Glutamate Mediates the Function of Melanocortin Receptor 4 on Sim1 Neurons in Body Weight Regulation

Cell Metabolism

Article Glutamate Mediates the Function of Melanocortin Receptor 4 on Sim1 Neurons in Body Weight Regulation Yuanzhong Xu,1 Zhaofei Wu,1 Hao Sun,1 Yaming Zhu,1 Eun Ran Kim,1 Bradford B. Lowell,2 Benjamin R. Arenkiel,3 Yong Xu,4 and Qingchun Tong1,5,6,* 1Brown

Foundation Institute of Molecular Medicine, University of Texas Health Science Center at Houston, Houston, TX 77030, USA of Endocrinology, Department of Medicine, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, MA 02215, USA 3Departments of Molecular & Human Genetics and Neuroscience, Jan and Dan Duncan Neurological Research Institute at Texas Children’s Hospital, Baylor College of Medicine, Houston, TX 77030, USA 4Children’s Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, One Baylor Plaza, Houston, TX 77030, USA 5Division of Endocrinology, Department of Internal Medicine, and Department of Neurobiology and Anatomy, the University of Texas Medical School at Houston, Houston, TX 77030, USA 6Programs in Neuroscience and Biochemistry, Graduate School of Biological Sciences, the University of Texas Health Science Center at Houston, Houston, TX 77030, USA *Correspondence: [email protected] http://dx.doi.org/10.1016/j.cmet.2013.11.003 2Division

SUMMARY

The melanocortin receptor 4 (MC4R) is a well-established mediator of body weight homeostasis. However, the neurotransmitter(s) that mediate MC4R function remain largely unknown; as a result, little is known about the second-order neurons of the MC4R neural pathway. Single-minded 1 (Sim1)expressing brain regions, which include the paraventricular nucleus of hypothalamus (PVH), represent key brain sites that mediate melanocortin action. We conditionally restored MC4R expression in Sim1 neurons in the background of Mc4r-null mice. The restoration dramatically reduced obesity in Mc4r-null mice. The anti-obesity effect was completely reversed by selective disruption of glutamate release from those same Sim1 neurons. The reversal was caused by lower energy expenditure and hyperphagia. Corroboratively, selective disruption of glutamate release from adult PVH neurons led to rapid obesity development via reduced energy expenditure and hyperphagia. Thus, this study establishes glutamate as the primary neurotransmitter that mediates MC4Rs on Sim1 neurons in body weight regulation.

INTRODUCTION The obesity epidemic has imposed a major social and economic burden on our society. As such, this epidemic demands a clear understanding of its mechanistic cause. Over the past decades, a large body of evidence has established the importance of the melanocortin pathway in body weight regulation. Inactivation of the melanocortin 4 receptor (Mc4r) gene produces massive

obesity in both rodents and humans (Farooqi and O’Rahilly, 2005; Huszar et al., 1997), suggestive of underlying common neural pathways that regulate body weight between these two species. MC4R-expressing neurons, which mediate the effects of a-melanocyte-stimulating hormone (a-MSH) released from proopiomelanocortin neurons and agouti-related protein (AgRP) released from AgRP neurons, have been directly linked to feeding behavior and energy expenditure (Cone, 2005; Elmquist et al., 2005). However, although there is a well-established role for MC4Rs in obesity development, the neurotransmitter(s) that mediate the function of MC4Rs remain unclear, and as a result, little is known about the second-order neurons that mediate the function of MC4R-expressing neurons. The paraventricular nucleus of hypothalamus (PVH) is known to control body weight and express abundant MC4Rs and has been identified as a major brain site of melanocortin action (Balthasar et al., 2005). Interestingly, hypomorphism of singleminded 1 (Sim1), a transcription factor required for PVH development, produces obesity, and genetic deletion of Sim1 selectively in the PVH leads to obesity (Michaud et al., 2001; Tolson et al., 2010). Notably, while MC4Rs are broadly expressed in the brain, selective MC4R restoration in Sim1 neurons (PVH and parts of the amygdala) greatly rescues the obesity in Mc4r-null mice (Balthasar et al., 2005; Kishi et al., 2003; Liu et al., 2003), suggesting a major role for MC4R function in Sim1 neurons toward body weight regulation. Consistent with the cell-type-selective function of MC4Rs in Sim1 neurons, MC4R restoration in other brain areas, such as hindbrain and spinal cord, showed little effect on obesity in Mc4r-null mice (Rossi et al., 2011; Sohn et al., 2013). Thus, despite broad expression of Mc4r in the brain, Sim1 neurons are one key site that mediates MC4R action on body weight regulation. Therefore, identification of the neurotransmitter(s) and signaling mechanisms that mediate MC4R function in Sim1 neurons is critical to delineate the MC4R neural pathway in body weight regulation. The PVH contains diverse groups of neurons that use peptides, glutamate, GABA, or dopamine as neurotransmitters

860 Cell Metabolism 18, 860–870, December 3, 2013 ª2013 Elsevier Inc.

Cell Metabolism Glutamate Mediates MC4R Function

Figure 1. Colocalization of Vglut2 and Mc4r in PVH (A–D) Shown is Vgat (A), Vglut2 (B), and Mc4r (C) mRNA expression by in situ hybridization in the PVH; Vglut2, but not Vgat, is largely colocalized with Mc4r (D). Scale bars: 50 mm. See also Figure S1.

RESULTS

(Cowley et al., 1999; Shi et al., 2013; Swanson and Sawchenko, 1980; Xu and Tong, 2011). Among those peptides released from PVH neurons are corticotrophin-releasing hormone (CRH), thyrotrophin-releasing hormone (TRH), oxytocin, and vasopressin (Swanson and Sawchenko, 1980). In addition to direct regulation of the autonomic nervous system, these peptides also regulate distinct endocrine functions through the hypothalamic pituitary axes (Swanson and Sawchenko, 1980). Although substantial evidence supports a role for these peptides in the regulation of energy balance (Kublaoui et al., 2008; Morton et al., 2012; Vella et al., 2011; Zhang et al., 2011), mice with loss of function of each of these neuropeptides exhibited few or no defects in body weight regulation (Jacobson, 1999; Vollmer et al., 2006; Wu et al., 2012; Yamada et al., 1997), suggesting a possible permissive role for neuropeptides within the PVH or, alternatively, adaptation to their developmental inactivation. GABA and dopamine have also been implicated in the PVH to mediate the melanocortin action and neuropeptide Y (NPY), respectively, in energy balance regulation (Cowley et al., 1999; Shi et al., 2013).Thus, similarly to AgRP neurons, both neuropetides and fast-acting neurotransmitters from PVH neurons mediate energy balance. Interestingly, GABA and NPY released from AgRP neurons functionally compensate for each other in mediating AgRP neuron action in feeding (Krashes et al., 2013). However, whether, and the extent to which, neuropeptides or neurotransmitters mediate MC4R action in the PVH is unknown. In line with this idea, we have identified that glutamate mediates the action of MC4Rs on Sim1 neurons in body weight regulation and that PVH neuropeptides are not capable of compensating for loss of glutamate release from Sim1 neurons.

MC4R-Bearing Sim1 Neurons Are Glutamatergic Vesicular glutamate transporter 2 (Vglut2, also known as Slc17a6), a gene expressed in glutamatergic neurons for loading glutamate in vesicles for presynaptic release, is abundant in the hypothalamus (Fremeau et al., 2004; Tong et al., 2007). To assess the extent to which VGLUT2 (the protein encoded by Vglut2) and MC4Rs colocalize in Sim1-expressing neurons, we performed RNAscope analysis, which can detect up to three distinct species of mRNA transcript simultaneously (Wang et al., 2012). We found that Vglut2 and Mc4r were abundantly expressed in the PVH (Figures 1B and 1C, dashed line regions) and the nucleus of lateral olfactory tract (NLOT; Figure S1 available online), suggesting that most neurons in these brain regions are glutamatergic and express Mc4r. In contrast, vesicular GABA transporter (Vgat, also known as Slc32a1, required for loading GABA in vesicles for presynaptic release) was only detected in a few neurons in the PVH (Figure 1A) and rarely detected in the NLOT (Figure S1). Moreover, Mc4r and Vglut2 signals were largely associated with each other within a given neuron in the PVH (Figure 1D) and NLOT (Figure S1). To quantify the extent to which Mc4r and Vglut2 colocalize, we chose to analyze 3 sections 20 mM thick with 120 mM apart, ranging from bregma 0.58 mm to bregma 0.82 mm of the mouse brain where both PVH and NLOT are located. Counting six sections from two mice showed that roughly 98.4% ± 6.2% of MC4R-expressing neurons were positive for Vglut2 signal, 90.2% ± 8.4% of VGLUT2-expressing neurons were positive for Mc4r signal in the PVH, and Mc4r and Vglut2 colocalized nearly 100% in the NLOT. These results suggest that VGLUT2, but not VGAT, is colocalized with MC4Rs in Sim1 neurons. Thus, within Sim1 brain regions, MC4R-expressing neurons are largely glutamatergic, but not GABAergic. Developmental Deletion of VGLUT2 in Sim1 Neurons Leads to Loss of Glutamate Release and Mild Obesity To examine the role of glutamate release from PVH neurons, we generated mice with conditional excision of Vglut2 selectively from Sim1 neurons (Sim1-Cre:Vglut2flox/flox mice). Whereas PVH neurons in control mice showed abundant Vglut2 expression (Figure 2A), in Sim1-Cre:Vglut2flox/flox conditional mice, Vglut2 mRNA signal was eliminated (Figure 2B inside the dashed area). Only a small region above the third ventricle showed

Cell Metabolism 18, 860–870, December 3, 2013 ª2013 Elsevier Inc. 861

Cell Metabolism Glutamate Mediates MC4R Function

Figure 2. Disruption of Glutamate Release from Sim1 Neurons Affects Body Weight (A–C) Selective deletion of Vglut2 in Sim1 neurons. In situ hybridization against Vglut2 transcripts in the PVH of Vglut2flox/flox mice (A) and Sim1Cre:Vglut2flox/flox mice (B). Vglut2 mRNA is absent in the PVH of Sim1-Cre:Vglut2flox/flox mice, but remains at sites immediately above the PVH (arrows in A and B). Sim1-Cre expression is shown by red fluorescent protein expression in Sim1Cre:Ai9 mice (C). The deletion pattern of Vglut2 in the PVH is consistent with Sim1-Cre expression pattern. (D) KCl-induced glutamate release from PVH sections micropunched from Vglut2flox/flox mice and Sim1-Cre:Vglut2flox/flox mice (n = 4–5). Data showing ratios of KCl versus base levels were presented. (E and F) Weekly body weight of Vglut2flox/flox and Sim1-Cre:Vglut2flox/flox 4- to 20-week-old male mice (n = 7–12) (E) and female mice (n = 14–15) (F). PVH, paraventricular hypothalamus; 3V, the third ventricle. Data are represented as mean ± SEM. Scale bars: 100 mm. *p < 0.05, analyzed by Student’s t test in (D) and one-way ANOVA in (E) and (F). See also Figure S2.

residual signal (Figure 2B, arrows), which reflected the known pattern of Sim1-Cre activity as indicated by robust DsRed fluorescence in the PVH of Sim1-Cre:Ai9 reporter mice (Figure 2C). In addition, the NLOT showed restricted expression of Sim1Cre, and in Sim1-Cre:Vglut2flox/flox mice, Vglut2 signal was eliminated in this region (Figure S2). Taken together, these data show that Vglut2 was selectively deleted in Sim1 neurons by Sim1-Cre. To verify that VGLUT2 deletion resulted in the loss of glutamate release, we directly measured glutamate release from micropunched PVH sections using a previously established method (Enriori et al., 2007; Kajimoto et al., 2007). In this assay, we used potassium chloride (KCl) to stimulate a nearly 2.5-fold increase of glutamate release in control PVH sections, whereas it only showed a 1.3-fold increase in PVH sections from Sim1Cre:Vglut2flox/flox mice (Figure 2D). The remaining 1.3-fold increase presumably resulted from non-Sim1 glutamatergic neurons (i.e., non-PVH tissues), which were present in the punched PVH sections and still capable of releasing glutamate. Together with the observed deletion of Vglut2 in the PVH (Figure 2B), these data demonstrate the efficacy of our conditional deletion model of VGLUT2 in Sim1 neurons to probe neurotransmitter function toward body weight control. To examine the effect of deficient glutamate release from Sim1 neurons, we monitored weekly body weight in our conditional

mouse models. Surprisingly, compared to littermate controls, male Sim1-Cre: Vglut2flox/flox mice showed mild lateonset obesity (Figure 2E), whereas females showed no signs of abnormal weight gain (Figure 2F). These data suggest, at least in females, that either glutamate release from Sim1 neurons is not required for body weight regulation or potentially redundant and/or gender-specific pathways exist to compensate for developmental disruption of glutamate release in embryonic Sim1 neurons. Deletion of VGLUT2 in Adult PVH Sim1 Neurons Leads to Rapid Obesity Development To address possible confounding factors associated with disrupting glutamate release early in development, we next conditionally deleted VGLUT2 in PVH neurons of 8- to 10week-old Vglut2flox/flox females through stereotaxic injection of adeno-associated viral vectors engineered to express Cre and GFP (AAV-Cre-GFP). Previous studies suggested that in Sim1Cre mice, nearly all PVH neurons express Cre recombinase (Kublaoui et al., 2008), and our results also verified efficient deletion of Vglut2 in the PVH by Sim1-Cre (Figure 1). Thus, as expected, bilateral delivery of AAV-GFP (control) (Figure 3A) or AAV-CreGFP (Figure 3B) to the PVH of Vglut2flox/flox mice showed robust targeted infection, and while Vglut2 was abundantly expressed in the PVH with control AAV-GFP injections (Figure 3C), it was dramatically reduced in the PVH following AAV-Cre-GFP injection (Figure 3D). The body weights of conditional knockout mice showed a rapid increase of 13 g within 4 weeks postinjection, whereas AAV-GFP-injected controls only gained 1 g during the same period (Figure 3E), showing the importance of

862 Cell Metabolism 18, 860–870, December 3, 2013 ª2013 Elsevier Inc.

Cell Metabolism Glutamate Mediates MC4R Function

Figure 3. Bilateral Deletion of Vglut2 in Female PVH Leads to Massive Obesity (A–D) Bilateral injections of either AAV-Cre-GFP or AAV-GFP were made in the PVH of 8- to 10-week-old Vglut2flox/flox females. GFP signal showing bilateral delivery and expression of AAV-GFP (A) and AAV-Cre-GFP (B) shown in representative PVH sections. Vglut2 in situ signal in the PVH of Vglut2flox/flox mice that received bilateral AAV-GFP (C) and AAV-Cre-GFP (D). (E) Weekly body weights for a period of 7 weeks following AAV vector injections. (F) Measurements of O2 consumption following AAV vector injection into Vglut2flox/flox mice. Measurements were made 4 weeks postinjection. The O2 consumption data were analyzed per individual animal. (G) Accumulated food intake of AAV vector-injected Vglut2flox/flox mice measured from 2–4 weeks following injections. Data are presented as mean ± SEM; n = 4–7. *p < 0.05 and ***p < 0.001, analyzed by Student’s t tests. Scale bars: 100 mM. See also Figure S3.

glutamate release from PVH Sim1 neurons toward body weight regulation. In these animals, increased body weight was associated with a significant increase in fat mass, typical of an obesity phenotype (Figure S3). Interestingly, the observed obesity phenotype was also accompanied by a slightly increased lean mass (Figure S3), similar to that seen in Mc4r-null mice. To evaluate other phenotypes following conditional loss of glutamate release, we also monitored readouts of altered metabolism. First, we assessed energy expenditure by measuring O2 consumption at 3 weeks following AAV vector delivery. Mice with AAV-Cre-GFP injections exhibited a significant reduction in O2 consumption as measured per individual (Figure 3F) or normalized to lean mass (Figure S3). Interestingly, mice with AAVCre-GFP injections also exhibited a significant reduction in locomotion, especially during the dark period (Figure S3). Interestingly, food intake measured during the same period exhibited an initial mild and statistically insignificant increase followed by a more pronounced and significant increase with time (Figure 3G). This observation suggests that changes in food intake may be secondary to increased body weight. Taken together, these data highlight an important role for glutamate release from PVH Sim1 neurons in regulating energy expenditure. A Conditional Approach for Concurrent MC4R Restoration and VGLUT2 Deletion in Sim1 Neurons The colocalization of MC4Rs and VGLUT2 in Sim1 neurons makes it feasible to specifically investigate the role of MC4Rs in controlling glutamate release only from these neurons. To

determine if MC4Rs act cell autonomously to regulate body weight control through glutamate release from PVH neurons, we generated mice with restored expression of MC4Rs selectively in Sim1 neurons while at the same time removing VGLUT2. To this end, we utilized a line of knockin mice in which a DNA fragment that harbors loxP-flanked transcription blockers (tb) had been inserted into the endogenous promoter region of Mc4r, referred to as Mc4rtb/tb mice. Since normal transcription is blocked at the promoter in these mice, Mc4rtb/tb mouse models are functionally null. However, in the presence of Cre recombinase the floxed transcriptional blocker is excised, and normal Mc4r expression is conditionally restored in a Credependent manner (Balthasar et al., 2005). These mice were first crossed with Sim1-Cre mice to restore MC4Rs in Sim1 neurons. We then introduced the Vglut2flox/flox allele to eventually generate Sim1-Cre:Vglut2flox/flox, Sim1-Cre:Mc4rtb/tb, Vglut2flox/flox, Mc4rtb/tb, and Sim1-Cre:Mc4rtb/tb:Vglut2flox/flox experimental mice. To evaluate the efficacy of our system, we first performed Mc4r in situ hybridization against Mc4r mRNA in Sim1 neurons of control animals. Mc4r was abundant in the PVH, NLOT, and hindbrain (Figure S4), which was consistent with earlier reports (Kishi et al., 2003). Moreover, confirming previous results (Balthasar et al., 2005), Mc4r was not detectable in any of these brain sites of Mc4rtb/tb mice (Figure S4). Finally, in Sim1-Cre:Mc4rtb/tb mice, Mc4r mRNA signal was strongly and selectively restored in the PVH (Figure 3G), NLOT, and supraoptic nucleus (SON) to levels similar to those of control mice (Figure S4). Taken

Cell Metabolism 18, 860–870, December 3, 2013 ª2013 Elsevier Inc. 863

Cell Metabolism Glutamate Mediates MC4R Function

Figure 4. Glutamate Mediates MC4R Action on Sim1 Neurons in Body Weight Regulation (A–D) Shown are the weekly body weight measurements of males (A) and their average daily food intake measured at the age of 19–20 weeks (n = 7–20) (B) as well as the weekly body weight of females (C) and their average daily food intake measured at the age of 19–20 weeks (n = 14–29) (D). Sim1-Cre:Mc4rtb/tb:Vglut2flox/flox mice showed identical body weight to that of Mc4rtb/tb mice at all time points measured in both males (A) and females (C). Sim1-Cre:Mc4rtb/tb:Vglut2flox/flox mice and Mc4rtb/tb mice showed the same level of hyperphagia as to other genotypes (B and D). All data are presented as mean ± SEM; *p < 0.05, one-way ANOVA test. See also Figure S4.

together, these results show that in our model, Mc4r expression can be selectively restored in Sim1 neurons in a Cre-dependent manner. Thus, Sim1-Cre:Mc4rtb/tb:Vglut2flox/flox mice show efficient and concurrent restoration of MC4Rs and deletion of VGLUT2 in Sim1 neurons. Glutamate Release Mediates the Action of MC4Rs on Sim1 Neurons in Body Weight Regulation We monitored weekly body weight in both males and females of five different genotypes that included Vglut2flox/flox, Sim1Cre:Vglut2flox/flox, Mc4rtb/tb, Sim1-Cre:Mc4rtb/tb, and Sim1Cre:Mc4rtb/tb:Vglut2flox/flox mice (Figures 4A and 4C). Consistent with previous data (Balthasar et al., 2005), MC4R restoration in Sim1 neurons (Sim1-Cre:Mc4rtb/tb mice) greatly reduced obesity of Mc4rtb/tb mice across all time points (Figures 4A and 4C). Interestingly, in both males (Figure 4A) and females (Figure 4C), Sim1-Cre:Mc4rtb/tb:Vglut2flox/flox mice showed body weight identical to that of Mc4rtb/tb mice at all time points measured, suggesting that disruption of glutamate release from Sim1 neurons completely reversed the body weightrescuing effects of MC4R restoration in Sim1 neurons of Mc4rnull mice. This effect is particularly striking in females, since deletion of Vglut2 alone in these mice showed no defects in body weight (Figures 1, 4A, and 4C). To examine the mechanism underlying the body weight phenotype, we measured daily food intake in adult mice (20 weeks old) of all experimental genotypes. As expected, Mc4rtb/tb mice exhibited hyperphagia, which was reduced in both genders of Sim1-Cre:Mc4rtb/tb mice to a level comparable to that of controls (Figures 4B and 4D). Strikingly, however, Sim1-Cre:Mc4rtb/tb:Vglut2flox/flox mice exhibited the same degree of hyperphagia as Mc4rtb/tb mice (Figures 4B and 4D), suggesting that disruption of glutamate release completely reversed the reduced food consumption effects of MC4R restoration in Sim1 neurons of Mc4rtb/tb mice. Taken together, these

results suggest that glutamate release mediates the body weight- and food intake-regulating effects of MC4Rs on Sim1 neurons. To avoid secondary effects that might be observed in already obese adult Mc4rtb/tb mice (Butler and Kozak, 2010; Tscho¨p et al., 2012), we measured food intake in 3- to 4-weekold female mice with minimal obesity development. Remarkably, the average daily food intake was not different among genotypes (Figure 5A). However, when measured during the same period, both Mc4rtb/tb and Sim1-Cre:Mc4rtb/tb:Vglut2flox/flox mice showed a significant increase in net body weight (Figure 5B), suggesting that the primary defect may be due to decreased energy expenditure. Indeed, both Mc4rtb/tb mice and Sim1Cre:Mc4rtb/tb:Vglut2flox/flox mice showed a significant reduction in O2 consumption at the age of 4–5 weeks, compared to control Vglut2flox/flox mice. This finding was consistent when the O2 data were either expressed per individual (Figure 5C) or normalized to lean mass (Figure S5). Interestingly, we observed a higher level of O2 consumption in Sim1-Cre:Mc4rtb/tb, than in Mc4rtb/tb mice (Figure 5C), arguing that the effects of body weight reduction by MC4R restoration in Sim1 neurons of Mc4r-null mice may be partially mediated by increased energy expenditure. Notably, the difference in O2 consumption among genotypes is not caused by initial body weight difference, since the starting weights of these mice were similar when O2 consumption was measured (Figure 5D). Similar results were also observed in males (Figure S5), except that male Sim1-Cre:Vglut2flox/flox mice showed reduced O2 consumption compared to control Vglut2flox/flox mice (data not shown). In addition, the respiratory exchange ratio, an indicator for preference in metabolizing carbohydrate or fat, was not different during day or night periods among genotypes (Figure S5). Energy Expenditure Response to MTII Whether MC4Rs on Sim1 neurons regulate energy expenditure remains controversial (Balthasar et al., 2005; Schneider, 2009; Skibicka and Grill, 2009). To ascertain the effect of glutamate release from Sim1 neurons on energy expenditure, we directly measured changes of O2 consumption in response to acute pharmacological action of melanotan II (MTII), an MC4R agonist,

864 Cell Metabolism 18, 860–870, December 3, 2013 ª2013 Elsevier Inc.

Cell Metabolism Glutamate Mediates MC4R Function

Figure 5. Effects on Food Intake and Energy Expenditure by Disruption of Glutamate Release in Young Mice (A–D) Shown is the average daily food intake (A) and net body weight increases (B) in 3- to 4-weekold females (n = 8-21). Female O2 consumption expressed per individual animal (C) and body weight (D) of the same cohort of mice used for measuring O2 consumption (n = 5–12). All data are presented as mean ± SEM; *p < 0.05, one-way ANOVA test. See also Figure S5.

in 4- to 5-week-old female mice. As expected, MTII greatly increased O2 consumption in control Vglut2flox/flox mice but had no effect in Mc4rtb/tb mice (Figures 6A and 6B). Interestingly, compared to Mc4rtb/tb mice, MTII also significantly increased O2 consumption in Sim1-Cre:Mc4rtb/tb mice, suggesting that MC4Rs on Sim1 neurons promote energy expenditure (Figures 6A and 6B). Strikingly, like Mc4rtb/tb mice, Sim1-Cre:Mc4rtb/tb: Vglut2flox/flox mice exhibited no change in energy expenditure to MTII (Figures 6A and 6B), suggesting that disruption of glutamate release reverses the effect of MC4R restoration in Sim1 neurons on energy expenditure. Consistent with data from females, similar responses in O2 consumption in response to MTII were also observed in males (Figure 6C), except that male Sim1-Cre:Vglut2flox/flox mice showed a significantly blunted response in O2 consumption compared to baseline controls, which was consistent with the mild obesity phenotype in these mice (Figure 2). Finally, consistent with their body weight, fat mass in Sim1Cre:Mc4rtb/tb:Vglut2flox/flox mice was increased to a level similar to that of Mc4rtb/tb mice in both males and females (Figure S6). Compared to control Vglut2flox/flox mice, fat mass in Sim1-Cre: Vglut2flox/flox mice was significantly increased in males, but not in females (Figure S6), which is in line with their body weight changes (Figure 1). Glutamate Mediates MC4R Action on Body Length Control in Females, But Not in Males Interestingly, in addition to the noted obesity phenotype, it has been reported previously that Mc4r-null mice also exhibit increased body length and associated lean mass (Balthasar et al., 2005). The mechanism underlying the increased body length of Mc4r-null mice is unknown. Strikingly, despite the similar degree of obesity compared to Mc4rtb/tb mice,

male Sim1-Cre:Mc4rtb/tb:Vglut2flox/flox mice showed no reversal in lean mass or body length, both of which remained at levels similar to those of Sim1Cre:Mc4rtb/bt mice (Figures 7A and 7C). In striking contrast, however, female mice Sim1-Cre:Mc4rtb/tb:Vglut2flox/flox showed a complete reversal and showed levels of lean mass and body length equivalent to those of Mc4rtb/tb mice (Figure 7B and 7D). These data suggest that, independent of its role toward body weight regulation, glutamate mediates MC4R action on body length control in females, but not in males. DISCUSSION Despite the proven importance of MC4Rs toward the development of obesity in both rodents and humans, the neurotransmitter(s) that mediate their action are not clear. As a result, the second-order neurons that mediate the action of MC4Rexpressing neurons in this circuit are unknown. Given the broad expression of MC4Rs throughout the brain, here we focused on the functional role of glutamate signaling specifically from Sim1 neurons. We took interest in these cells due to their abundant Mc4r expression, known roles in body weight control, and the Mc4r-null obesity phenotype (Balthasar et al., 2005; Skibicka and Grill, 2009). To specifically examine the role of glutamate release in mediating MC4R function in Sim1 neurons, we generated a mouse model that allowed concurrent restoration of MC4Rs while at the same time disrupting glutamate release selectively from Sim1 neurons. This unique model allowed us to address whether, and the extent to which, glutamate release mediates MC4R activity on Sim1 neurons in body weight regulation. Consistent with previous data (Balthasar et al., 2005), the restoration of MC4R function selectively in Sim1 neurons reduced the obesity of Mc4r-null mice. Interestingly, the reduction of obesity by MC4R restoration was reversed by concomitant disruption of glutamate release from the same Sim1 neurons. The reversal was complete, and the resulting obesity was indistinguishable from that of Mc4r-null mice. We reasoned that this reversal was specific to MC4Rs on Sim1 neurons since (1) MC4Rs are largely colocalized with VGLUT2 in Sim1 neurons and (2) the reversing effect on body weight perfectly counters the

Cell Metabolism 18, 860–870, December 3, 2013 ª2013 Elsevier Inc. 865

Cell Metabolism Glutamate Mediates MC4R Function

Figure 6. Effects on O2 Consumption by MTII Injection (A and B) Response in O2 consumption to MTII injection in females. (A) Traces of O2 consumption of five experimental genotypes of female mice in response to saline or MTII injection (n = 5–12). Mice (4–5 weeks old) with food removed were i.p. injected at 12:00 pm with saline on day 1 and MTII (200 mg/mice) on day 2, followed by measurement of O2 consumption. (B) The increase in O2 consumption was calculated based on changes induced by MTII relative to saline injections for a 1.5 hr period after saline or MTII injection. (C) Response in O2 consumption to MTII injection in males (n = 5–9). All data are presented as mean ± SEM; *p < 0.05, one-way ANOVA test.

rescuing effect by MC4R restoration. It is important to point out that the complete reversal of body weight by disruption of glutamate release does not affect the expression of known peptides (e.g., neuropeptides in PVH and SON neurons; data not shown), suggesting that neuropeptides alone are not capable of compensating for loss of glutamate release from Sim1 neurons. This is in contrast to GABA release from AgRP neurons in which GABA and NPY could compensate each other in mediating the feeding behavior of AgRP neurons (Krashes et al., 2013). Thus, we have identified glutamate as the key neurotransmitter that mediates the action MC4Rs on Sim1 neurons in body weight regulation. Future studies that target glutamate signaling from these neurons should allow the identification of direct downstream target neurons that mediate MC4R action on Sim1 neurons, as previously shown for GABA release from AgRP neurons (Atasoy et al., 2012; Wu et al., 2009). Together, our data from AAV-Cre-mediated VGLUT2 deletion and from restoration of MC4Rs in Sim1 neurons demonstrate the importance of glutamate release in body weight regulation, which is in stark contrast to mild and gender-specific changes in body weight that were observed when glutamate release was disrupted early in embryonic Sim1 neurons. These results suggest that functionally redundant pathways exist to compensate for disruption of glutamate release during developmental periods. Indeed, previous studies have shown that compensatory mechanisms can effectively mask detrimental phenotypes and perform essential physiological roles when deletion is

created during developmental periods (Luquet et al., 2005). Such plasticity in gene compensation likely derives from the fact that MC4Rs are widely expressed in the brain (Kishi et al., 2003; Liu et al., 2003) and that the action of melanocortin influences multiple physiological processes, including body weight, throughout multiple brain areas. Given the lack of significant changes in body weight (females) and mild obesity (males) by disruption of glutamate release from embryonic Sim1 neurons, we speculate that lack of glutamate signaling blocks the downstream pathway mediated by MC4Rs on Sim1 neurons and invokes the pathway mediated by MC4Rs on non-Sim1 neurons for body weight maintenance (Figure 7E, left panel). Disruption of glutamate release leads to mild obesity in males, but not in females, suggesting that MC4R action on non-Sim1 neurons is sufficient for body weight homeostasis in females. In mice with targeted conditional restoration of MC4Rs in Sim1 neurons of Mc4r-null mice, lack of MC4R function on non-Sim1 neurons does not provide functional compensation by the melanocortin signaling. Thus, selective disruption of glutamate release from Sim1 neurons, which blocks MC4R action in these neurons, causes reversal of obesity to a level comparable to that of Mc4r-null mice (Figure 7E, right panel). The location of such proposed non-Sim1 neurons is unknown but could involve acetylcholinergic neurons in the spinal cord, since restoration of MC4Rs in these neurons increases energy expenditure and reduces obesity (Rossi et al., 2011; Sohn et al., 2013). Nevertheless, our results demonstrate a powerful developmental compensation within the melanocortin pathway toward regulating body weight. These results imply that care should be taken in data interpretation when the function of the melanocortin pathway is examined using animal models with early embryonic loss of gene function. In addition to glutamate, the PVH also releases diverse neuropeptides. Interestingly, genetic deletion of major PVH neuropeptides produces little or no effects on body weight (Jacobson, 1999; Vollmer et al., 2006; Yamada et al., 1997), while

866 Cell Metabolism 18, 860–870, December 3, 2013 ª2013 Elsevier Inc.

Cell Metabolism Glutamate Mediates MC4R Function

Figure 7. Body Length Control by Glutamate Release from Sim1 Neurons and Proposed Model for Glutamate in Mediating MC4R Action on Sim1 Neurons (A–D) Bar graphs showing lean mass in males (n = 10–22) (A) and females (n = 7–17) (B) and body length in males (n = 10–22) (C) and females (n = 7–17) (D). Both lean mass and body length were measured at 20 weeks of age. All data are presented as mean ± SEM; *p < 0.05, one-way ANOVA test. See also Figure S6. (E) A proposed model for the melanocortin action on body weight regulation. The melanocortin pathway regulates body weight through MC4R expression in both Sim1 neurons and non-Sim1 neurons. Disruption of glutamate release from Sim1 neurons blocks the action of MC4Rs on Sim1 neurons. Left panel shows that in Sim1-Cre: Vglut2flox/flox mice, disruption of glutamate release from Sim1 neurons invokes activation of a pathway mediated by MC4Rs on non-Sim1 neurons (larger black arrow), which is sufficient for body weight maintenance in females (no obesity), but not males (mild obesity). Right panel shows that in Sim1-Cre:Mc4rtb/tb:Vglut2flox/flox mice, only MC4Rs on Sim1 neurons are functional and that MC4Rs on non-Sim1 neurons are inactivated; thus, disruption of glutamate release is incapable of invoking activation of the pathway mediated by MC4Rs on non-Sim1 neurons, therefore leading to obesity comparable to that of Mc4r-null mice.

pharmacological application of these peptides results in altered energy balance (Parker and Bloom, 2012). Previous results suggest a role for TRH, CRH, and oxytocin in body weight regulation (Kublaoui et al., 2008; Lu et al., 2003; Vella et al., 2011; Zhang et al., 2011). Thus, it would be highly informative to examine whether, and the extent to which, PVH peptides mediate MC4R action on Sim1 neurons in body weight control. Given that our data support a role for glutamate in mediating the body weight-regulating effects of MC4Rs on Sim1 neurons, we speculate that PVH peptidergic action, if at all important for body weight regulation, might be mediated by altering glutamate release, as previously proposed for oxytocin and vasopressin action (Bailey et al., 2006; Peters et al., 2008). One striking observation in this study is that Mc4r-null mice exhibit severe hyperphagia in adults, with no obvious defects in feeding at a young age. Notably, food intake in young Mc4rnull mice was either equivalent to or slightly greater than that observed in controls (Albarado et al., 2004; Shaw et al., 2005; Vella et al., 2011). Consistent with these data, deletion of

MRAP2 (Mrap2 / mice), an MC4Rinteracting protein, leads to an obesity phenotype similar to that observed in Mc4r-null mice, with no defects in food intake prior to obesity development (Asai et al., 2013). In contrast, we observed significantly reduced energy expenditure in young Mc4r-null mice. In addition, selective restoration of MC4Rs in Sim1 neurons increased energy expenditure of Mc4r-null mice, which was reversed by disruption of glutamate release from those same neurons (Figure 4). The changes in energy expenditure were not secondary to differences in body weight, since the starting body weights of both experimental and control groups were comparable when energy expenditure was measured at young ages. Thus, glutamate mediates MC4Rs on Sim1 neurons in energy expenditure regulation at young ages and, when altered, results in obesity. This conclusion is consistent with the anatomical data that reveal prominent projections from PVH MC4R-expressing neurons to the brown adipose tissues (Song et al., 2008), the major thermogenic organ in rodents. Our current data on food intake in adult mice, as well as previous results from pair feeding Mc4r-null mice or treating with MC4R agonists, suggest that MC4Rs regulate food intake independent of body weight changes (Balthasar et al., 2005; Fan et al., 1997; Marsh et al., 1999; Ste Marie et al., 2000). In agreement with this, Mrap2 / mice show both normal feeding and normal response in feeding to MTII at a young age but ultimately

Cell Metabolism 18, 860–870, December 3, 2013 ª2013 Elsevier Inc. 867

Cell Metabolism Glutamate Mediates MC4R Function

develop hyperphagia as adults (Asai et al., 2013). This hyperphagia in adult mice might be caused by an impaired coordination of nutrient intake and substrate oxidation (Albarado et al., 2004; Butler et al., 2001; Srisai et al., 2011). Collectively, our results suggest that reduced energy expenditure in Mc4r-null mice is responsible for initial obesity, which is acerbated by subsequent hyperphagia. In summary, our study establishes a critical role for glutamate in mediating the function of MC4Rs on Sim1 neurons in the regulation of both energy expenditure and food intake. EXPERIMENTAL PROCEDURES Animal Care Mice were housed at 22 C–24 C in a 12 hr light/12 hr dark cycle, with food and water provided ad libitum. Animal care and procedures were approved by the University of Texas Health Science Center at Houston Institutional Animal Care and Use Committee. Glutamate Release Assay The assay for glutamate release from PVH tissues was derived from combined methods from previous publications (Enriori et al., 2007; Kajimoto et al., 2007). After isoflurane anesthesia, brains were taken out from Vglut2flox/flox and Sim1Cre: Vglut2flox/flox mice. Two slices (400 mm thick) containing PVH were treated for 1 hr at 37 C with artificial cerebrospinal fluid (aCSF; w/0.6 internal unit aprotinin/ml), which had been equilibrated with 95% O2 and 5% CO2, and then challenged with 56 mM KCl for 1 min. Medium (50 ml) was taken from basal and KCl-stimulated conditions for measuring glutamate concentration for an ELISA kit. AAV Vector Stereotaxic Injections Both AAV-Cre-GFP and AAV-GFP vectors were purchased from the viral core facility of the University of Pennsylvania. Either vector (20 nl) of was stereotaxically injected into bilateral PVH with the following coordinates: bregma 0.7 mm; midline ± 0.3 mm; dorsal surface 4.5 mm and 4.3 mm using a 0.5 ml Hamilton syringe controlled by a nanoinjector. The injection speed was controlled at 0.5 nl/min, and the syringe was withdrawn 15 min after the final injection. In Situ Hybridization RNAscope Multiplex Fluorescent Assay, one in situ hybridization (ISH) technique designed to visualize multiple cellular RNA targets in fresh frozen tissues (Wang et al., 2012), was used for detecting the colocalization of Vgat, Vglut2, and Mc4r in the brain (Advanced Cell Diagnostics). The digoxigenin-labeled cRNA probes were generated against mouse Vglut2 or Mc4r mRNA, and ISH was also performed as we previously described for validating the deletion of Vglut2 or restoration of Mc4r mRNA in Sim1 neurons (Xu et al., 2012, 2013; Kishi et al., 2003). See Supplemental Information for details. Body Weight, Food Intake, and Energy Expenditure Weekly body weight was monitored on all genotypes fed standard mouse chow (8664Teklad F6 Rodent Diet: 4.05 kcal/g, 3.3 kcal/g metabolizable energy, 12.5% kcal from fat; Harlan Teklad) from 4- to 20-week-old mice. Body composition (fat mass and lean mass) was measured at the indicated times by using the EchoMRI system (EchoMRI). In young mice, all study subjects were individually housed after weaning, and daily food intake was monitored for 1 week after 3 days of acclimation. Body weights of these mice were also recorded at the beginning and end of the measurement period. In adult mice (19–20 weeks old), all study subjects were preacclimated for at least 1 week by single housing, and then daily food intake was monitored for 1 week. Energy expenditure was measured by oxygen consumption by indirect calorimetry. Individually housed mice maintained on chow diet at 4–5 weeks old were placed at room temperature (22 C–24 C) in chambers of a Comprehensive Lab Animal Monitoring System (CLAMS; Columbus Instruments). Food and water were provided ad libitum. Mice were acclimatized in the chambers for at least 48 hr prior to data collection.

MTII Action on Energy Expenditure Mice, at 4–5 weeks old, were acclimated in CLAMS chambers for at least 2 days. As previously described (Balthasar et al., 2005), all mice were intraperitoneally (i.p.) injected with normal saline at 12:00 pm on day 1; their food was removed and the O2 consumption was measured for the ensuing 2 hr. On day 2, all mice were i.p. injected with 200 mg MTII (H3902; Bachem) at 12:00 pm, their food was removed, and O2 consumption was monitored for the ensuing 2 hr. Statistical Analysis Data represent means ± SEM. Statistical analysis was performed using GraphPad Prism. Statistical significance among the groups was tested using one-way ANOVA followed by a post hoc least significant difference test or an unpaired Student’s t test, when appropriate. Statistical significance was defined as p < 0.05. SUPPLEMENTAL INFORMATION Supplemental Information includes Supplemental Experimental Procedures and six figures and can be found with this article online at http://dx.doi.org/ 10.1016/j.cmet.2013.11.003. AUTHOR CONTRIBUTIONS Q.T. conceived and designed the study. Yuanzhong Xu, Z.W., H.S., Y.Z., and E.R.K. performed the experiments. B.B.L., B.R.A., and Yong Xu provided essential reagents for the study. Yuanzhong Xu and Q.T. analyzed the data and wrote the paper. All authors agreed on the final version of the manuscript. ACKNOWLEDGMENTS We thank V. Narkar and N. Justice for critically reading the manuscript and provide helpful input. Q.T. is the holder of Becker Family Foundation Professor in Diabetes Research. This work was supported by NIH R01DK092605 and a Scientist Development Award from American Heart Association (to Q.T.); NIH R01DK093587, R00DK085330, P30 DK079638-03, and grants from the American Diabetes Association, the Klarman Family Foundation, Naman Family Fund for Basic Research, and Curtis Hankamer Basic Research Fund (all to Yong Xu); NIH R01 NS078294 (to B.R.A.); and NIH R01 DK075632, R01 DK07105, R01 DK089044, P30 DK046200, and P30 DK057521 (to B.B.L.). The authors would like to thank Dr. Zhengmei Mao for the assistance in imaging. The authors declare no competing interests. Received: April 22, 2013 Revised: August 26, 2013 Accepted: October 25, 2013 Published: December 3, 2013 REFERENCES Albarado, D.C., McClaine, J., Stephens, J.M., Mynatt, R.L., Ye, J., Bannon, A.W., Richards, W.G., and Butler, A.A. (2004). Impaired coordination of nutrient intake and substrate oxidation in melanocortin-4 receptor knockout mice. Endocrinology 145, 243–252. Asai, M., Ramachandrappa, S., Joachim, M., Shen, Y., Zhang, R., Nuthalapati, N., Ramanathan, V., Strochlic, D.E., Ferket, P., Linhart, K., et al. (2013). Loss of function of the melanocortin 2 receptor accessory protein 2 is associated with mammalian obesity. Science 341, 275–278. Atasoy, D., Betley, J.N., Su, H.H., and Sternson, S.M. (2012). Deconstruction of a neural circuit for hunger. Nature 488, 172–177. Bailey, T.W., Jin, Y.H., Doyle, M.W., Smith, S.M., and Andresen, M.C. (2006). Vasopressin inhibits glutamate release via two distinct modes in the brainstem. J. Neurosci. 26, 6131–6142. Balthasar, N., Dalgaard, L.T., Lee, C.E., Yu, J., Funahashi, H., Williams, T., Ferreira, M., Tang, V., McGovern, R.A., Kenny, C.D., et al. (2005). Divergence of melanocortin pathways in the control of food intake and energy expenditure. Cell 123, 493–505.

868 Cell Metabolism 18, 860–870, December 3, 2013 ª2013 Elsevier Inc.

Cell Metabolism Glutamate Mediates MC4R Function

Butler, A.A., and Kozak, L.P. (2010). A recurring problem with the analysis of energy expenditure in genetic models expressing lean and obese phenotypes. Diabetes 59, 323–329. Butler, A.A., Marks, D.L., Fan, W., Kuhn, C.M., Bartolome, M., and Cone, R.D. (2001). Melanocortin-4 receptor is required for acute homeostatic responses to increased dietary fat. Nat. Neurosci. 4, 605–611. Cone, R.D. (2005). Anatomy and regulation of the central melanocortin system. Nat. Neurosci. 8, 571–578. Cowley, M.A., Pronchuk, N., Fan, W., Dinulescu, D.M., Colmers, W.F., and Cone, R.D. (1999). Integration of NPY, AGRP, and melanocortin signals in the hypothalamic paraventricular nucleus: evidence of a cellular basis for the adipostat. Neuron 24, 155–163. Elmquist, J.K., Coppari, R., Balthasar, N., Ichinose, M., and Lowell, B.B. (2005). Identifying hypothalamic pathways controlling food intake, body weight, and glucose homeostasis. J. Comp. Neurol. 493, 63–71. Enriori, P.J., Evans, A.E., Sinnayah, P., Jobst, E.E., Tonelli-Lemos, L., Billes, S.K., Glavas, M.M., Grayson, B.E., Perello, M., Nillni, E.A., et al. (2007). Dietinduced obesity causes severe but reversible leptin resistance in arcuate melanocortin neurons. Cell Metab. 5, 181–194. Fan, W., Boston, B.A., Kesterson, R.A., Hruby, V.J., and Cone, R.D. (1997). Role of melanocortinergic neurons in feeding and the agouti obesity syndrome. Nature 385, 165–168.

Morton, G.J., Thatcher, B.S., Reidelberger, R.D., Ogimoto, K., WoldenHanson, T., Baskin, D.G., Schwartz, M.W., and Blevins, J.E. (2012). Peripheral oxytocin suppresses food intake and causes weight loss in dietinduced obese rats. Am. J. Physiol. Endocrinol. Metab. 302, E134–E144. Parker, J.A., and Bloom, S.R. (2012). Hypothalamic neuropeptides and the regulation of appetite. Neuropharmacology 63, 18–30. Peters, J.H., McDougall, S.J., Kellett, D.O., Jordan, D., Llewellyn-Smith, I.J., and Andresen, M.C. (2008). Oxytocin enhances cranial visceral afferent synaptic transmission to the solitary tract nucleus. J. Neurosci. 28, 11731– 11740. Rossi, J., Balthasar, N., Olson, D., Scott, M., Berglund, E., Lee, C.E., Choi, M.J., Lauzon, D., Lowell, B.B., and Elmquist, J.K. (2011). Melanocortin-4 receptors expressed by cholinergic neurons regulate energy balance and glucose homeostasis. Cell Metab. 13, 195–204. Schneider, J.E. (2009). If I only had a whole brain: the importance of extrahypothalamic areas in the energy balance equation. Endocrinology 150, 5195– 5198. Shaw, A.M., Irani, B.G., Moore, M.C., Haskell-Luevano, C., and Millard, W.J. (2005). Ghrelin-induced food intake and growth hormone secretion are altered in melanocortin 3 and 4 receptor knockout mice. Peptides 26, 1720–1727.

Farooqi, I.S., and O’Rahilly, S. (2005). Monogenic obesity in humans. Annu. Rev. Med. 56, 443–458.

Shi, Y.C., Lau, J., Lin, Z., Zhang, H., Zhai, L., Sperk, G., Heilbronn, R., Mietzsch, M., Weger, S., Huang, X.F., et al. (2013). Arcuate NPY controls sympathetic output and BAT function via a relay of tyrosine hydroxylase neurons in the PVN. Cell Metab. 17, 236–248.

Fremeau, R.T., Jr., Voglmaier, S., Seal, R.P., and Edwards, R.H. (2004). VGLUTs define subsets of excitatory neurons and suggest novel roles for glutamate. Trends Neurosci. 27, 98–103.

Skibicka, K.P., and Grill, H.J. (2009). Hypothalamic and hindbrain melanocortin receptors contribute to the feeding, thermogenic, and cardiovascular action of melanocortins. Endocrinology 150, 5351–5361.

Huszar, D., Lynch, C.A., Fairchild-Huntress, V., Dunmore, J.H., Fang, Q., Berkemeier, L.R., Gu, W., Kesterson, R.A., Boston, B.A., Cone, R.D., et al. (1997). Targeted disruption of the melanocortin-4 receptor results in obesity in mice. Cell 88, 131–141.

Sohn, J.W., Harris, L.E., Berglund, E.D., Liu, T., Vong, L., Lowell, B.B., Balthasar, N., Williams, K.W., and Elmquist, J.K. (2013). Melanocortin 4 receptors reciprocally regulate sympathetic and parasympathetic preganglionic neurons. Cell 152, 612–619.

Jacobson, L. (1999). Glucocorticoid replacement, but not corticotropinreleasing hormone deficiency, prevents adrenalectomy-induced anorexia in mice. Endocrinology 140, 310–317.

Song, C.K., Vaughan, C.H., Keen-Rhinehart, E., Harris, R.B., Richard, D., and Bartness, T.J. (2008). Melanocortin-4 receptor mRNA expressed in sympathetic outflow neurons to brown adipose tissue: neuroanatomical and functional evidence. Am. J. Physiol. Regul. Integr. Comp. Physiol. 295, R417–R428.

Kajimoto, T., Okada, T., Yu, H., Goparaju, S.K., Jahangeer, S., and Nakamura, S. (2007). Involvement of sphingosine-1-phosphate in glutamate secretion in hippocampal neurons. Mol. Cell. Biol. 27, 3429–3440. Kishi, T., Aschkenasi, C.J., Lee, C.E., Mountjoy, K.G., Saper, C.B., and Elmquist, J.K. (2003). Expression of melanocortin 4 receptor mRNA in the central nervous system of the rat. J. Comp. Neurol. 457, 213–235. Krashes, M.J., Shah, B.P., Koda, S., and Lowell, B.B. (2013). Rapid versus Delayed Stimulation of Feeding by the Endogenously Released AgRP Neuron Mediators GABA, NPY, and AgRP. Cell Metab. 18, 588–595. Kublaoui, B.M., Gemelli, T., Tolson, K.P., Wang, Y., and Zinn, A.R. (2008). Oxytocin deficiency mediates hyperphagic obesity of Sim1 haploinsufficient mice. Mol. Endocrinol. 22, 1723–1734. Liu, H., Kishi, T., Roseberry, A.G., Cai, X., Lee, C.E., Montez, J.M., Friedman, J.M., and Elmquist, J.K. (2003). Transgenic mice expressing green fluorescent protein under the control of the melanocortin-4 receptor promoter. J. Neurosci. 23, 7143–7154. Lu, X.Y., Barsh, G.S., Akil, H., and Watson, S.J. (2003). Interaction between alpha-melanocyte-stimulating hormone and corticotropin-releasing hormone in the regulation of feeding and hypothalamo-pituitary-adrenal responses. J. Neurosci. 23, 7863–7872. Luquet, S., Perez, F.A., Hnasko, T.S., and Palmiter, R.D. (2005). NPY/AgRP neurons are essential for feeding in adult mice but can be ablated in neonates. Science 310, 683–685. Marsh, D.J., Hollopeter, G., Huszar, D., Laufer, R., Yagaloff, K.A., Fisher, S.L., Burn, P., and Palmiter, R.D. (1999). Response of melanocortin-4 receptordeficient mice to anorectic and orexigenic peptides. Nat. Genet. 21, 119–122. Michaud, J.L., Boucher, F., Melnyk, A., Gauthier, F., Goshu, E., Le´vy, E., Mitchell, G.A., Himms-Hagen, J., and Fan, C.M. (2001). Sim1 haploinsufficiency causes hyperphagia, obesity and reduction of the paraventricular nucleus of the hypothalamus. Hum. Mol. Genet. 10, 1465–1473.

Srisai, D., Gillum, M.P., Panaro, B.L., Zhang, X.M., Kotchabhakdi, N., Shulman, G.I., Ellacott, K.L., and Cone, R.D. (2011). Characterization of the hyperphagic response to dietary fat in the MC4R knockout mouse. Endocrinology 152, 890–902. Ste Marie, L., Miura, G.I., Marsh, D.J., Yagaloff, K., and Palmiter, R.D. (2000). A metabolic defect promotes obesity in mice lacking melanocortin-4 receptors. Proc. Natl. Acad. Sci. USA 97, 12339–12344. Swanson, L.W., and Sawchenko, P.E. (1980). Paraventricular nucleus: a site for the integration of neuroendocrine and autonomic mechanisms. Neuroendocrinology 31, 410–417. Tolson, K.P., Gemelli, T., Gautron, L., Elmquist, J.K., Zinn, A.R., and Kublaoui, B.M. (2010). Postnatal Sim1 deficiency causes hyperphagic obesity and reduced Mc4r and oxytocin expression. J. Neurosci. 30, 3803–3812. Tong, Q., Ye, C., McCrimmon, R.J., Dhillon, H., Choi, B., Kramer, M.D., Yu, J., Yang, Z., Christiansen, L.M., Lee, C.E., et al. (2007). Synaptic glutamate release by ventromedial hypothalamic neurons is part of the neurocircuitry that prevents hypoglycemia. Cell Metab. 5, 383–393. Tscho¨p, M.H., Speakman, J.R., Arch, J.R., Auwerx, J., Bru¨ning, J.C., Chan, L., Eckel, R.H., Farese, R.V., Jr., Galgani, J.E., Hambly, C., et al. (2012). A guide to analysis of mouse energy metabolism. Nat. Methods 9, 57–63. Vella, K.R., Ramadoss, P., Lam, F.S., Harris, J.C., Ye, F.D., Same, P.D., O’Neill, N.F., Maratos-Flier, E., and Hollenberg, A.N. (2011). NPY and MC4R signaling regulate thyroid hormone levels during fasting through both central and peripheral pathways. Cell Metab. 14, 780–790. Vollmer, R.R., Li, X., Karam, J.R., and Amico, J.A. (2006). Sodium ingestion in oxytocin knockout mice. Exp. Neurol. 202, 441–448. Wang, F., Flanagan, J., Su, N., Wang, L.C., Bui, S., Nielson, A., Wu, X., Vo, H.T., Ma, X.J., and Luo, Y. (2012). RNAscope: a novel in situ RNA analysis platform for formalin-fixed, paraffin-embedded tissues. J. Mol. Diagn. 14, 22–29.

Cell Metabolism 18, 860–870, December 3, 2013 ª2013 Elsevier Inc. 869

Cell Metabolism Glutamate Mediates MC4R Function

Wu, Q., Boyle, M.P., and Palmiter, R.D. (2009). Loss of GABAergic signaling by AgRP neurons to the parabrachial nucleus leads to starvation. Cell 137, 1225– 1234.

Xu, Y., Kim, E.R., Zhao, R., Myers, M.G., Jr., Munzberg, H., and Tong, Q. (2013). Glutamate release mediates leptin action on energy expenditure. Mol Metab 2, 109–115.

Wu, Z., Xu, Y., Zhu, Y., Sutton, A.K., Zhao, R., Lowell, B.B., Olson, D.P., and Tong, Q. (2012). An obligate role of oxytocin neurons in diet induced energy expenditure. PLoS ONE 7, e45167.

Yamada, M., Saga, Y., Shibusawa, N., Hirato, J., Murakami, M., Iwasaki, T., Hashimoto, K., Satoh, T., Wakabayashi, K., Taketo, M.M., and Mori, M. (1997). Tertiary hypothyroidism and hyperglycemia in mice with targeted disruption of the thyrotropin-releasing hormone gene. Proc. Natl. Acad. Sci. USA 94, 10862–10867.

Xu, Y., and Tong, Q. (2011). Expanding neurotransmitters in the hypothalamic neurocircuitry for energy balance regulation. Protein Cell 2, 800–813. Xu, Y., O’Brien, W.G., 3rd, Lee, C.C., Myers, M.G., Jr., and Tong, Q. (2012). Role of GABA release from leptin receptor-expressing neurons in body weight regulation. Endocrinology 153, 2223–2233.

Zhang, G., Bai, H., Zhang, H., Dean, C., Wu, Q., Li, J., Guariglia, S., Meng, Q., and Cai, D. (2011). Neuropeptide exocytosis involving synaptotagmin-4 and oxytocin in hypothalamic programming of body weight and energy balance. Neuron 69, 523–535.

870 Cell Metabolism 18, 860–870, December 3, 2013 ª2013 Elsevier Inc.