we asked whether the age-associated onset of leptin resistance in Y985F mice is coupled to alterations in leptin-stimulated activation of the STAT3 pathway. In contrast to animals at the age of 7 weeks, in which leptin injection resulted in similarly robust stimulation of hypothalamic STAT3 phosphorylation in homozygous Y985F mice (Fig. (Fig.5A)5A) (~6.5-fold for Y985F versus ~9.0-fold for WT), Y985F mice at the age of 52 weeks showed prominently reduced leptin stimulation of STAT3 phosphorylation relative to their age-matched WT littermates (Fig. (Fig.5B)5B) (1.7-fold for Y985F versus ~6.4-fold for WT). Notably, 52-week-old Y985F mice had higher basal levels of STAT3 phosphorylation (~2.5-fold) than WT mice. This correlated with the hyperleptinemia observed in these mice (Table (Table1)1) and was consistent with disruption by Tyr985 mutation of the postulated autoinhibitory action upon STAT3 phosphorylation (10). Nonetheless, exogenous leptin failed to further augment the activation of STAT3, suggesting that attenuation of the leptin-activated STAT3 pathway as a result of loss of Tyr985 signaling may at least partially contribute to the observed adult-onset leptin resistance and energy imbalance in Y985F mice.
Leptin is known to inhibit the expression of orexigenic neuropeptide Y (NPY) and agouti-related protein (AgRP) while stimulating the expression of anorexigenic pro-opiomelanocortin (POMC) in the arcuate nucleus of hypothalamus (ARC) (12), which represent critical components of leptin's central regulatory circuitry. To further investigate the mechanistic basis of the age-dependent metabolic effects of abrogation of Tyr985 signaling, we measured the mRNA levels of hypothalamic NPY, AgRP, and POMC by quantitative RT-PCR. In the fed state, no significant changes were detected in homozygous Y985F females at 7 or 55 weeks of age (Fig. (Fig.6A).6A). Therefore, the observed alterations in the control of energy balance in young versus adult Y985F mice cannot be attributable to changes in these neuropeptides, indicating that Ob-Rb Tyr985-mediated mechanisms likely operate independently of regulating the NYP/AgRP and POMC neurons.
Several molecules have been implicated in either negatively or positively modulating leptin signaling transduction, among which are SOCS3 (a STAT3 target gene) (24, 31), SHP2 (28, 41), and PTP1B (3), as well as adaptor protein SH2B1, which enhances leptin signaling (38). To determine whether changes in the expression of these regulatory proteins contributed to the observed leptin resistance in adult Y985F animals, we assessed their hypothalamic expression levels by quantitative RT-PCR. Interestingly, among all the molecules tested, only the abundance of SOCS3 mRNA was markedly elevated (~2.5-fold) in 55-week-old homozygous Y985F mice compared to in age-matched WT littermates (Fig. (Fig.6B),6B), in accordance with the observed increases in the basal levels of STAT3 phosphorylation (Fig. (Fig.5B);5B); in contrast, comparable SOCS3 levels were detected in 7-week-old animals. Given SOCS3's ability to antagonize the STAT3-mediated signaling actions (6, 24), these data support an important role of age-dependent upregulation of SOCS3 expression in attenuating leptin's metabolic actions in adult obese Y985F mice.
Given the observation that Y985F animals developed adult-onset leptin resistance and impairment in energy balance, we wondered whether loss of Tyr985-mediated actions could affect the susceptibility of mice to overnutrition-induced obesity. We first examined the effect of a high-fat diet challenge upon adult Y985F mice after they had adiposity comparable to that of their WT counterparts. Male homozygous Y985F animals, maintained on a diet of normal chow (NC) until 15 weeks of age, showed dramatically increased sensitivity to feeding of a high-fat diet (HFD) (60% fat), weighing ~12 g more and accumulating ~17% more body fat than their WT littermates after 4 weeks (Fig. 7A and B). Moreover, Y985F mice fed the HFD for 13 weeks displayed higher degrees of hyperleptinemia (~9-fold versus ~7-fold increases) and hyperinsulinemia (~7-fold versus ~4-fold increases) than their WT counterparts (Table (Table2).2). Consistent with more severe diet-induced impairment in energy balance, Y985F mice fed the HFD for 9 weeks exhibited a significantly decreased metabolic rate (by ~10%) and physical activity (by ~25%) through the dark cycle relative to their WT littermates (Fig. 7C and D).
We next investigated whether a high dietary fat intake could accelerate the adult onset of obesity in Y985F mice. In contrast to the reported l/l mice, in which mutation of Tyr985 to a leucine residue led to protection of female animals against HFD-induced obesity at 5 to 12 weeks of age (10), male homozygous Y985F mice fed an HFD at the age of 5 weeks showed a sustained acceleration of body weight gain relative to mice fed NC, reaching body weights and adiposity levels at the age of 9 weeks that were similar to those in age-matched WT littermates (Fig. 8A and B). Moreover, HFD-fed Y985F mice showed a body weight gain exceeding that of WT mice by 7.4 to 11.3 g (Fig. (Fig.8A)8A) and accumulated ~13% more body fat (Fig. (Fig.8B)8B) at 15 to 19 weeks of age, displaying more pronounced diet-induced obesity (Fig. (Fig.8C).8C). Taken together, these results demonstrate that the Tyr985 signaling deficiency amplifies HFD-induced metabolic impairment in energy control, suggesting that Ob-Rb Tyr985 exerts important metabolic actions that counteract the effects of a high dietary intake of fat.
To further determine whether deficiency in Tyr985 signaling enhanced the impairment of HFD on energy metabolism through affecting the hypothalamic STAT3 pathway, we quantitated leptin-stimulated STAT3 phosphorylation in HFD-fed Y985F mice that exhibited pronounced obesity after 21 weeks of high-fat-diet feeding. In contrast to NC-fed Y985F animals, which showed robust STAT3 phosphorylation upon leptin treatment, HFD-fed Y985 animals exhibited dramatically blunted leptin-induced STAT3 activation (~3.0-fold [NC] versus ~1.5-fold [HFD] stimulation) with markedly increased basal STAT3 phosphorylation levels (Fig. (Fig.8D),8D), whereas in their WT littermates, HFD feeding resulted in only marginal attenuation of stimulation by leptin of STAT3 phosphorylation (~6.8-fold [NC] versus ~5.6-fold [HFD]). Consistently, analysis by quantitative immunohistochemistry of hypothalamic sections revealed that in both NC-fed Y985F and WT animals, leptin induced marked increases in the intensities of phospho-STAT3 signals within the arcuate nucleus (ARC) as well as the ventromedial hypothalamic nucleus (VMH) and dorsomedial hypothalamic nucleus (DMH) (Fig. 8E and F). In contrast, HFD feeding for 15 weeks dramatically decreased leptin's ability to stimulate STAT3 phosphorylation throughout all the three hypothalamic nuclei of Y985F mice relative to WT control mice (Fig. 8E and F), largely due to pronounced increases in the levels of phosphorylated STAT3 under basal conditions. Therefore, these data not only demonstrate that leptin responsiveness is prominently reduced in HFD-fed Y985F mice but also indicate that the metabolic effects of Tyr985 signaling are mediated through the leptin-responsive neurons located in multiple hypothalamic regions, deficiency of which may confer increased sensitivity to the deleterious effects of overnutrition upon energy homeostasis.
Similar to the case for the adult obese Y985F animals, quantitative RT-PCR assessment revealed no prominent alterations in the mRNA expression of AgRP, NPY, or POMC neuropeptides in the hypothalami of female Y985F mice fed the HFD for 15 weeks, which exhibited similarly exacerbated diet-induced adiposity relative to their WT littermates (Fig. 9A and B). On the other hand, in parallel with the observed HFD-induced elevation in the basal levels of STAT3 phosphorylation, HFD feeding in Y985F animals led to markedly increased hypothalamic mRNA expression of SOCS3 (by ~1.5-fold) in comparison with WT mice (Fig. (Fig.9C).9C). These results thus suggest that loss of Tyr985-medaited actions may confer increased susceptibility to age-dependent and diet-induced dysregulation of energy balance through similar mechanisms, whereby elevated expression of SOCS3 may contribute in large part to the occurrence of leptin resistance.
In the present study, we investigated the role in vivo of intracellular Tyr985 signaling of the leptin receptor in energy homeostasis in the face of aging and overnutrition. Abrogation of Tyr985-mediated actions of mouse Ob-Rb by introduction of a knock-in mutation, while conferring upon mice at young ages (e.g., 7 weeks) a lean phenotype with increased leptin sensitivity, led to adult-onset or diet-induced leptin resistance and obesity. Our findings demonstrate hitherto unappreciated binary modes of actions of Ob-Rb Tyr985 signaling as a regulatory switch, which can shift from an autoinhibitory action toward an age-dependent or diet-responsive positive role in energy balance, underscoring the physiological importance of interactions of leptin signaling mechanisms with environmental influences in energy metabolism.
Reported results from cell culture experiments have indicated that Ob-Rb Tyr985 phosphorylation may mediate multiple signaling actions: serving as a docking site for SOCS3, thereby antagonizing Tyr1138-mediated STAT3 activation (8), and/or directly binding SHP2 and thus stimulating ERK activation (9). Moreover, physiological studies of l/l knock-in mice at young ages (<12 weeks old) led the authors to postulate an autoinhibitory action by Tyr985-mediated signaling upon leptin's metabolic actions in energy balance (8, 10). Consistent with this, homozygous Y985F animals at 7 weeks of age were also slightly leaner than their WT littermates and displayed decreased food intake and increased energy expenditure as well as enhanced sensitivity to exogenous leptin. However, the occurrence of late-onset obesity in both male and female Y985F mice with apparent leptin resistance and marked impairment in energy balance clearly uncovers an age-dependent positive regulatory role for Tyr985 signaling in mediating leptin's functions. For reasons that are currently unclear, Y985F animals did not exhibit the sex-biased metabolic phenotypes reported for the l/l mice (10). In contrast to female l/l mice, which were shown to be leaner and resistant to diet-induced obesity, both male and female Y985F mice showed increased susceptibility to HFD-induced obesity, further supporting a positive action of signaling through Tyr985 in attenuating diet-induced impairment of energy metabolism. This discrepancy might result from the unknown effects of the different residue (leucine versus phenylalanine) substituted for Tyr985 upon the structural integrity of the receptor protein. On the other hand, while recent studies have implicated hypothalamic ERK activation in the regulation of food intake and body weight (37), we have been unable to detect age- or diet-dependent changes in leptin-responsive hypothalamic ERK activation as a result of Tyr985 signaling deficiency, largely due to the high basal levels observed in the hypothalamus for ERK phosphorylation. Thus, it remains unclear whether the SHP2-ERK pathway in leptin-responsive neurons is physiologically involved in mediating the metabolic actions through Tyr985 in energy balance, particularly in the context of aging or overnutrition.
Although the exact mechanism that causes the age-associated reversal of leptin sensitivity in Y985F animals remains to be elucidated, there emerged some interesting common features that accompanied the adult-onset and HFD-induced leptin resistance and energy imbalance due to disruption of Tyr985-signaling, i.e., markedly elevated basal phosphorylation levels in multiple hypothalamic regions of STAT3 that failed to be further activated in response to leptin stimulation, as well as the associated increases in the expression of SOCS3, which is known to be implicated in eliciting leptin resistance (5, 6, 8, 31). Therefore, in the face of aging or overnutrition, loss of Tyr985-mediated actions could cause similar molecular defects (e.g., SOCS3 upregulation as a result of chronic STAT3 activation in select hypothalamic neurons) that were responsible for the observed impairment in energy homeostasis. Our findings also suggest that leptin-responsive activation of Ob-Rb STAT3 signaling exerts important actions upon energy balance in young animals, but increased STAT3 activation in the long term may provoke increased adiposity at older ages with elevated levels of endogenous leptin (Table (Table1).1). This is in line with a recently reported study demonstrating through overexpression of a constitutively active form of STAT3 that chronically elevated basal STAT3 activation in the POMC neurons leads to obesity with attenuated leptin and insulin signaling due to increased SOCS3 expression (19). In addition, leptin is thought to act upon heterogeneous populations of Ob-Rb-expressing neurons, including those outside the ARC in the brain (34). For instance, it has been shown that disruption of leptin-induced signaling events specifically in the VMH is associated with leptin resistance (4, 15). Consistent with a previously reported study (32), we observed almost completely retained leptin stimulation of STAT3 phosphorylation in the VMH and DMH of HFD-challenged WT mice. In this context, it is worth noting that in the absence of Tyr985-mediated actions, HFD feeding completely abolished leptin responsiveness with respect to STAT3 phosphorylation not only in the ARC but also in the VMH and DMH of Y985F mice (Fig. (Fig.8).8). However, it currently remains unknown whether these nonarcuate neurons play an important role in integrating Ob-Rb Tyr985 signaling in response to overnutrition and thereby affecting the homeostatic control of energy balance.
Despite the fact that the precise mechanisms by which the increased abundance of SOCS3 in the hypothalamus disrupts leptin's metabolic actions have yet to be deciphered, SOCS3 may exert its antagonizing effects by blocking the nuclear or mitochondrial activities of STAT3 protein (23, 40), rather than by directly inhibiting STAT3 phosphorylation (5, 24). Moreover, deficient Tyr985 signaling led to adult-onset or diet-induced energy imbalance in Y985F mice in the absence of prominently dysregulated expression of hypothalamic NPY/AgRP or POMC neuropeptides. Thus, signaling through Tyr985 may be able to exert in vivo its metabolic effects through mechanisms independent of regulating the expression of these neuropeptides, especially acting through leptin-responsive neurons in hypothalamic regions aside from the ARC.
The leptin receptor employs a multitude of tyrosyl-phosphorylation-mediated signaling pathways to exert leptin's metabolic functions. Other key components, such as phosphatidylinositol 3-kinase (PI3K) (35), AMP kinase (AMPK) (30), and mTOR (13), are also thought to be involved in mediating leptin's central control of energy metabolism. However, how the complex Ob-Rb signaling mechanisms operate under various environmental influences remains poorly understood. Here, our results reveal the functional connection of one signaling arm of Ob-Rb through Tyr985 to the metabolic influences exerted by aging and overnutrition in the control of energy balance. Because leptin resistance is closely associated with the common form of human adult obesity in the settings of aging or chronic high-energy food intake, our findings unmask a potential pathogenic mechanism that may underlie the progressive development of obesity and type 2 diabetes.
We thank J. Repa (UTSW) for animal care and handling, J. Shen (INS, CAS) for assistance with CLAMS, and X. Y. Lu (UTHSC) for insightful discussions.
This work was supported by grants from the National Natural Science Foundation (no. 30988002, 30830033, and 90713027), the Ministry of Science and Technology (973 Programs 2006CB503900 and 2007CB947100), the Chinese Academy of Sciences (Knowledge Innovation Programs KSCX1-YW-02 and KSCX2-YW-R-115, CS Program SIBS2008006, and CAS/SAFEA International Partnership Program), and the Science and Technology Commission of Shanghai Municipality (no. 08dj1400601) to Y.L., W.L., and Z.K. and by NIH grant R01-DK60137 to C.L.
Published ahead of print on 19 January 2009.